Home

parametrics editors for structured documents

1. lt meta id metalex AADCFAA name pubblicazione norm gt lt meta gt lt mcontainer gt lt mcontainer gt lt container gt basic name intestazione id metalex EEAA gt lt htitle name tipoDoc id metalex EFEAA gt DECRETO LEGISLATIVO lt htitle gt 25 gennaio 1992 n inline name numDoc id metalex _ECEAA gt 84 lt inline gt milestone name h br id metalex_EBEAA xml lang it gt lt milestone gt lt htitle name titoloDoc id titolo xml lang it gt iE e n 88 220 CEE relative agli organismi di investimento collettivo in Attuazione delle direttive n 85 611 C I I I valori mobiliari operanti nella forma di societa di investimento a capitale variabile SICAV 40 3 Structural differences in XML dialects lt htitle gt lt basic gt omissis basic name articolato id metalex_ECAA gt lt hcontainer name articolo gt lt htitle name num id metalex_EEPCAA gt Art 1 lt htitle gt lt htitle name rubrica gt Societa di investimento a capitale variabile e autorizzazione alla costituzione lt htitle gt lt hcontainer name comma gt lt htitle name num id metalex_EBCPCAA gt 1 lt htitle gt lt basic name corpo id metalex EACPCAA gt La societa per azioni cost
2. VN AN q6I se VN AN q8T S VN AN ALI se VN AN qII S VN AN qcT 4se VN AN qvT se VN AN q T se VN AN qcI sey VN AN qII se VN AN qor S 0 SN q6 se 09 SN q8 4se4 09 SN q4 se 09 SN q9 HSE 09 SN qs XS8 OcI SN qf se3 OcI SN qE 3 e3 0 SN qg 3 5 e3 08I SN qT 5e3 ou 10 so4 peusip duroooe seg 99s eurrj posdeyy 28s qu eurrg addy seg Tequinu ysey 0UIl pug OUIT 31e3g I 2SN 6 4 103 6 4 Analysis of the results The usability test was successfully completed in six hours and all ques tionnaires except one were filled in properly Indeed an user completely forgot to fill in the questionnaire related to the evaluation of satisfaction and I excluded it in the analysis of the results So for the analysis of the results I used the remaining nine tests 6 4 1 Summary of the testers The information about users show that my sample of users was quite representative of the scenario in which a legal markup editor should be used I expected the users of a new legal markup editor to be professionals and to have a good knowledge of computer science or laws and jurisprudence I also expected them to be skilled in computers online and desktop applications and words processors and that some of them also had some knowledge of legal informatics and XML editors Indeed we can see in table that the average age of the users was thirty three years old Almost all of them use often computer
3. Times New Roman 10 G C a E 4 oam P Commenti finali gt md uotations Body of provision Conc Dea Sw BSa 2 ATs 3 E EE EEEE EET SEET TERE EE Standard v Insertion of new section 45C in the principal enactment ak Part 3 The following new section is hereby inserted immediately after section 45B of the p Pise have effect as section 45C of that enactment f Section Article Facilitating persons to leave Sri Lanka lt m Number H5C 1 Any person who 4 Multiple partions Clauses a organizes one or more personsto leave Sri Lanka in contravention of any of the provisions Latere Numbers Title Sidenote b attempts or does any act preparatory to or aids and abets any other person to so organ Insert object be guilty of an offence 2 8 z 2 Any person guilty of am offence under subsection 1 shall upon conviction by zl imprisonment of either description for aterm not less than one year and not more than tive yed Article 1 Clause 1 Article 2 Clause 1 Clause 2 Article 3 Clause 1 3 Inthis section the expression organize with its grammatical variations includes a the recruitment of a person on a promise of securing employment outside Sri Lanka 4 c c c b knowingly making false promises of employment im a foreign country or making disseminating misleading information with the intention of inducing
4. 24 2 The markup of structured documents and attributes The differences between the two standards regard the SGML declaration that in XML is substituted by the XML own declaration and the character set because XML adopts unicode US91 Like SGML valid XML documents must pass two kinds of validation the well formed validation and the schema validation The well formed valida tion is a kind of syntax validation For example it ensures that the document has a unique root container and that all elements are formed by an opening tag and a closing tag The schema validation checks if the document fol lows the rules that are defined in the schema referred by its document type declaration The schema of XML documents can be created using different technologies such as DTD XML Schema BMC 04 RELAX NG CMO1 Schematron and others XML is nowadays the most used descriptive markup language and it is used to markup a large variety of documents ranging from the documents that describe user interfaces to the documents that express actions within business processes 7 but it is especially suitable for the markup of text based resources 2 4 XML for the markup of text based re sources In this chapter we have seen that a structured document both a proce dural one and a text based one is a resource formed at least by a logical structure a human readable content and some other background informa tion usually referred as metadat
5. legal RuleML Even if this demonstrates that the requirement of indepen dence was met LIME could be improved in many ways For example two of the most requested features are the possibility not to use the FRBR storage system and the independence of the LIME core from its interface The FRBR storage is one of the mandatory LIME requirements because for example parliaments legal drafters want both to markup documents and to easily catalog resources This is true but it is also true that creating a LIME API would enable the creation of plugins In this way LIME should continue to use the FRBR storage system but plugins can be created with the purpose to override the LIME default storage system and supply storage system more suitable for other specific contexts The independence of the LIME core from its interface is not a trivial task but it can be achieved because LIME relies on the Model View Controller architectural style For this reason it is possible to detach LIME from its interface and to supply a LIME core framework that others can use with their own interface However the current LIME interface is specifically designed to meet the second requirement of a markup editor A markup editor must drive users to correctly markup XML documents To evaluate the LIME interface I designed an usability test for markup editors and applied it to LIME The test inspected the efficacy and the effi ciency of LIME and the satisfaction of users
6. 22 oko ks 91 FEE eS Gaon obs 93 6 3 The LIME s user experience test less 94 6 3 1 OMeshodolEi 3 2 fa see sa sed ERI E 95 Du uu SON EU EGRE ME A 96 6 3 3 The complete test 9 4 mace ooo x Fey 96 6 4 Analysis or the results uus Yo a 2 eae RACES ORE Y 103 T 103 6 4 2 Analysisoftheefficacy a ln 105 6 4 3 Analysis of the efficiency 107 T 113 6 5 LIME s strengths and weaknesses LL 116 iv CONTENTS Conclusions 119 y 123 ist of Figures 133 ist of Tables 135 E E E os 2 Ce H 9 ke a istings 138 Chapter 0 Introduzione Lo scopo di questa dissertazione di identificare le tecnologie pi appro priate per la creazione di editor parametrici per documenti strutturati e di descrivere LIME un editor di markup parametrico e indipendente dal lin guaggio La recente evoluzione delle tecnologie XML ha portato ad un utilizzo sempre pi consistente di documenti strutturati Oggigiorno questi vengono utilizzati sia per scopi tipografici sia per l interscambio di dati nella rete internet Per questa ragione sempre pi persone hanno a che fare con docu menti XML nel lavoro quotidiano Alcuni dialetti XML tuttavia non sono semplici da comprendere e da utilizzare e per questo motivo si rendono nec essari editor XML che possano guidare gli autori di documenti XML durante tutto il processo di markup In alcuni contesti specialmente in quello dell info
7. Never ug Oe a ae Vey oten 7 How often do you navigate the web Never O O O O O O O Very often 8 How often do you use desktop applications Never qq qp 0O Very otten 9 How often do you use online applications Never OL c Very aite 10 How often do you use desktop word processors i e word open office Never O O OO 2 Very Olen 11 How often do you use in browser word processors i e google docs Never DOE Very otten 12 How often do you use desktop XML editors i e oxygen XML altova XML spy Never O O Very often 13 How often do you use legislative markup editors i e NIR editor bungeni editor Never p S ED Very iis Foresee the effort you will make to complete the following tasks la How difficult do you expect it will be to sign up to the system Very diicul _ _ Very easy 2a How difficult do you expect it will be to log in to the system Very difficult O O O O O O O Very easy 98 6 Evaluation of LIME s user experience 3a 4a 5a 6a Ta 8a 9a 10a lla 12a How difficult do you expect it will be to change the editor s language to your preferred one Very difficult O O O O O O O Very easy How difficult do you expect it will be to import a Microsoft word s file from your desktop Very difficult O 0O O O
8. lt section id sct1 gt num id sct1 nm1 gt 1 lt num gt heading id sct1 hdngl gt Definitions and application lt heading gt clause id sctl clal gt num id sctl clal nml gt 1 lt num gt lt content gt lt blockList id sctl clal ull gt lt listIntroduction id sctl clal ull 1lstntrdctnl gt In this Act unless the context indicates otherwise lt listIntroduction gt lt item id sctl clal itml gt def id sct1 clal itml df1 gt area of jurisdiction def gt means the area of jurisdiction designated for a traditional community and traditional council that have been recognised as provided for in omissis lt item gt lt blockList gt lt content gt lt clause gt clause id sctl cla2 gt num id sctl cla2 nm1 gt 2 lt num gt lt content gt lt p gt Nothing contained in this Act may be construed as precluding members of a traditional community from addressing a traditional leader by the traditional title accorded to him or her by custom but such traditional title does not derogate from or add anything to the status role and functions of a traditional leader as provided for in this Act lt p gt lt content gt lt clause gt lt section gt lt chapter gt omissis lt body gt lt bill gt 3 4 45 lt akomaNtoso gt 3 4 Abstraction of the XML dialects The XML patterns are the key element for the abstraction of
9. section id sec2 l1st1l itmb modl qtdS1 sec47A sub3 gt lt num gt 3A lt num gt lt content gt lt p gt Notwithstanding the provisions of this section but without prejudice to lt ref id ref6 href ak act 2010 08 27 1 main sec47A sub2 itmb gt 2 b lt ref gt an Act of Parliament may provide for re publication of the draft Constitution and its re introduction into the National Assembly for re consideration lt p gt lt content gt lt section gt quotedStructure mod lt p gt lt content gt omissis lt article gt omissis lt body gt omissis lt akomaNtoso gt 3 5 A generic markup process for distinct di alects In this chapter I pointed out differences and similarities among different XML languages I described examples of dialects used to markup textual resources and examples of standards used for the markup of legal and leg islative documents I ended my XML languages review with Akoma Ntoso 3 5 51 that supplies a well pattern based architecture Patterns are really important concepts If used correctly they enable a generic markup process for distinct documents marked up with different di alects The Akoma Ntoso developers are strongly convinced that the logical structure of every kind of document can be described using only six pat terns Nothing more true than this Even if a schema is not designed to be pattern based we can abstract its
10. O O O Very easy How difficult do you expect it will be to open one of the examples supplied by the editor Very difficult O O O O O O O Very easy How difficult do you expect it will be to save a file under a different name Very difficult O O O O O O O Very easy How difficult do you expect it will be to open the XML preview of a document Very difficult O O O O O O O Very easy How difficult do you expect it will be to save the XML ver sion of a document on your desktop Very difficult O O O O O O O Very easy How difficult do you expect it will be to log out from the system Very difficult O O O O O O O Very easy Having a partially marked up document how difficult do you expect it will be to markup its preface Very difficult O O O O O O O Very easy Having a partially marked up document how difficult do you expect it will be to markup its document date Very difficult O O O O O O O Very easy Having a partially marked up document how difficult do you expect it will be to markup its main body Very ditheu P m ppp Very cacy 6 3 99 13a Having a partially marked up document how difficult do you expect it will be to markup one of its articles with its number and its heading Very difficult O O O O O O O Very easy 14a Having a marked up article how difficult do
11. based content models BPVC1 The pattern based design has two purposes On the one hand it allows future extensions of the schema to be carried out in a simple and backward compatible way On the other hand it permits to specify a small set of patterns to which all the elements of the schema must belong to in order to define in a way as simple as possible how every element interacts with the other For example elements that belong to the block pattern can contain elements that belong to the inline pattern but can not contain other block elements I will describe deeply the Akoma Ntoso patterns in the next sec tion The Akoma Ntoso standard differs from the CEN Metalex standard be cause it does not supply only very generic elements that reflect the pattern based content models It also supplies a full legal vocabulary that can be l Akoma Ntoso means united hearts in Akan language of West Africa and it is an acronym for Architecture for Knowledge Oriented Management of African Normative Texts using Open Standards and Ontologies fUNDESA a department of the United Nations that helps countries around the world to meet their economic social and environmental goals 42 3 Structural differences in XML dialects used by the majority of legal drafters from all over the world For instance Akoma Ntoso supplies elements for the creation of sections and articles and it provides elements for the markup of the preamble and
12. 1 1 1 1 0 1 1 0 1 78 9 1 1 1 1 0 0 1 1 1 78 avg 78 89 100 89 33 78 100 89 100 84 The eighty four percent of the tasks was successfully completed More than fifty percent of the task was successfully completed by all users The only exception was found when the users were asked to open one of the marked up examples supplied by the editor The majority of the users were not able to figure out how to do it In order to open an example in LIME users have to click on the file menu and then on the open menu Examples are stored in a folder called example and can be opened like any other file When I asked the users to open a marked up example they seemed puzzled and started to look for the example allover the application except in the open menu Other two tasks were completed only by seventy eight percent of the users Some users got confused when they had to sign up to the system and when they had to save the file under a different name I expected this difficulty because the LIME registration mask is still a prototype and the save functionality needs a certain expertise in FRBR Overall the efficacy evaluation of LIME gave good results Except for 6 4 107 one critical issue and two improvable functionality I can consider LIME an effective markup editor d 100 2 co a S 75 o v 3g 50 3 5 S 25 E v e d 0 Tib T3b T5b T7b T9b T2b T4b T6b T8b Tasks Figure 6 1 Th
13. Table 6 8 The evaluation of LIME efficiency Task completion rate Task time cs Efficiency T1b 78 1100 7 T2b 89 120 74 T3b 100 100 100 T4b 89 150 59 T5b 33 280 11 T6b 78 140 55 T7b 100 130 76 T8b 89 130 68 T9b 100 150 66 T10b 89 150 59 T11b 89 140 63 T12b 89 140 63 T13b 78 220 35 T14b 78 140 55 T15b 89 130 68 T16b 78 320 24 T17b 89 140 63 T18b 89 140 63 T19b 78 220 35 112 6 Evaluation of LIME s user experience 100 S F E 5 10 u Lu 25 0 lt 39 40 19 9 MP NS NS Xi NP Task Figure 6 2 The efficiency evaluation by tasks The evaluation of the LIME efficiency proves that for the majority of the tasks LIME is efficient Only five of the nineteen tasks had an evaluation below fifty percent The inefficiency and inefficacy of two of these tasks T1b and T5b was already proved by the LIME efficacy evaluation This means that the parts of the LIME user interface involved in these tasks must severely be redesigned Another task that revealed big issues in usability was the one that in volved the user in the creation of a new version expression of a document Like in the efficacy evaluation users failed again in using the save system of the LIME editor and spent a lot of time to completely save new versions of the document This is surely related to the complexity of the FRBR notation but the test proved that this part of the
14. WebHelp old ee Section 2 2 External SVG Images DocBook EPUB 4 i imogedita These images are referred using the bfileref4 attribute 9 svg svg 5cm af the bimaaenhienct4 element x ST para Thisis an inline image _ lt m SK m oss Text Grid 5 Transf Entities x Elements x D samples docbook v5 sampleMathMLandSVG xml U 0000 Figure 4 2 The author view in Oxygen editor 4 3 Markup editors Markup editors are WYSIWYG editors that do not allow users to di rectly modify the XML code On the one hand markup editors are identical to classical WYSIWYG words processors because during the edit mode documents are displayed in a view that is very similar to the final result On the other hand they differ from other WYSIWYG editors because they force users to create documents that are compliant to the XML schema but they assure at the same time a certain freedom of editing The difference between markup editors and the Oxygen s author view is that users are not required to have any knowledge of the rules specified by the document type declaration 58 4 Software for XML markup For example if a user is editing a TEI document the editor does not al low him to insert a physDesc element inside a msContents element because the TEI schema does not allow this but it is the editor s interface itself that guides the user not to do it In the same way markup editors
15. a lot of legal tradition simply modifying some of its interface s components and the XSLT that transforms the final document into an Akoma Ntoso document The other side of the coin is that the Bungeni editor still carries some problems because it is created as an extension of an existing software like Norma editor Its interface is strictly related to the OpenOffice s one For this reason users can get confused because they can find a lot of tools that are not useful in the legal context This also forces designers and developers to use the technologies that OpenOffice supports Last but not least as a desktop software it must be installed on the computer it will be used on In some complex situation like a parliament this can be a big problem because it must be installed on hundreds of machines The Bungeni editor is used in a lot of African parliaments and its open source ness makes it the most reliable solution in case a desktop application is strictly needed 1CXSLT is a language used to transform document from many formats XML plain text HTML and so on to any XML dialects 62 4 Software for XML markup erd_2011 7 13_en ke debaterecord 2011 10 31 en currently open documents iaterecord 2011 713 enodt v a Baal aa Vis KS g eS y AJAAWA Q ifu v4 2 4 i i 2 A a Pai LEN EIN ITER FINITO LEE 000 N asa Structure NATIONAL ASSEMBLY genera aetiore event action
16. and the at tributes of the element selected in the left column The author view is an intermediate view between the regular text view and a WYSIWYG markup Eclipse is a free and open source Integrated Development Environment IDE It sup plies a base workspace and can be extended or customized using its plug in system Syncro Soft is a software development company located in Craiova SW of Romania It is specialized in developing Java based XML solutions 56 4 Software for XML markup editor Indeed in author view XML elements are presented in a more human readable way but the semantic and the nesting of XML documents remain clear and accessible In figure and figure 4 2 it is possible to see the oxygen grid and author view respectively Oxygen natively supports the most popular XML dialects including Doc Book TEI and XHTML and currently it is widely used both by XML experts and by new XML users 6 carspec xml x lt xml version 1 0 encoding UTF 8 v stock id 143334 date 10 10 2006 available gt entry ype quantity spec 4 rows 1BMw 2 luxury 2 Renault 1 luxury 3 Renault 2 budget 4 Mini 17 tiny v spec type wheelbase front track 3 rows 1 luxury 3008mm 1602mm 2 budget 2800mm 1513mm N 3 tiny 2601mm 1413mm Text Grid Figure 4 1 The grid view in Oxygen editor 5Courtesy of Courtesy of 4 3 57 E sampleMathM
17. de multis lt quote gt In Latin p lt msContents gt lt physDesc gt lt p gt lt material gt Parchment lt material gt written in more than one hand 7 1 4 x 5 3 8 in i 55 leaves in double columns with a few coloured capitals lt p gt lt physDesc gt lt history gt lt p gt Written in lt origPlace gt England lt origPlace gt in the lt origDate gt 13th cent lt origDate gt On fol 54v very faint is lt quote gt Iste liber est fratris guillelmi de buria de Roberti ordinis fratrum Pred icatorum lt quote gt 14th cent lt quote gt hanauilla lt quote gt is written at the foot of the page 15th cent Bought from the rev W D Macray on March 17 1863 for L 1 10s lt p gt lt history gt lt msDesc gt 32 3 Structural differences in XML dialects 3 2 2 DocBook DocBook is a descriptive and semantic markup language that aims to allow the creation of technical documentations Originally it was in tended for the markup of programming languages documentation but cur rently it can be used for any kind of technical documentation or also for other purposes such as the markup of e learning materials MOMGSFM06 O Reilly P and HaL Computer Systems O started the developing of Doc Book in 1991 in a discussion group on Usenet F and moved in 1998 to the SGML Open Consortium which is now known as OASIS DocBook started as a SGML application but later an
18. drive users to create documents that are also semantically correct For instance the Akoma Ntoso schema allows to insert articles inside paragraphs This is not semantically correct but Akoma Ntoso creators allowed it for the sake of modularity extensibility and customizability In these situations inexperienced users can make semantic mistakes and moreover if the editor is limited to infer the rules from the schema it would not be able to report these kinds of errors For these reasons it is very difficult to design markup editors that can be used with more than one XML schema or that can be used in different contexts In the next section I will describe the most used markup editors in the legal context 4 4 Markup editors for the legal context In the last two decades a lot of markup editors for legislative and legal documents were produced The legal context is one of the most difficult to handle because there are many constraints originated by parliaments and public offices Moreover users of legal markup editors have their own tra ditions and have to follow either stricter or looser workflows This means for example that editors of the Italian Camera dei Deputati should use a markup editor in a completely different way from the one of the drafters belonging to the Italian Senato For these reasons legal markup editors are usually created to permit the markup of the legislative and legal documents of a specific tradition In the
19. equivalent version in XML that now has replaced the SGML one for the majority of the uses was developed The latest version of DocBook is the 5 1 and it is still maintained by OASIS The DocBook language is defined by a RELAX NG schema and by a set of Schematron rules that are integrated in the main schema Like the other descriptive markup languages its aim is to give tools to describe the meaning of the content rather than the way in which it should be presented Doc Book s elements can be clustered in three main groups the structural level elements the block level elements and the inline level elements The structural level elements are those ones that are used to describe the logical structure of the document To this category belong elements like ar ticle that is used to markup unnumbered collections of block level elements chapter that is used to define numbered collections of block level elements and part that is used to markup a titled collection of chapters Block level elements are used to markup sequential parts of the content but they can or they can not contain text Paragraphs lists and titles are is an American media company that publishes books and web sites and produces technical conferences regarding computer technology HaL Computer Systems is a Californian computer company funded in 1990 by Andrew Heller whose goal was to build computers for the commercial market Usenet is a worldwide distributed discuss
20. how it harmonizes the whole markup In the 3 3 43 next section I will explain how to abstract the logical structure of documents starting by the identification of the patterns that they contain Listing 3 7 A sample of Akoma Ntoso for the markup of a comma of an Italian law using the comma element lt xml version 1 0 encoding UTF 8 lt akomaNtoso xmlns http docs oasis open org legaldocml ns akn 3 0 CSD05 xmlns html http www w3 0rg 1999 xhtml bill meta identification source somebody gt lt FRBRWork gt lt FRBRthis value za bill 2003 09 04 76 main gt lt FRBRuri value za bil1 2003 09 04 76 gt lt FRBRdate name enactment date 2003 09 04 gt lt FRBRauthor as author href parliament gt lt FRBRcountry value za gt lt FRBRWork gt omissis lt identification gt lt meta gt lt preface id prfc1 gt lt p class heading gt REPUBLIC OF SOUTH AFRICA lt p gt lt p class subheading gt docTitle id 2 prfcl dcTtll TRADITIONAL LEADERSHIP AND P GOVERNANCE FRAMEWORK BILL lt docTitle gt lt p gt omissis lt preface gt omissis lt body id bdy1 gt lt chapter id chal gt lt num id chal nm1 gt CHAPTER 1 lt num gt heading id chal hdng1 INTERPRETATION AND APPLICATION lt heading 3 Structural differences in XML dialects
21. in the document or to markup placeholders in the text The figure displays the content model of the Akoma Ntoso noteRef element that is used to markup a reference to a note 48 3 Structural differences in XML dialects markeropt extension base attributes 8 coreopt noteRef Type extension of markeropt eer ihutes 8 notes 8 link a Figure 3 4 The content model of the noteRef elements in Akoma Ntoso 3 4 4 Blocks Blocks are elements that splits vertically the text and that can contains text inline elements or markers Usually there is only one content model shared by all the blocks This means that wherever any block is allowed all blocks are allowed too The p elements in the listing 3 8 are examples of blocks elements Listing 3 8 A fragment of an Akoma Notoso document showing the usage of blocks and inline elements lt xml version 1 0 encoding UTF 8 gt lt akomaNtoso xmlns http docs oasis open org legaldocml ns akn 3 0 CSD05 xmlns html http www w3 0rg 1999 xhtml gt omissis lt preface id prfcl gt lt p class heading gt REPUBLIC OF SOUTH AFRICA lt p gt lt p class subheading gt lt docTitle id prfcl dcTtl1 TRADITIONAL lt b gt LEADERSHIP lt b gt AND lt br gt GOVERNANCE FRAMEWORK BILL lt docTitle gt lt p gt lt preface gt omissis 3 4 lt akomaNto
22. is DocBook lt title gt lt para gt DocBook is an SGML dialect developed by O Reilly and HaL Computer Systems in 1991 It is currently maintained by the Organization for the Advancement of Structured Information Standards OASIS DocBook describes I3 Courtesy of www informatik tu cottbus de 34 3 Structural differences in XML dialects the content of articles books technical manuals and other documents Although DocBook is focused on technical writing styles it is general enough to describe most prose writing In this article I 11 discuss an XML variant of the DocBook DTD that is also available lt para gt lt section gt lt article gt 3 3 Examples of dialects used in the legal con text Laws judgments amendments and all the documents that belong to the legal context are resources whose logical structure is really difficult to markup with a generic markup language Due to the fact that parliamentary work flows can be completely different among different countries the laws should be written in different ways and should follow formal or informal rules dic tated by the country s legal system Moreover even in the same country it is possible to find different kinds of laws that should be written in different ways for example a bill can follow determinate drafting rules and an act can follow other rules In the last two decades a large community of scholar worked to accom plish thi
23. istics Satisfaction is about the users feelings while they interact with the edi tor or when they are performing tasks In section 6 2 3 1 will explain how I examined users satisfaction 6 2 91 6 2 1 Examining the efficacy The aim of this part of the test is to evaluate the editor s navigation system that should be used to navigate and manage files and to handle the editor s interface The navigation system is the very first part of the software to which the users are exposed To evaluate this part of the software I have chosen to submit to users nine tasks that are fairly representative of the interaction with the navigation system The nine tasks are listed below e Can you sign up to the system e Can you log in with your account e Can you change the editor s language to your preferred one e Can you import a Microsoft word s file from your desktop e Can you open one of the examples supplied by the editor e Can you save the file under a different name e Can you open the XML preview of the example you are looking at e Can you save the XML version of the document on your desktop e Can you log out These are critical tasks that users must complete successfully Therefore it is important to check both that the user can complete the tasks because a negative response results in a fatal interaction s issue and the ease of the interaction because an uneasy interaction with the navigation system can lead
24. listing it is possible to see that the main body of an Italian law is marked up with the element articolato that is basically a container of numbered hierarchical structure These kinds of numbered hierarchical structure are used all over the world for drafting laws even if other terms are used For this reason it is possible to generalize the schema and the legal XML standards of the third generation try to introduce an abstraction layer in order to guarantee the reusability of the standard Listing 3 3 A fragment extracted from an Italian law marked up using NIR lt xml version 1 0 encoding UTF 8 lt NIR xmlns http www normeinrete it nir 2 2 xmlns dsp http www normeinrete it nir disposizioni 2 2 xmlns h http www w3 org HIML 1998 htm14 xmlns xlink http www w3 0rg 1999 xlink tipo monovigente gt lt Legge gt lt meta gt lt descrittori gt pubblicazione norm 20130831 num 204 tipo gt redazione id 13G00144 nome norm gt urn valore urn descrittori lt meta gt lt intestazione gt lt tipoDoc gt monovigente lt tipoDoc gt lt dataDoc norm 20130831 gt 31 agosto 2013 lt dataDoc gt lt numDoc gt 101 lt numDoc gt lt titoloDoc gt Disposizioni urgenti per I perseguimento di obiettivi di razionalizzazione nelle pubbliche amministrazioni 13G00144 3 3 37 lt titoloDoc gt lt intestazione gt lt formulainiziale gt lt articolato g
25. ma allo stesso tempo flessibili per permettere di descri vere al meglio tutte le varie tipologie di documento che si possono incontrare in uno specifico contesto Uno dei contesti pi insidiosi quello giuridico I documenti legali e leg islativi possono avere strutture molto differenti anche se creati dalla stessa istituzione Ad esempio la struttura delle leggi italiane pu essere differente dalla struttura delle proposte di legge Per questo sono stati creati negli anni molti linguaggi XML per la marcatura di documenti legali Attualmente il pi completo Akoma Ntoso che con la sua architettura basata su pattern pu essere utilizzato per la marcatura di documenti provenienti da tutte le istituzioni Gli esperti di informatica giuridica sono stati pionieri nella creazione di editor di markup In questo contesto infatti gli autori di documenti debbono concentrarsi sulla struttura dei documenti e non possono essere distratti dalla complessa sintassi XML Insieme a linguaggi XML specifici sono stati creati anche editor di markup per le varie istituzioni Solitamente essi sono svilup pati come plug in di processori di testo esistenti come Microsoft Word o apache OpenOffice Questo molto importante in quanto fornisce agli autori 0 0 un interfaccia grafica gi nota e con la quale si sentono a loro agio Tut tavia molto difficile adattare questi software alla marcatura di documenti provenienti da istituzioni per le quali
26. ntoso experience In Proceedings of the V legislative XML workshop pages 67 86 Florence Italy European Press Academic Publishing 2007 Norman Walsh DocBook the definitive guide volume 1 Oreilly amp Associates Incorporated 1999 Oscar Wilde Oscar Wilde s wit and wisdom A book of quotations Courier Dover Publications 1998 7 0 131 WKLW98 Stuart Weibel John Kunze Carl Lagoze and Misha Wolf Dublin core metadata for resource discovery Internet En gineering Task Force RFC 2413 222 1998 132 BIBLIOGRAPHY List of Figures 2 1 The output generated by the lines of code in the listing Sar 17 3 1 A sample of a manuscript s description 30 3 2 The content model of the act element in Akoma Ntoso 46 3 The content model of the article elements in Akoma Ntoso 47 3 4 The content model of the noteRef elements in Akoma Ntoso 48 Je Led ei 56 3 bo 7 M BOE RUE d E eg Ead br 4 8 A screenshot of the Norma editor 60 TETTE he ee S 62 ID 63 LU a ME UE E de 73 5 2 A screenshoot of the LIME top bar 74 SEE 75 IT 76 Le ie T7 5 6 The LIME editor architecture rn 78 th Boga 80 puke RUE Sane posse kee 107 TETE ee ee tei 112 6 3 Average Expectation and Experience Ratings per Task 115 133 LIST OF FIGURES List of Tables 6 1 The sheet used to collect user s data 102 6 2 The summary of the tes
27. of legislative documents they are able to simply markup the main structure of acts and bills such as their preface their preamble and their article and sections The test also highlighted that users are quite satisfied while using LIME They were asked to foresee the expected effort to complete some tasks and after the completion they were asked to indicate the real effort they experienced The majority of the users found the main part of the tasks as simple as they expected or in some situations they found them even simpler The test also highlighted some issues related to the LIME usability and some bugs that will be fixed in the next version of the editor LIME is currently requested by several parliaments political and apoliti 1 0 11 cal organizations from overall the world to markup their legal and legislative documents It is also requested by history scholars to markup descriptions and transcriptions of ancient manuscripts Hopefully after the release of the first beta version many language plugins will be developed by third parties and that LIME will be widely used both in the legal context and in other contexts 12 1 Introduction Chapter 2 The markup of structured documents The aim of this chapter is to give an introduction about structured doc uments about their evolution in time and about the markup languages used to create them In the first section I will produce a definition of structured d
28. persons to lec t E 34 Disegno Forme I8 420M D L A ERMEL Correct Pg 1 Sez 1 1 2 A9 1cm Ri 12 Coll Italiano Ital Figure 4 3 A screenshot of the Norma editor 4 4 2 Bungeni Editor The Bungeni editor is the successor of the Norma Editor BvE11 and it is a markup editor built on the Apache OpenOffice P suite It is component of the Bungeni project P and was developed under the supervision of the Apache Open Office is an open source software suite for word processing It stores the data in an international open standard format called Open Document Format ODF is a Parliamentary and Legislative Information System that aims to make Parliaments more accessible to citizens 4 4 61 United Nations Department for Economics and Social Affairs The Bungeni editor solved many problems of the Norma editor First of all it was created on the Apache OpenOffice suite This means that the editor can be released as an open source software because OpenOffice is released under the apache license Ros04 This also means that documents are natively stored in an XML format the ODF format and this is of great help because they can be simply translated into any other XML format using XML related technologies like XSLT Another improvement in the Bungeni editor is that it is designed around Akoma Ntoso In this way the editor can be adapted to
29. the elements enabled in that position Figure 5 4 displays the LIME markup menu for Akoma Ntoso documents The last part of the LIME editor s interface is the document outline which is displayed in figure It is used to display the outline of the 76 5 LIME a parametric editor for structured documents document This is useful to see the hierarchical nesting of the already marked up document and to quickly navigate among them NS NS DL NS DU DS DU UE Figure 5 4 A screenshoot of the LIME markup menu 5 5 77 Outline 3 60 3 preface legislature docType gt docDate gt docNumber legislature C docType docTitle 3j authorialNote 3j preamble 4 body H conclusions Figure 5 5 A screenshoot of the LIME documents outline LIME architecture revolves on the getting real method God06 For this reason we started from the application interface and created the system ar chitecture on it 5 5 LIME architecture LIME is based on a four tier architecture The application logic of LIME completely relies on its client side components and the server side compo nents are charged to manage documents and database transaction LIME uses two database The first one is a classic relational MySQL database and it stores information related to users and statistic information about the access to the system The second database is an eXist database instance and is used to store
30. the preface of the laws But like CEN Metalex it also supplies generic elements with the same name of the patterns in order to allow also the markup of elements that are not provided as basic elements by the schema For instance the comma element that is largely used in the Italian laws is not supported by Akoma Ntoso but it can be marked up using the line of code in the listing that have exactly the same semantic meaning of the markup in listing 3 6 Listing 3 5 A sample of Akoma Ntoso for the markup of a comma of an Italian law using the hcontainer element lt article id art1 gt lt hcontainer name comma id artl coml lt num gt comma 1 lt num gt lt content gt A hierarchical container with the name comma is equivalent to a comma element lt content gt lt hcontainer gt lt article gt Listing 3 6 A sample of Akoma Ntoso for the markup of a comma of an Italian law using the comma element article id art1 gt comma name comma id artl_coml gt lt num gt comma 1 lt num gt lt content gt A hierarchical container with the name comma is equivalent to a comma element lt content gt lt comma gt lt article gt The fragment of document in listing3 Tis a full example of Akoma Ntoso and it is extracted by a markup of a Swiss law In this example it is possible to see how the use of patterns in the markup makes immediately identifiable all the hierarchical containers and
31. the user to think that the editor is not working properly 6 2 2 Examining the efficiency This part of the test is intended to evaluate how much effort users make when they use the editor to markup a legislative document In order to 92 6 Evaluation of LIME s user experience markup a legislative document users must identify the legal parts of the document and must assign to each of them the correct label On the one hand the markup s tasks are secondary because the users start to use them only after they feel comfortable with the navigation system On the other hand they are fundamental tasks because they are performed to accomplish the editor s main objective In order to evaluate the markup features of the editor I asked the testers to perform the following tasks e After a partially marked up document was opened Can you set the preface of this document Can you set the document date Can you set the main body of the document Can you set an article its number and its heading Can you markup a bold text in the article that you created Can you create a table in the article that you created Can you set a quoted structure in the article that you created e After a complete marked up example was opened Can you show me the subsection 2 of the section 3 of this docu ment Can you un mark the preface of this document and all the elements that it contains Can
32. was originally designed to allow a large scale elec tronic publishing but it is currently also used to exchange a large variety of data on the web The first version of XML XML 1 0 was defined in 1998 and it is cur rently in its fifth edition published on November 26th 2008 The second version of XML XML 1 1 was published on February 4th 2004 and it is currently in its second edition published on August 16th 2006 Prior to the fifth edition XML 1 0 differed from XML 1 1 because it had stricter require ments for characters used in elements names attribute names and unique identifiers The fifth edition adopted the mechanism of XML 1 1 specifying that only certain characters are forbidden in names and everything else is allowed Even if no organization has announced plans to work on an hypo thetical version 2 0 a skunkworks written by one of the XML developers contains some ideas of what an XML 2 0 could be As I said previously a big part of XML comes from SGML unchanged XML like SGML uses the angle brackets syntax it separates the logical and the physical structure of the document and it supports a grammar based va lidity through a document type declaration It also allows mixed content and processing instructions used to separate the processing from the representa tion and it permits the separation of data and metadata by using elements Extensible Markup Language SW XML SW Skunkworks 10 February 2002 Tim Bray
33. when they were using it The test highlighted that for the majority of the tasks LIME is usable Even if users have no experience in Akoma Ntoso they were able to markup documents using many of the elements that the standard supplies This proves that the LIME interface actually guides users through a markup pro cess that results in a complete and if needed valid XML document The usability test also pointed out some LIME weaknesses Not all users were able to use properly the save functionality and some of them failed to figure out how to markup some elements specifically related to the legal context The problems of the storage system as said before can technically be 7 0 121 solved by enabling plugins But this would not fix the usability issues re lated to the interface s parts that allow users to save documents by taking advantage of FRBR notation It would be ideal if LIME users could prop erly use the storage system even they are not skilled in FRBR This can be achieved by improving the save functions parts of the interface and by making them in some way more easily learnable By doing so users would spend few minutes to learn the usage of the interface but then they could take advantage of it In order to understand how to create a learnable inter face I will organize other hackatons and comparative usability tests in which I will ask users to exploit different interfaces to complete the same task Usability issues
34. 18b 18c 19b 19c How difficult was to complete this task Very difficult O O O O O O O Very easy Can you set an article its number and its heading How difficult was to complete this task Very difficult O O O O O O O Very easy Can you markup a bold text in the article that you created How difficult was to complete this task Very difficult O O O O O O O Very easy Can you create a table in the article that you created How difficult was to complete this task Very difficult O O O O O O O Very easy Can you set a quoted structure in the article that you cre ated How difficult was to complete this task Very difficult O O O O O O O Very easy Can you show me the subsection 2 of the section 3 of this document How difficult was to complete this task Very difficult O O O O O O O Very easy Can you un mark the preface of this document and all the elements that it contains How difficult was to complete this task Very difficult O O O O O O O Very easy Can you save this document as a new expression in the same work How difficult was to complete this task Very difficult O O O O O O O Very easy 6 Evaluation of LIME s user experience 102 Table 6 1 The sheet used to collect user s data
35. ALMA MATER STUDIORUM UNIVERSIT DI BOLOGNA SCUOLA DI SCIENZE Laurea Magistrale in Scienze di Internet PARAMETRICS EDITORS FOR STRUCTURED DOCUMENTS Tesi di Laurea in Interazione Persona Computer Relatore Presentata da Chiar mo Prof Luca Cervone Fabio Vitali Correlatore Chiar ma Prof ssa Monica Palmirani Sessione II 2012 2013 To my parents CARMELA and EMANUELE and to my childhood town SANNICANDRO GARGANICO they built my hardware and developed my operating system To BOLOGNA and the people that I have met here they have fixed so many bugs Contents 0_ Introduzione 1 Introduction 2 The markup of structured documents 3 2 1 Structured documents h Pe edu PE Ed ea ee OX X 2 3 1 The Standard Generalized Markup Language perenni T HIT Structural differences in XML dialects 3 1 Different dialects for different contexts 3 2 Examples of dialects for textual resources 3 2 1 Text Encoding Initiative 2 2 lt 5 442454 24 us 3 2 0 _ DocBook 3 3 Examples of dialects used in the legal context 3 3 1 Norme In Rete 3 3 2 CEN Metalex 3 3 3 Akoma Ntoso 13 13 15 18 18 21 23 24 CONTENTS CONTENTS 3 4 Abstraction of the XML dialects 45 3 4 1 Comtamers lt e kocsoe e e x oko ee a i a 45 3 4 2 Hierarchical containers 46 dipl MR 47 Sdi Blocks cia EE n EON Se 48 ciaoo iaia a En 49 TT
36. BER DOCKET NUMBER RU 11 529 DOCKET NUMBER e DOCKET NUMBER CS 3 3 DOCKET NUMBER DOCUMENT TITLE Legge federale sulla esecuzione e sul fallimento DOCUMENT TITLE SHORT TITLE LEF SHORT TITLE dell DOCUMENT DATE 11 aprile 1889 DOCUMENT DATE Stato DATE 1 gennaio 2012 DATE L Assemblea federale della Confederazione Svizzera RECITALS RECITAL visto l articolo 64 della Costituzione federale2 Cost 32 NOTE Nuovo testo giusta il n I della LF del 24 mar 2000 in vigore dal 1 gen 2001 RU 2000 2531 FF 1999 8077 8458 NOTE CS 1 3 A questa disp corrisponde l art 122 cpv 1 della Cost del 18 apr 1999 RS 101 ENACTING FORMULA decreta BODY TITLE NUMBER Titolo primo NUMBER HEADING Disposizioni generali HEADING SECTION NUMBER I NUMBER HEADING Della organizzazione HEADING ARTICLE NUMBER Art 14 NUMBER NOTE Path act gt body gt title Titolo primo gt num Figure 5 3 A screenshoot of the LIME word processor In order to allow users to do it LIME supplies a markup menu in the right side of the interface that is contextual to the part of the document users are editing For example if users are marking up an Akoma Ntoso document in the first step of the markup the menu will display the top level elements made available by Akoma Ntoso Later when they have already marked up some chapters or articles users can position the mouse s cursor inside these elements and the markup menu will display
37. D cC Bet VES ee Be es 49 50 4 Software for XML markup 53 4 1 WYSIWYG editors s sche E UR xor x a 53 4 2 Generic XML editors aoa a a a 54 4 2 1 oXygen XML editor 42 ome a 55 RE CE ee oe 57 4 4 Markup editors for the legal context 58 4 4 1 Norma editor 1 54 a x 6 xk EE OS 59 See Em ee E ee E ee 60 Pow ee ee eee 62 See i a EM RP AP ee E 63 65 ue an Or Hm 65 boue ee 66 66 5 2 2 Frameworks for cross browser software 67 SI Ettore 68 ndg ICT ee esca e Eae ai a i 68 P3 ri nn de 68 5 2 5 eXist Database si end 69 5 2 5 1 FRBR storage dsc ee te 9o eS ee YS 69 5 2 5 2 XQuery language and XSLT 2 71 5 2 6 XML patterns and XML guidelines 71 CONTENTS iii COTTE CERETTO TO 71 a aes ees 72 5 4 Overview of LIME features a 73 Bad amp ea m te ee eS 77 DANILO a gee 79 Lama Sa es REI 80 ee a ee ee ee ee ee ee ee 81 5 6 1 XML guidelines and patterns used by LIME 81 5 6 1 1 Patternless the patterns wildcard 82 5 6 1 2 Guidelines for elements unique identifiers 83 5 6 1 3 Guidelines for elements classes 83 or ak ar Ao mom A RE ae 84 5 6 2 1 LIME language plugin 85 5 7 Evaluating markup editors usability 87 89 CRE E O dn ee 89 6 2 Goals of the tes ces ewe a el ee RR ERS 90 6 2 1 Examining the ellicacy 22 R9 Rx ee ees 91 6 2 2 Examining the efliciency
38. LIME interface must be improved The other two tasks that were rated less than fifty percent were the ones that involved the user in the markup of legal elements of documents I no 6 4 113 ticed slowdowns when users had to markup an article with its number and its heading and when they had to markup an element called quoted structure In order to markup an article its number and its heading users had to select the text of the article in the document and click the set article button on the right Then they had to select the number of the article and click the button set num in the markup toolbar and again they had to select the heading of the article and click the button set heading in the markup toolbar The problem in this workflow was that even if users had no problem marking up the article they struggled to find the buttons to markup the number and the heading In order to markup a quoted structure users had to select the text that had to be marked up and then they had to click on a toolbar called com mon elements finally they had to find and click the button called set quoted structure The common elements toolbar contains buttons grouped by their thematic area Users failed to find the button to markup a quoted structure because the majority of them did not know which was the correct thematic area that had to be open and abandoned before they had inspected all the areas There are other parts of the interface that must be
39. LandSVG xml D Projects eXml_SVN samples docbook w5 sampleMathMLandSVG xml oXygen XML Author lelak File Edit Find Project Options Tools DocBook5 Document Window Help Dm c m a dan a p p DEEPA DINE Ker smote We amp OK iS d q Ver i amp 69 s iB2ZUa NSME HEEE LE BARS Project o A X esampleMathMLandsSvGxmi x 4 b E Attributes E pae Ell aride secti set ttle Ho m Wh wsd e qa T3 BB attribute Value 7 B xe Section 2 SVG frameworks arch i x A F J dita Section 2 1 Embedded SVG audience 4 k docbook condition J v conformance ae A i dir The SVG content is wrapped into an Pimageobject4 Li images 1 P linkend E photo album xml element E Z sample xml vi 1 remap e sampleMathMLandSVG xml ed G figure revision i lt e sampleXInclude xml m J mio 1 Attributes Model x lt e section1 xml Xe section2 xml m Transformation Scenarios sampleMalo 9 X Outline D a Elemen ne Fill Type Filter text Q ST para Inline equation z 3 Association Scenario sect2 External MathML Equation bb 44 4 DocBook 5 9 4 set SVG DocBook HTML cda title SVG DocBook HTML Chunk 4 sect2 Embedded SVG Images DocBook POF ts title Embedded SVG Images P DocBook XHTML T para The SVG content is wrap DocBook XHTML Chunk 4 E figure An SVG figure DocBook WebHelp As title An SVG figure This is an inline bbb 4444 image DocBook WebHelp with Feedback 4 mediaobject Y S z 7 DocBook
40. M Document structure Common elements DOCUMENT NUMBER 261 1 DOCUMENT NUMBER DOCKET NUMBER RU 11 s29 DOCKET NUMBER e BOCKET NUMBER cS 3 3 DOCKET NUMBER GGCURERF STRE Ls DGCURENT TITLE SHORTT er EROR TITLE Er E del EGCURERT DATE 11 aprile 1869 DOCUMENT DATE Stato DATE 1 gennaio 2012 DATE sa I L Assemblea federale della Confederazione Svizzera RECITALS RECITAL Visto l articolo 64 della Costituzione federale Cast 33 NOTE Nuovo testo giuste il n 1 della LF del 24 mar 200 in vigere dal 1 gen 2001 RU 2000 2531 FF 1999 0077 0450 NOTE CS 1 3 A questa disp corrisponde lart 122 cpv 1 della Cost del 18 apr 1999 RS 101 ENACTING FORMULA decreta Tm GRISE ritoo primo NUMBER READING pisposizioni generali FEASINE paragraph 3 SECTION rum HEATING Della organizzazione HEADING ec ARTICLE NUMBER rt 14 NUMBER wore Path Figure 5 1 A screenshoot of the LIME editor 5 4 Overview of LIME features The LIME editor permits to markup documents in various XML lan guages by using an interface that is quite similar to the one of desktop word processors LIME allows users to register and to create and save their documents no matter the language they are marked up in the cloud So users can import documents edit them and save them in their dedicated eXist database space These functionality are all managed using the to
41. ML Addison Wesley 1988 Luci Berkowitz Karl A Squitier and William Allen John son Thesaurus Linguae Graecae canom of Greek authors and works Oxford University Press 1986 Alexander Boer and Tom van Engers A metalex and metadata primer Concepts use and implementation In Legislative XML for the Semantic Web pages 131 149 Springer 2011 James Clark Steve DeRose et al Xml path language xpath 1999 James Clark and Makoto Murata Relax NG specifica tion 2001 R Cover Theological markup language thml The XML Cover Pages http www oasis open org cover thml html 2000 7 0 125 CRD87 James H Coombs Allen H Renear and Steven J DeRose Cro06 D I78 Fie00 G 05 GMEO7 GNK 99 God06 Gol81 Gol91 Markup systems and the future of scholarly text processing Commun ACM 30 11 933 947 November 1987 Douglas Crockford The application json media type for javascript object notation json 2006 Mary E D Imperio The voynich manuscript an elegant enigma Technical report DTIC Document 1978 Roy Thomas Fielding Architectural Styles and the Design of Network based Software Architectures PhD thesis UNI VERSITY OF CALIFORNIA IRVINE 2000 Jesse James Garrett et al Ajax A new approach to web applications 2005 Danny Goodman Michael Morrison and Brendan Eich Javascript bible John Wiley amp Sons Inc 2007 David T Gering A
42. X few people write in plain TeX anymore The current version is LaTeXe end document 18 2 The markup of structured documents 2 3 Descriptive markup languages Descriptive markup is used to ascribe a semantic meaning to each part of the document through labels called tags In this way the author can focus on the document s logical structure regardless of any particular treatment of rendition of it Unlike the procedural or the presentational markup the goal of the descriptive markup languages is to institute a one to one mapping between the logical structure of the text and the markup of elements that constitute it A first attempt of markup standardization for the encoding and analysis of literary texts was made by the COCOA encoding scheme that was originally developed for the COCOA program in the 1960s and 1970s Rus67 but that was later used as an input standard by the Oxford Concordance Pro gram in the 1980s HM88 and by the Textual Analysis Computing Tools in the 1990s LBM 96 Another language that reached a certain level of stan dardization was the Beta transcription encoding system that was used to transcript and encode the classical Greek texts BSJ86 The first descriptive markup language that was really standardized and that accomplished to the task of formally separating information from meta data also ensuring interchange reusability and portability was the Standard Generalized Markup La
43. a Earlier in section 2 1 I gave a definition of logical structure that must be intended as the description of the objects forming the document and the relations between them The content of a textual resource is obviously the readable text inside it and the metadata are 16The XML User Interface Language is a markup language developed by Mozilla whose aim is creating user interfaces of applications like the Firefox web browser 7The Business Process Execution Language is an OASIS standard executable language that enables users to describe business process activities 2 4 25 data that describe other data In some situations metadata can be mixed in the content of the document but they are often peripheral data contained in resources related to the main one for example a book s cover or in a specific part of the document that if missing does not jeopardize the actual readability of the document For instance a book is constituted by chapters sections the actual text of the chapters and of the sections and by other relevant information like its author the publication date and the publisher s name On the other hand a law is made by sections and their text pointed lists and their text and by a lot of legal info such as the efficacy date the number of the law and the law s proponent In section 2 2 I described three different families of markup languages and in section 2 3 I explained thoroughly the descriptiv
44. act the orig inal javascript s ones and behave in the same way on different browsers Most recent frameworks also allow to quickly create animations visual effects and to easily communicate with the server side of the software Other javascript frameworks known as component based frameworks Lew98 allow developers to create with few lines of code visual elements that are similar to the ones used in desktop applications These are the most suitable ones for the creation of a parametric markup editor 68 5 LIME a parametric editor for structured documents 5 2 2 1 ExtJS Ext JS is a component based javascript framework for building interactive web applications It allows to create easily the core of the appli cation by following the Model View Controller pattern ExtJS comes along with a big range of user interface widgets but it is possible to extend it by creating new components and these can be combined with the default ones to create rich user interfaces ExtJS is completely cross browser and applications that exploit Ext JS can be used both on all browsers running on all operating systems and on modern tablets and smart phones 5 2 3 TinyMCE TinyMCE is a platform independent web based javascript HTML WYSIWYG editor control It can convert HTML5 text area fields or other HTML5 elements into editor instances It relies on an user interface very similar to the one of the most used word processors like Micro
45. age system and to the interface that is probably not simple to understand Future versions of LIME will try to fix these issues and to improve the LIME overall usability Chapter 7 Conclusions In this essay I described how to create a parametric editor for structured documents and then I described LIME a parametric and language indepen dent markup editor Currently there are many software suitable for the creation of structured documents Some of them are able to markup all XML languages but users need a good knowledge of the XML language they wish to use These editors are the WYSIWYG editors Other editors permit to markup XML documents without a deep knowl edge of the XML language but they are created for only one specific dialect These are called markup editors and some of these are used in the legal con text LIME is a mix of these two types of editors On the one hand LIME allows users to markup documents through many XML languages and on the other hand it drives users to create a correct markup even if they do not know the XML language In order to be independent LIME relies on parameters Its architecture is completely based on JSON configuration files and by creating packages of these files called language plugins it is possible to allow LIME users to markup documents through any XML language Currently LIME supports three XML languages Akoma Ntoso in its 2 0 and 3 0 versions TEI and 119 120 7 Conclusions
46. al structure Descriptive markup languages allow authors to describe objects and the relations between them by labeling them with tags In this way authors can focus on documents logical structure regardless of any particular treatment of their rendition The first really standardized descriptive markup language was the Stan dard Generalized Markup Language whose purpose was to define a gener alized markup language for documents By using SGML it is possible to markup structured documents according to a Document Type Declaration SGML laid the foundations for the two most common descriptive markup languages Hyper Text Markup Language and eXtensible Markup Language HTML is currently the most suitable technology to create web pages Web browsers interpret the HTML tags in the documents and present the whole document in a readable or audible way HTML and chiefly HTML 5 are good descriptive markup languages but they are not intended to markup documents that must have their specific in context semantically relevant el 1 0 ements In these situations XML must be used XML defines a set of rules aimed to produce human readable and machine readable documents It was originally designed to allow a large scale electronic publishing but it is currently also used to exchange data on the web There are many XML dialects used for the markup of text based re sources These XML languages must be specific but at the same time they must be a
47. aptive learning content in technical domains JCSA 3 2 91 108 2006 Brad A Myers A brief history of human computer interac tion technology interactions 5 2 44 54 1998 William Newman Design case study the bravo text editor interactions 19 1 75 80 2012 Edward T O Neill Frbr Functional requirements for bibli ographic records Library resources and technical services 46 4 150 159 2002 Leslie M Orchard Ara Pehlivanian Scott Koon and Harley Jones Professional JavaScript Frameworks Prototype YUI ExtJS Dojo and MooTools Wrox Press Ltd 2009 128 BIBLIOGRAPHY P 00 PB02 PB03 PC09 PC13 PCRO9 PLO9 Steven Pemberton et al Xhtml 1 0 the extensible hy pertext markup language W9C Recommendations pages 1 11 2000 Monica Palmirani and Raffaella Brighi Norma system A legal document system for managing consolidated acts In Database and Expert Systems Applications pages 310 320 Springer 2002 Monica Palmirani and Raffaella Brighi An xml editor for legal information management In Electronic government pages 421 429 Springer 2003 Monica Palmirani and Luca Cervone Legal change man agement with a native xml repository In Proceedings of the 2009 conference om Legal Knowledge and Information Systems JURIX 2009 The Twenty Second Annual Con ference pages 146 155 Amsterdam The Netherlands The Netherlands 2009 IOS Press Monica Palmirani and Luca Cerv
48. arkup 6 4 6 4 2 Analysis of the efficacy The efficacy of the editor was evaluated asking the users to complete nine binary tasks These tasks had a time limit and if users did not complete them in time the task was classified as not completed Table shows the tasks used to evaluate the editor efficacy their time limit and their code Table 6 3 The tasks used to evaluate LIME efficacy Task Task code Time limit sec Sign up to the system T1b 180 Log in to the system T2b 30 Change the editor s language T3b 30 Import a Microsoft word s file from the user s desktop ay Ep Open one of the examples supplied by T5b G0 the editor Save a file under a different name T6b 60 Open the XML preview of a file T7b 60 Save the AL version of a document T8b 60 on the user s desktop Log out from the system T9b 30 In table and in figure the results of the efficacy evaluation are displayed 106 6 Evaluation of LIME s user experience Table 6 4 The summary of the tasks completed by users 1 indicates completed tasks 0 indicates not completed task Tib T2b T3b T4b T5b T6b T7b T8b T9b avg pl 0 1 1 1 0 1 1 1 1 78 p2 1 1 1 1 0 1 1 1 1 89 p3 1 1 1 1 1 0 1 1 1 89 p4 0 0 1 1 0 1 1 1 1 67 p5 1 1 1 0 1 1 1 1 1 89 p6 1 1 1 1 0 1 1 1 1 89 p7 1 1 1 1 1 1 1 1 1 100 p8
49. ata means that there are no big surprises here although combining this data with the one received in the efficiency evaluation of the same tasks it is clear that there are many opportunities of improvement The last sector is the upper right one To this sector belong tasks that users thought would be easy and indeed they were This means that these features of LIME must not be changed and that it is already usable for the completion of those tasks Having eleven points in the upper right sector of the diagram and three points in the upper left sector I can conclude that except for five tasks users are overall satisfied when they use LIME 6 5 LIMESs strengths and weaknesses In this chapter I analyzed the LIME usability Firstly I described the usability test that I created to evaluate markup editors and then I described the results that I obtained applying it to LIME Results demonstrated that LIME is effective and efficient and that users are quite satisfied when they use it LIME seems to be very usable for markup Users take advantage of its functionality to quickly mark up the structure of legal documents Except for some tasks they feel comfortable with the markup toolbar supplied by the editor Moreover they can use easily the interface when they have to navigate through documents and their various views However other tasks like the save system frustrated users This is due 6 5 117 both to the particular LIME stor
50. ate Modifies document Editor Ask to update Figure 5 7 The interaction among LIME client side components 5 5 2 Server side components The server side components have three main tasks proxying parsing and translating The requests proxy is a PHP module that simply dispatches the requests towards the two kinds of database used by the application When a docu ment is requested or must be saved the module retrieves or sends the document to the web services built on the top of the eXist database When users information must be retrieved or must be saved from the MySQL database it sends the request to the MySQL server and returns the informa tion to the client The document parsing component is used for the smart markup function ality provided by the editor This permits to parse the document trying to automatically find and markup some parts of the document For instance if 5 6 81 a law file opened in the editor contains some references to other laws they will be found and labeled as ref elements The document translations component is the module that translates the document from the HTML version created in the editor into the XML format that users need 5 6 Three commandments to be parametric and language independent In order to be parametric and language independent LIME relies on three main concepts XML guidelines XML patterns and JSON configura tions files Guidelines and pattern
51. content models and we can assign one of these six patterns to each element In chapter 5 section 5 6 1 1 will explain that exceptions i e elements that do not belong to any pattern can simply be labeled as patternless and this does not jeopardize a hypothetical generic markup process But well designed schemata are not the only think that we need in or der to enable such kind of markup process We need powerful and usable software that guide us thorough a markup process that results in documents compliant to the specific schema and to the patterns As for the legal standards in the last two decades example of software were produced These software aim to allow the users to markup legal and legislative documents even if they do not know the XML syntax In the next chapter I will describe some of the legislative markup software and I will compare them to the software used for the pure XML editing 52 3 Structural differences in XML dialects Chapter 4 Software for XML markup This chapter aims to introduce the software used for XML markup Firstly I will give an introduction about WYSIWYG editors In the sec ond section I will describe generic editors for XML documents In section I will introduce markup editors and later in section I will describe the most common markup editors used to markup legal and legislative docu ments Lastly in section 4 5 I will discuss the challenges that markup editors still have to meet i
52. d to improve the user experience 6 3 The LIME s user experience test The LIME s user experience test was built on Akoma Ntoso markup lan guage and was intended to capture both users performance and satisfaction For these reasons it was submitted using specific methodologies and users were chosen in order to meet specific requirements In this section I will explain the methodologies that I used to submit the tests how I chose the testers and eventually I will report the complete questionnaires submitted to the subjects 6 3 95 6 3 1 Methodology The user experience s test was divided in four sessions A registration session a presentation session an effort foresee session and a task perform ing session All the subjects were volunteers and I will explain in the next section how I recruited them For the registration I asked testers their age their degree of experience in certain fields and how often they uses some kinds of software I have chosen not to ask their name and their surname in order to ensure their anonymity during the results evaluation and for this reason I assigned to each of them an individual ID number The other information were asked in order to make it possible to cluster the results and infer more specific data During the presentation session a teacher in legal informatics and cre ator of Akoma Ntoso Monica Palmirani explained testers the basis of the language and shared all the k
53. describe the whole plugin Each file describes a different layer of the plugin from the user interface to the patterns mapped to each element and the set of elements specified by the language The root directory of all languages configurations files is languagesPlug ins in which some directories and one file are stored The resources contained in the languagesPlugins folders are the following e One or more directories having the name of the XML language whose markup must be enabled in the editor These directories contain two nested folders and a file client an optional folder containing plugins written in pure javascript interface the main folder of language configuration structure json a JSON file containing the structure of interface folder e default this is a directory containing a dummy language which con tains some default files that every language can override e config json A configuration file that contains the list of the languages enabled in the editor The interface directory contains all the files used to describe the configuration of a specific XML language It contains the following files 86 5 LIME a parametric editor for structured documents e viewConfigs json a file containing configurations about views in the editor it allows to enable or disable views e markupMenu json in this file all the elements of the language with their patterns are stored e markupMenu rule
54. digm in the smalltalk 80 system Journal of object oriented program ming 1 3 26 49 1988 Leslie Lamport LaTeX A Document Preparation System Addison Wesley 1 edition 1986 Ian Lancashire John Bradley Willard McCarty Michael Stairs and TR Wooldridge Using TACT with Electronic Texts A Guide to Text analysis Computing Tools Version 2 1 for MS DOS and PC DOS Modern Language Associ ation of America 1996 Scott M Lewandowski Frameworks for component based client server computing ACM Computing Surveys CSUR 30 1 3 27 1998 7 0 LRV 08 Mar92 Mar94 Mei03 MOMGSFM06 Mye98 New12 O N02 OPK J09 Effie Law Virpi Roto Arnold P O S Vermeeren Joke Kort and Marc Hassenzahl Towards a shared definition of user experience In CHI 08 Extended Abstracts on Human Fac tors in Computing Systems CHI EA 08 pages 2395 2398 New York NY USA 2008 ACM Nenad Marovac Document recognition concepts and im plementations SIGOIS Bull 13 3 28 38 December 1992 Fred Garth Martin Circuits to control Learning engineer ing by designing LEGO robots PhD thesis Massachusetts Institute of Technology 1994 Wolfgang Meier exist An open source native xml database In Web Web Services and Database Systems pages 169 183 Springer 2003 Iv n Mart nez Ortiz Pablo Moreno Ger Jos Luis Sierra and Baltasar Fern ndez Manj n Using docbook and xml technologies to create ad
55. ding this essay and if you fell in love with me while reading you can give me a call at 39 3482627545 Tutto cominci con un commodore 16 Ma che bella invenzione Passavo ore a giocare con quell artefatto Io lui mio fratello Angelo e scimmia magica formavamo un quartetto perfetto Poi arrivarono l Amiga 500 super frog e sensible soccer Spettacolo Ma con l Amiga non c erano solo i giochi c erano anche QBasic e l enciclopedia del computer Diamine quei cosi rettangolari non servivano solo per giocare Si potevano programmare Ringrazier per sempre i miei genitori per avermeli comperati Frequentai le scuole medie e le superiori arriv la playstation e il mio 486 Conobbi un losco figuro col nome francese il fratello di un amica di famigla e suo padre che mi insegnarono a fare il vino e a berlo e con alcuni cugini e cugine imparai che le feste non erano poi cosi male E intanto gli anni delle superiori passavano ero curioso e sperimentavo tutto Anche le cose pi pericolose Tipo perdere un anno di scuola E arriv la dreamcast e la prima connessione internet E arriv la prima chat con una tizia che mi disse faccio la webmaster a Londra e mi pagano molto bene La webmaster Porca paletta c gente che si guadagna da vivere con questa roba Quel piccolo paesino che tanto mi aveva dato non poteva insegnarmi a diventare uno scienziato dei computer Dovevo andare 139 140 LISTINGS via E Bologna fu Bologna B
56. ditors and I will show the results that I obtained applying it to LIME Chapter 6 Evaluation of LIME s user experience The aim of this chapter is to describe the process for the evaluation of LIME s user experience In the first section I will synthetically explain what is user experience and what are the goals of a generic user experience s test In the second section I will show the goals of LIME s user experiences test and in the third section I will deeply describe the methodologies used to develop the test to submit it and to collect the data that it produces In the fourth section I will analyze the results of the test and eventually I will point out the strengths and the weaknesses of the LIME editor 6 1 The study of the user experience Even if many scholars gave their own definition of user experience here after UX LRV 08 the most of them agree that it describes the thoughts the feelings and the perceptions that result from an interaction between a human and an artifact no matter if it is a computer or a corkscrew For this reason the aim of a user experience s test is to collect the behav iors the attitudes and the emotions that emerge from an interaction with a system 89 90 6 Evaluation of LIME s user experience For example if we are analyzing the user experience of a person who wants to bake parmigiana we will count how many movements he performs in order to open the oven pu
57. documents logical structure Indeed in order to abstract the logical structure of a docu ment we need to highlight the parts of its structure that recurs several times and we need to markup them in the same or in a similar way And identify a pattern means exactly the same think We find content models that are shared between more elements There are a lot of officially recognized patterns 5 that can describe more or less all the most used content models Akoma Ntoso uses six of these patterns that I will describe in the following sections 3 4 1 Containers A container is an element that contains sequences of specific elements Each container has its specific list of contained element For this reason it is not possible to create a generic content type for all the containers but they share the same basic characteristics Some of the elements that they contain can be optional and they can not contain directly text An example of container in Akoma Ntoso is the act element that is used to markup the main container of an act Figure 3 2 shows the content model s diagram of the act element as extracted from the Akoma Ntoso schema I The XML patterns community currently lists twenty eight patterns 46 3 Structural differences in XML dialects hierarchicalStructure O attributes name contains Type versionType Default originalVersion act coverPage Type emu Ty
58. e of difficulty they found in accomplishing the task In section 6 3 3 I will report the questionnaires and the sheet that I used to collect the information 6 3 2 Choosing the testers In order to recruit the testers we at CIRSFID organized a hackaton Its objective was to markup by using Akoma Ntoso as many documents as possible in five hours During the hackaton I asked the users to join me in a separate room to perform the fourth session of the user experiences test Ten people joined the hackaton and ten people agreed to do the test In section 6 4 1 I will summarize the users information 6 3 3 The complete test The following are the questionnaires submitted to the users and the sheets that I used to report information The questionnaire submitted before the users performed the tasks About you 1 Your ID number 2 How old are you Iam ______ years old 3 What is your knowledge in computer science None O O 0 Im an eger 1A hackaton is an event in which individuals are involved to collaboratively contribute to a project A lot of software or software functionality were prototyped or implemented during hackatons for example the Facebook s chat PL09 6 3 97 4 What is your knowledge in law and jurisprudence None O O O O O O O I m an expert 5 What is your knowledge in legal informatics Neme G O 0 0 0 O 0O Im an egit 6 How often do you use computers
59. e Web Consortium recommendation and it is the most used markup language to create web pages Web browsers interpret the HTML tags in the documents and present the whole document in a readable or audible way The history of HTML started in 1980 when Tim Berners Lee proposed ENRIQUE a system whose purpose was to allow researchers to share docu ments Later in 1989 Berners Lee wrote an article where he advanced the idea of an internet based hypertext system BL89 In 1990 he designed HTML and developed the language and the first web browser that was able to read HTML documents The first article related to HTML was published in 1991 it was called HTML tags and described eighteen elements and a first simple design of HTLM BLB The first HTML standardized specifica tion was completed in 1995 when HTML 2 0 was released BLC95 Since 1996 the HTML specifications have been maintained by the World Wide Web Consortium 2 but in 2000 HTML also became an international stan dard The current stable standardized version is HTML 4 01 that was published in 2001 In 2004 development began on HTML 5 that will be officially released in 2014 The design of HTML was inspired by SGML and still nowadays it has SGML based specification But on January 26 2000 the World Wide Web Consortium released XHTML 1 0 P 00 that has XML based specification XHTML is intended to be identical to HTML 4 01 except where limitations of XML over the more complex SGML
60. e efficacy evaluation by tasks 6 4 3 Analysis of the efficiency To evaluate the efficiency of LIME I asked the users to complete nineteen tasks In addition to the tasks described in the previous section I asked the users to perform ten additional tasks All tasks were treated as timed task For this reason I timed the users and reported the time that it took to complete the tasks Table 6 5 lists the tasks and their code 108 6 Evaluation of LIME s user experience Table 6 5 The tasks used to evaluate the LIME efficiency Task Task code Sign up to the system Tib Log in to the system T2b Change the editor s language T3b Import a Microsoft word s file from the user s desktop T4b Open one of the examples supplied by the editor T5b Save a file under a different name T6b Open the XML preview of a file T7b Save the XML version of a document on the user s desktop T8b Log out from the system T9b Markup the preface of a document T10b Markup the document s date T11b Markup the main body of the document T12b Markup an article its number and its heading T13b Markup a bold text T14b Create a table T15b Markup a quoted structure T16b Find the subsection 2 of the section 3 of a document T17b Unmark the preface of a document and all the contained elements T18b Save a document as a new expression in the same work T19b As said in the previous section the first
61. e examples of XML languages that are used for the markup of generic textual resources and for the markup of legal resources 3 2 Examples of dialects for textual resources A book a technical manual or a description of an ancient manuscript are all examples of textual resources For the markup of these documents XML languages are needed These must be specific but at same time as flexible as possible in order to hit all the peculiarities of the different areas in the same context For example a book about Oscar Wilde s aphorisms has a quite different structure from a novel regarding a pseudoscientific book that promises everlasting life Tho10 Similarly the structure of a technical manual about RecipeML also known as DESSERT Document Encoding and Structuring Specifica tion for Electronic Recipe Transfer is an XML language created in 2000 by the company FormatData and provides detailed markup for ingredients cooking time and so on 3 2 29 a programming language is completely different from the structure of a manual about an airplane made with LEGO M construction toys Differ ences can also be identified in ancient manuscripts and for this reason using a specific XML dialect that I will describe in the next section it is possible to markup the description of extremely different manuscripts i e the descrip tion of the Voynich manuscript and the description of a legal ancient manuscript PC13 3 2 1 Text Enc
62. e markup languages that are the currently most adopted languages for the markup of structured documents The rest of this research is focused on XML and on the textual resources XML due to its intrinsic nature and syntax is the best suitable choice for the markup of text based documents It allows to characterize their structure by using elements it enables the creation of metadata by using elements at tributes and last but not least thanks to XML it is possible to evidence the document s structure without changing or disassembling the textual content of the document In chapter 3 1 will explore various XML dialects in order to highlight differences and similarities between them 26 2 The markup of structured documents Chapter 3 Structural differences in XML dialects In this chapter I will deeply explore the use of XML to markup textual resources In the first section I will give an introduction about the differences among the XML dialects that are used for documents belonging to different contexts In the second section I will examine some examples of dialects that are used for the markup of generic textual resources and then in the third section I will show the most popular XML dialects that are used to create structured laws In section 3 4 I will the describe how it is possible to abstract the logical structure of the documents using patterns I will conclude this chapter examining the possibility to identify a ge
63. e six XML patterns supported by LIME LIME manages patternless elements by simply transforming them into structures that are compliant to HTML So the example in listing is translated into the markup in listing 5 2 In a nutshell patternless elements are wildcards that can be used every where and that do not follow any rule Listing 5 1 An example of patternless element lt xml version 1 0 encoding UTF 8 gt lt root gt lt patternlessExample gt Hello lt block gt world lt block gt lt hcontainer gt lt inline gt I am lt inline gt lt container gt lt block gt a patternless element lt block gt lt container gt lt hcontainer gt lt patternlessExample gt lt root gt Listing 5 2 A translated patternless element lt xml version 1 0 encoding UTF 8 gt lt root gt lt div gt 5 6 83 Hello lt div gt world lt div gt lt div gt lt div gt I am lt div gt lt div gt lt div gt a patternless element lt div gt lt div gt lt div gt lt div gt lt root gt 5 6 1 2 Guidelines for elements unique identifiers LIME follows guidelines for elements unique identifiers Unique identi fiers are built in specific ways that allow both to infer some semantic infor mation of the document just looking at them and to aid automatic processes that have to parse documents First of all all elements in a LIME documen
64. each document intended to be compliant to SGML must be composed of three parts the SGML declaration the pro logue containing a DOCTYPE definition that refers to a Document Type Definition and the content itself that is composed by a root element con taining all the other elements of the document The document type declaration defines the rules that the logical struc ture of the document must follow and other information that is useful for the documents processors such as the attribute values and their types It is used to check one of the two validity kinds of the SGML documents the so called type validity In order to be type valid a SGML document must be compliant to its document type declaration A fragment of a document type declaration is reported in listing 2 3 Although the document can be type valid it should also be tag valid in order to be full valid To accomplish this the text in the document must be completely tagged It is important to keep in mind that the type validity check is the first one to be performed because a failure means that the document is not syn tactically correct so it is not possible to process it in any way On the other hand even if the type validity check can be performed or not it is the most important one because if the documents does not follow the rules specified in its document type declaration it has no semantic meaning 20 2 The markup of structured documents Listing 2 4 repor
65. ection I will describe the requirements that these technologies must meet 5 2 Technologies for parametric markup edi tors A good parametric markup editor must meet many requirements First of all It must be an open source software because it should be used in contexts where open sourceness is required It must also be an on line in browser software so that users can avoid installing software on their computers but at the same time it must run correctly on all the most used browsers Moreover the editor must have a Model View Controller MVC KP 88 architecture because its interface must be easily modifiable in or der to meet the users requirements Furthermore the software must use an XML storage because it is intended to manage XML files and all the communication with the database must be done using a REST style infrastructure It must also exploit all the XML patterns to abstract XML languages and last but not least all of its parameters must be specified in configurations files that must be read and edited also by those who have no knowledge in programming languages In the next sections I will describe the technologies that can be used in order to create a markup editor that follows all the above requirements 5 2 1 Ajax javascrip and HTLM5 for in browser soft ware Ajax G 05 javascript GMEO7 and HTML HH10 are the leading technologies used for developing in browser software 5 2 67 Ajax Asynch
66. ed action so it will delete that resource REST is a natural style for the architecture of web applications 5 2 5 eXist Database eXist is a native XML database that is completely built on XML technology The database interacts with Ajax applications by supplying a RESTful interface A unique resource locator URI is assigned to each resource in the database which can be accessed using it This is important because by using a standard conceptual schema for the creation of the URI for instance FRBR it is possible to navigate the storage without using a query language The query language must be used to perform more complex actions for instance to modify specific fragments of documents or to retrieve specific subsets of documents Unlike relational database management systems eXist uses XQuery to access and manage the data that are stored in it 5 2 5 1 FRBR storage FRBR Functional Requirements for Bibliographic Records O N02 is a conceptual entity relationship model that allows users for example to retrieve and access resources in an online library by using a human friendly 70 5 LIME a parametric editor for structured documents syntax FRBR provides hierarchical links to navigate resources that are composed by a specific set of items The main entities in FRBR are the work the expression the manifestation and the item The work represents a distinct intellectual or artistic creation Pla98 For examp
67. eir content models XML patterns are the keystone for language independent markup editors In legal context markup editors are widely used and in the last two 1 Introduction decades many editors were developed As a matter of fact legal informatics experts were pioneers in the markup editors field They immediately under stood that XML is a powerful technology to ensure interchange reusability and portability of legal and legislative documents But they also understood that legal and law experts can not rely on pure XML editors because of the difficulty to write a proper XML legal document by using XML syntax Through legal markup editors legal drafters are not in charge of checking XML well form and XML validity They focus on the structure of the law and the editor drive them to create a correct XML too The most common legal markup editors are built as plug ins of well known word processors like Microsoft Word or Apache OpenOffice In this way users are facilitate in their work because they use an interface they are already familiar with By using macros legal drafters ascribe the logical structure to legal documents the legal markup editor prompts them the cor rect macros to be used and eventually it creates the final XML document by transforming the word processors format into XML The other side of the coin is that these software are hardly customizable because of three reasons Firstly they are built on the top
68. elda 21 02 1986 0carina jp8011 1998 game nod Link png 5 2 71 5 2 5 2 XQuery language and XSLT XQuery is a functional programming language used to query large sets of structured documents in XML format It relies on a set of XPath CD 99 expression used to address specific fragments of XML documents XQuery can also be used to create or modify XML documents indeed it supplies functions to dynamically or statically create nodes and attributes Even if like XSLT XQuery can be used to transform XML documents into other XML documents it is better to rely on XQuery only to query create and modify documents This because XSLT is stronger for simple tasks such as to transform all div elements to span elements Moreover the template architecture of XSLT is perfectly suitable for transformations based on patterns 5 2 6 XML patterns and XML guidelines As said in section XML patterns are powerful instruments that can be used to abstract XML dialects A parametric markup editor must rely on patterns in order to specify common procedures for elements belonging to the same pattern Also XML guidelines are useful in the design and the developing of a para metric markup editor XML guidelines specify requirements for the usage of XML that are not strictly required by the schema but that simplify the pro duction of homogeneous documents A parametric markup editor that aims to be used for the markup of many XML languages must specify and
69. eric XML editors are usually used by XML experts because they are comfortable with the XML syntax However the use of these editors is dis couraged when a lot of documents must be created The following section describes the Oxygen XML editor that is one of the most used generic XML editors 4 2 1 oXygen XML editor The Oxygen XML editor Sof03 is a multi platform software written in Java It runs on Windows Mac OS X and Linux as a standalone software or it can be installed as an Eclipse IDE s plugin P It is a proprietary software and has been developed since 2002 by SyncRO Soft Oxygen is currently released in its 15 1 version Oxygen offers a lot of features for editing XML documents As a generic XML editor it supplies tools to check the well formness and to validate them against a schema Schema compliance can be checked both when the docu ment is completely marked up or while it is being edited It is also possible to create plug in software in order to extend the Oxygen s native validation system Oxygen supplies three different views that users can exploit to edit XML documents The first one is the classical text view It simply shows the XML document as text and users can directly manipulate the code finding solace in tags auto completion or live validation The grid view displays the doc ument like a spreadsheet The left column shows the document s elements while the right column displays a contextual list of the children
70. f the standard was released in April 2007 and the final version was released later in January 2010 Originally the CEN Metalex standard was not intended for the actual markup of legal documents but it aimed to be an interchange format for le gal and legislative resources This is useful when public administrations need to manage sets of documents coming from different countries For instance the European Parliament needs to receive documents from all the European countries that format them with their favorite standard and needs to trans late them into another format i e Akoma Ntoso PC09 To accomplish this CEN Metalex strongly relies on the concepts of con tent models instead of elements With this approach the elements names are not semantically charged as they have to do according to the descriptive markup philosophy because they do not exactly specify the elements role in the logical structure of the document but they simply are labels that identify the content model they belong to Introducing content models like hierarchical containers containers blocks and inlines and elements with names that strongly reflect their content model like hcontainer container block and inline CEN Metalex aims to completely separate the descriptive role of the element from its role in the log ical structure of the document It is possible to see an example of CEN Met alex in listing 3 4 The CEN Metalex approach could seem per
71. fect for legal drafting but The CEN Workshop on Open XML interchange format for legal documents WS METALEX officially started on July 7th 2006 and ended in April 2007 The second phase of the meeting started on June the 4th 2008 and ended in January 2010 with the release of the final version of the Metalex standard 3 3 39 this has a long tradition and countries often have their own standard that has to be strictly followed On the other hand there are a lot of exceptions in concrete examples of laws and the possibility to rely on very generic elements would be a real godsend Akoma Ntoso tries to fill the gap between complete elements abstraction and strict descriptive markup Listing 3 4 A fragment extracted from an Italian law marked up using CEN Metalex lt xml version 1 0 encoding utf 8 lt root name NIR id metalex_EA tipo originale xmlns metalex xmlns xs http www w3 0rg 2001 XMLSchema instance xs schemaLocation metalex DTD e_fork_four xsd gt container name DecretoLegislativo id 1992 01 25 dlgs 84 xml lang it gt container name meta id metalex EFAA gt lt mcontainer name descrittori id metalex ECFAA gt mcontainer name pubblicazione id metalex EDCFAA gt lt meta id metalex ACDCFAA name pubblicazione tipo gt lt meta gt meta id metalex ABDCFAA name pubblicazione num gt lt meta gt
72. ficant For example a task can be rated as dreadful by the usability expert because it takes five minutes rather than the expected one minute But then the users can rate the same task as amazing because they had a lot of fun while performing it and as said before in some situations this can be the only thing that matters The testers where asked to answer two sets of questions The first ques tionnaire was completed before they performed the tasks listed in section and section The second one was compiled after they have com pleted the tasks The following are two examples of questions asked before and after the task completion respectively 94 6 Evaluation of LIME s user experience An example of a question submitted before the users completed the task la How difficult do you expect it will be to sign up to the system Very difficult O O O O O O O Very easy An example of a question submitted after the users completed the task 1c How difficult was to complete this task Very difficult O O O O O O O Very easy After each task was completed users were asked to indicate the degree of difficulty they found in the task s completion using the same scale they used in foreseeing their effort for the same task The complete questionnaire is reported in section 6 3 3 Thanks to this kind of tests it is possible to compare the expectation rating and the experience rating TA08 this can be use
73. follow some XML guidelines these are useful for example to create generic parser and generic queries for documents retrieval 5 2 7 JSON JSON JavaScript Object Notation is a text based and human readable open standard used for data interchange It is specifically designed to represent data structures using associative arrays 72 5 LIME a parametric editor for structured documents Even if JSON is language independent and there are many parser avail able for many programming languages it was derived from Javascript and for this reason it is the leading solution used for javasript applications con figurations files JSON is also often used in Ajax applications to pass data from the server to the client and viceversa This is useful for parametric editors because JSON is less verbose than XML and this results in smaller files An in browser parametric markup editor must use XML to describe struc tured documents and JSON to describe all the configuration files both for the client and the server side of the software and all the applications data that must be interchanged between the two sides 5 3 LIME a Language Independent Markup Editor LIME is the parametric web based language independent markup editor that I designed and partially developed to prove my thesis It drives users through the markup of non structured documents into well formed option ally valid structured XML documents compliant to the XML lan
74. guage cho sen by the user The LIME editor is an open source software and relies on many open source technologies It is currently under development by CIRSFID and the University of Bologna Works on LIME started when some parliaments asked professor Monica Palmirani and professor Fabio Vitali to create a web markup editor to markup their legal and legislative documents At the same time some scholars of his tory requested a markup editor that would be able to markup descriptions of ancient manuscripts CIRSFID is an inter departmental research center of the University of Bologna Its main legal researches are focused on legal informatics law and philosophy and sociology of law 5 4 73 Even if the two contexts are completely different by revolving on our experience on XML languages we started to imagine a markup editor that would be suitable for all XML dialects without the need to modify the code We started to focus on the legal context and by using the technologies reviewed in previous sections we designed and developed the first parametric and language independent markup editor LIME Language Independent Markup Editor gt F hutp time cirsfid unito it demo QF Goo ee M HB Dottorato di ricercay Apple Yahoo Google Maps YouTube Wikipedia Notizie amp 83 7 pl conosciutiv Welcome Demo User gt E E Document editor Document markup LPs Ge ee E d E Upload
75. he Common Industry Format For Usability Reports specifies that the core measure of efficiency is the ratio of the task completion rate to the mean time per task TA08 Table 6 8 and figure show the efficiency task by task Table 6 7 the average time in seconds for completion of tasks NC indicates the tasks that were not completed pl p2 p3 p4 p5 p6 p7 p8 p9 avg Tib NC 100 87 NC 118 163 65 89 151 110 T2b 6 5 12 NC 27 17 11 5 11 12 T3b 17 10 8 5 9 15 4 13 4 10 T4b 20 14 23 11 NC 12 15 10 16 15 T5b NC NC 24 NC 48 NC 11 NC NC 28 T6b 13 15 NC 14 17 11 15 14 NC 14 T7b 7 8 23 10 15 11 6 16 21 13 T8b 12 9 38 6 7 17 8 NC 12 13 T9b 6 17 23 22 4 18 15 21 11 15 T10b 13 10 14 15 NC 15 12 37 8 15 T11b 6 5 33 NC 5 43 13 4 7 14 T12b NC 11 13 13 13 20 16 7 23 14 T13b 3 NC 41 19 26 NC 17 20 28 22 T14b 10 12 10 5 NC 6 4 NC 50 14 T15b NC 13 10 16 12 13 8 16 16 13 T16b 24 12 48 10 18 NC 78 35 NC 32 T17b 7 0 11 31 20 11 11 20 17 14 T18b NC 7 30 23 6 10 10 8 19 14 T19b 23 17 25 18 15 26 NC 31 NC 22 6 4 111
76. he second generation web Scientific American 280 5 89 93 1999 C Biagioli E Francesconi P Spinosa and M Taddei The nir project Standards and tools for legislative drafting and legal document web publication In Proceedings of ICAIL workshop on e government modelling norms and concepts as key issues pages 69 78 2003 Alexander Boer Erik Hupkes Fabio Vitali Monica Palmi rani and Balazs Ratai Metalex cen workshop proposal Technical report CEN Workshop on an Open XML Inter change Format for Legal and Legislative Resources Met alex 2008 Tim Berners Lee Information management A proposal 1989 Tim Berners Lee and XHTML Basic First specifications Tim Berners Lee and Dan Connolly Hypertext markup language 2 0 Technical report RFC 1866 November 1995 123 124 BIBLIOGRAPHY BMC 04 BPVC11 Bry88 BSJ86 BvE11 CD 99 CM01 Cov00 Paul Biron Ashok Malhotra World Wide Web Consor tium et al Xml schema part 2 Datatypes World Wide Web Consortium Recommendation REC xmlschema 2 20041028 2004 Gioele Barabucci Monica Palmirani Fabio Vitali and Luca Cervone Long term preservation of legal re sources In KimNormann Andersen Enrico Francesconi Ake Gronlund and TomM Engers editors Electronic Gov ernment and the Information Systems Perspective volume 6866 of Lecture Notes in Computer Science pages 78 93 Springer Berlin Heidelberg 2011 Martin Bryan SG
77. icatorum 14th cent hanauilla is written at the foot of the page 15th cent Bought from the rev W D Macray on March 17 1863 for 1 10s Now MS Add A 61 Figure 3 1 A sample of a manuscript s description Listing 3 1 A TEI document describing the source in figure lt msDesc gt lt msIdentifier gt 5The quantitative codicology is the ability to make complex queries on large sets of manuscripts description such as Find all manuscripts that have the same distribution of jer letters and the same watermarks Find all manuscripts that share at least three descriptive features with one another and so on Courtesy of the The is an independent federal agency and is one of the largest funders of humanity programs in the United States 8The Modern Language Association is an association founded in 1983 by teachers and scholars that promotes the study and the teaching of languages 3 2 lt settlement gt Oxford lt settlement gt lt repository gt Bodleian Library lt repository gt lt idno gt MS Add A 61 lt idno gt lt altIdentifier type SC gt lt idno gt 28843 lt idno gt lt altIdentifier gt lt msIdentifier gt lt msContents gt lt p gt lt quote gt Hic incipit Bruitus Anglie lt quote gt the lt title gt De origine et gestis Regum Angliae lt title gt of Geoffrey of Monmouth Galfridus Monumetensis beg quote Cum mecum multa amp amp
78. icked the editor must markup the text selected by users with an element having the same name of the button The other described button is docTitle that when pressed must markup the selected 5 7 87 text with the docTitle element and must display a widget asking for a short title Then the text that users input in the widget will be inserted in the docTitle element s attribute called shortTitle Listing 5 4 Am example of a LIME configuration file elements acti 4 Children preface preamble conclusions docTitle askFor docTitle label short title type text insert attribute name shortTitle 5 7 Evaluating markup editors usability In this chapter I described the technologies suitable for the creation of a parametric markup editor and I explained how we used these technologies to design and develop LIME that is the first in browser parametric and language independent markup editor LIME is not simply a markup editor It is also supposed to drive the documents drafters to obtain a correct markup even if they do not know the XML language they are using The second challenge of my thesis is to 88 5 LIME a parametric editor for structured documents demonstrate that LIME is absolutely suitable for this task because it relies on a usable user interface In the next chapter I will describe how I created a usability test for markup e
79. il software non stato originariamente pensato Partendo dall analisi di questi software possibile progettare e im plementare un editor di markup indipendente dal linguaggio XML utilizzato LIME un editor di markup open source parametrico e indipendente da qualsiasi linguaggio XML La sua architettura basata sui pattern XML su alcune linee guida per la creazione di documenti XML e soprattutto su parametri inseriti all interno di file di configurazione JSON L idea alla base di LIME che possibile astrarre un linguaggio XML assegnando ad og nuno dei suoi content model uno dei pattern utilizzati dal linguaggio Akoma Ntoso Questa astrazione pu essere descritta con parametri array associa tivi all interno di file di configurazione JSON In questo modo in LIME possibile abilitare altri linguaggi XML semplicemente scrivendo un insieme di file JSON impacchettati in un modo comprensibile al software Questo significa anche che ogni eventuale estensione non necessita di competenze specifiche in nessun linguaggio di programmazione Altro scopo di questa dissertazione quello di identificare un test di us abilit adatto agli editor di markup e utilizzarlo per testare LIME Il test stato creato al fine di valutare l efficacia e l efficienza degli editor di markup e la soddisfazione degli utenti nell usare le loro funzionalit Il test stato sottomesso a dieci utenti per valutare LIME e ha evidenziato l eccel
80. iner gt div id b111 class bill container gt omissis lt div id bll1 bdy1 class body container gt lt div id blll bdyl chal class chapter hcontainer gt lt span id blll bdyl chal nml class num inline gt CHAPTER 1 lt span gt lt span id blll bdyl chal hdng1 class heading inline gt INTERPRETATION AND APPLICATION lt span gt lt div id bl11 bdyl chal sct1 class section inline gt omissis lt div gt lt div id blll bdyl chal sct2 class section inline gt omissis lt div gt lt div gt lt div id blll bdyl cha2 class chapter inline gt omissis lt div gt omissis lt div gt lt div gt lt div gt 5 6 2 JSON configuration files JSON configuration files are used to specify LIME parameters for each XML language These files are collected in packages called language plugins A language plugin is a collection of settings regarding both the language 5 6 85 used to markup documents and the behavior of the user interface when that language is used Each element specified in the plugin is connected with one or more buttons and one or more markup elements Many buttons can be inserted in a single configuration file 5 6 2 1 LIME language plugins A language plugin is not made of just one file of configuration Each plu gin provides a set of well structured directories and JSON files that
81. ion system that enables users to create and follow discussions about new technologies reading and posting messages on the system 1 At write moment the technical committee is discussing the version 5 1b2 of the stan dard 3 2 33 samples of block level elements These elements cause a vertical break of the text in the position they are fitted in The inline level elements are used to wrap a part of the text and to give to it some kind of different presentational rules or semantic meaning Unlike block level elements they do not split the text Listing 3 2 reports an example of a simple resource marked up using Doc Book The DocBook s distinction among different clusters of elements is really important because as I will discuss in section B 4 and later in chapter 5 the languages based on this kind of element s abstraction lay the foundations for a definition of a generic markup process Listing 3 2 An example of DocBook markup language lt xml version 1 0 encoding UTF 8 gt lt DOCTYPE article PUBLIC OASIS DTD Simplified DocBook XML V1 0 EN http www oasis open org docbook xml simple 1 0 sdocbook dtd article lt title gt DocBook Tutorial lt title gt lt articleinfo gt lt author gt lt firstname gt Adrian lt firstname gt lt surname gt Giurca lt surname gt lt author gt lt date gt April 5 2005 lt date gt lt articleinfo gt lt section gt lt title gt What
82. it was coordinated in conjunction with the Italian Ministry of Justice BFSTO3 The NIR standard was designed around the Italian legislative system and its schema is specified using DTD NIR supplies three different schemata that are used for specific purposes The flexible DTD nirloose does not specify any mandatory rule and is used for those documents that do not have to follow drafting rules The basic DTD nirlight is a subset of the complete schema and is used for purposes other than the legal drafting for example for teaching purposes The complete DTD nirstrict contains about one hundred and eighty elements specifies a lot of mandatory legal drafting rules and permits to completely markup the Italian laws As a second generation legal XML standard NIR does not try to abstract the logical structure of the laws in order to be reused I have to point out APA a per l informatica nella pubblica amministrazione Currently it is known as CNIPA Centro nazionale per l informatica nella pubblica amministrazione Na tional Center For The Information Technology in the Public Administration 36 3 Structural differences in XML dialects that this does not mean that NIR is not able to markup the structure of the document of course it can because it is a descriptive markup language But NIR identifies the logical structure of Italian laws and uses elements that are specific for the Italian legal system In the example reported in
83. ituita ed operante in conformita alle disposizioni del presente decreto che ha per oggetto esclusivo 1 investimento collettivo in valori mobiliari del patrimonio raccolto mediante l offerta al pubblico in via continuativa di proprie azioni e denominata societa di investimento a capitale variabile SICAV lt basic gt lt hcontainer gt omissis lt hcontainer gt omissis lt basic gt omissis lt container gt omissis lt root gt 3 3 41 3 3 3 Akoma Ntoso Akoma Ntoso 4 is an XML standard developed in 2005 under a project of the United Nations Department for Economics and Social Affairs whose aim is to aid the African legislatures to better accomplish their democratic functions using the information technologies VZ07 The standard went through various versions till today and nowadays it is in its third version and is an OASIS standard In order to match the nam ing convention of the OASIS XML standard the Akoma Ntoso namespace was renamed in LegalDocumentML but it is still referred to as Akoma Ntoso in literature and in technical documentations Akoma Ntoso has a lot of features that make it usable in a large variety of countries respecting their specific legal system it separates the authorita tive content from the non authoritative one it clearly separates the metadata from the data and above all it is constituted by simple and reusable pattern
84. kup languages can be roff nroff and troff all typesetting packages for the UNIX operating system that read a file written with a special syntax and output another file suitable for the presen tation For example the listing 2 1 shows how to set a line spacing and how 4See the page on wikipedia to read more about the history of the markup 5Read more about Microsoft Word on the word web page 16 2 The markup of structured documents to center the text using groff a Linux porting of nroff and troff Listing 2 1 An example of groff markup language ls 2 ce 1 Hello world I m a centered text in a document with a specific line spacing Another typesetting program introduced in 1978 is TeX that uses the meta font language for font description and the modern typefaces to allow the users to produce well formatted documents with a minimum effort Later in time in early 1980s LaTeX was introduced a document markup language for the TeX typesetting program LaTeX allows to create high quality documents with a reasonably simple syntax Lam86 For instance the code in listing 2 2 produces the document in figure Procedural markup languages was massively used in the 1980s and 1990s and LaTeX remains till today the most used instrument to edit scientific papers this dissertation is written in LaTeX too Even if both procedural markup languages and presentational markup languages are useful to decouple the logica
85. l structure from the document content this work focuses on the descriptive markup languages that I will extensively explain in the next section 8For more information read the groff s project page Courtesy of 2 3 17 XT RX IATEX is a document preparation system for the TEX typesetting pro gram It offers programmable desktop publishing features and extensive fa cilities for automating most aspects of typesetting and desktop publishing including numbering and cross referencing tables and figures page layout bibliographies and much more IXTpX was originally written in 1984 by Leslie Lamport and has become the dominant method for using TEX few people write in plain TEX anymore The current version is BTEX 22 Figure 2 1 The output generated by the lines of code in the listing 2 2 Listing 2 2 An example of LaTeX markup language documentclass 12pt article usepackage amsmath title LaTeX date begin document maketitle LaTeX is a document preparation system for the TeX typesetting program It offers programmable desktop publishing features and extensive facilities for automating most aspects of typesetting and desktop publishing including numbering and cross referencing tables and figures page layout bibliographies and much more LaTeX was originally written in 1984 by Leslie Lamport and has become the dominant method for using NTe
86. l web services O Reilly 2008 DB Russel Cocoa A word count and concordance gener ator for atlas Atlas Computer Laboratory Chilton 1967 Brian Shackel Usability context framework definition de sign and evaluation Human factors for informatics usabil ity pages 21 37 1991 C Michael Sperberg McQueen Lou Burnard et al Guide lines for electronic text encoding and interchange volume 1 Text Encoding Initiative Chicago and Oxford 1994 SyncRo Soft oxygen xml editor 2003 130 BIBLIOGRAPHY TA08 TDK 99 Tho04 Tho10 US91 Van04 VZ07 Wal99 Wil98 Tom Tullis and Bill Albert Measuring the user experi ence Collecting Analyzing and Presenting Usability Met rics 2008 Charles A Taylor Mary T Draney Joy P Ku David Parker Brooke N Steele Ken Wang and Christopher K Zarins Predictive medicine computational techniques in thera peutic decision making Computer Aided Surgery 4 5 231 247 1999 Henry S Thompson Xml schema part 1 Structures second edition 2004 Scarlett Thomas Our tragic universe Canongate Books 2010 CORPORATE Unicode Staff The Unicode Standard Worldwide Character Encoding Addison Wesley Longman Publishing Co Inc 1991 Edward Vanhoutte An introduction to the tei and the tei consortium Literary and linguistic computing 19 1 9 16 2004 Fabio Vitali and Flavio Zeni Towards a country independent data format the akoma
87. le in video game contexts the concept The Legend of Zelda is a work Ina FRBR URI this concept must be expressed as the following jp zelda 21 02 1986 The URI expresses that Zelda is a work published in Japan on February the 21st 1986 The expression is the specific intellectual or artistic form that a work takes each time it is realized Pla98 In Zelda example an expression TM can describe the Ocarina of the time episode of the game released in November 1998 in Japanese The URI of this expression is the following jp zelda 21 02 1986 0carina jp811 1998 The manifestation is the physical embodiment of an expression of a work As an entity manifestation represents all the physical objects that bear the same characteristics in respect to both intellectual content and physical form Pla98 For the Zelda work a manifestation can be the Ocarina of the time version stored on Nintendo Optical Disk released for Nintendo Game Cube T9 The URI of this manifestation can be the following jp zelda 21 02 1986 0carina jp8011 1998 game nod The item is a single exemplar of a manifestation The entity defined as item is a concrete entity Pla98 For example an item of Zelda Ocarina of the Time in its Game Cube TM version can be the file character png describing the physical file that contains the art of the main character of the game Items can be expressed in URI format as the following jp z
88. le to identify these three conceptual frameworks In a structured document the focus is on the logical structure of the document with no concerns about how it must be printed For this reason a structured document contains a lot of semantic information about its logical structure which can be easily interpreted by computers in order to create different kinds of presentations suitable for a variety of devices computers screens mobile phones paper sheets and so on or in order to perform com plex computations on large sets of documents Such structure can be ascribed to the document using appropriate and standardized markup languages that are used to add semantic markup to each section of the document that is relevant to its logical organization In the next section I will explain how these languages can be used to create the logical structure of the document and the differences among the most known markup languages used to create structured documents 2 Although some studies highlight four conceptual structures for example the one of Mar92 this research is focused on only three of them Tt is important not to confuse the logical structure that refers to the organization as it is seen by the author and the physical structure that is the organization as it is seen by the publisher 2 2 15 2 2 Markup languages for structured docu ments A markup language is a language used to annotate some parts of the text in o
89. lente us abilit della maggior parte delle sue funzionalit Numerose istituzioni parlamentare politiche e apolitiche attendono il ri lascio della prima versione beta di LIME per il markup dei loro documenti legali e legislativi I successivi capitoli di questa dissertazione verranno redatti in lingua in glese per facilitare la comprensione a ricercatori e sviluppatori non italiani 0 Introduzione Chapter 1 Introduction The purpose of this dissertation is to pinpoint the proper technologies for the development of parametric editors for structured documents and to characterize LIME a parametric and language independent markup editor The recent evolution of XML related technologies has led to a massive use of structured documents both for typographical and for data interchange purposes and in the last decade several XML languages have been developed in order to meet the requirements of an incredibly diverse set of contexts Due to this many people with many different skills started to handle XML documents in their everyday work But specific XML dialects can be difficult to understand and to be used properly and for this reason XML editors that allow users to markup documents in a proper way are needed In specific contexts especially the legal one XML drafters find help in markup editors Markup editors are WYSIWYG software that assist users in the creation correct XML documents by driving drafters through the correct
90. low associated to each XML language This means that it 10 1 Introduction is also possible to modify a part of the LIME business logic without any effort in terms of coding The other challenge for LIME is to ensure users a great user experience Since users are driven by the editor to create a proper XML file in some situations they may feel frustrated because this can be interpreted as a lack of freedom during the markup workflow To evaluate the LIME interface s usability a usability test appropriate for markup editors was created The usability test was submitted to ten users during a hackaton held at CIRSFID Centro Interdipartimentale di Ricerca in Storia del Diritto Filosofia e Sociologia del Diritto e Informatica Giuridica of the University of Bologna The hackaton was also intended to find the bugs of the first alpha version of LIME The test was performed on LIME with its Akoma Ntoso configuration After a brief introduction to LIME overall features and to the structure of legislative documents two questionnaires were submitted to the users and they were asked to perform nineteen tasks in order to evaluate the LIME efficacy the LIME efficiency and the users satisfaction while they use LIME The test results demonstrate that users are able to easily use the LIME features related to the navigation system such as import export files open files and save them It also demonstrated that even if users have no knowl edge
91. n markup editors In this chapter I described existing WYSIWYG editors and markup edi tors I analyzed some of the most used markup editors for legal and legislative documents The legal context is the perfect use case to understand the problems that the markup editors have to face if they aim to be completely customizable and portable Even if they allow users to markup documents using just one XML language they need to be extremely modular in order to fit all the specifics needs of the legal traditions Current markup editors have some strengths and some weaknesses Some of them are not open source software and can not be used in some parlia ments in which open sourceness is required Others are hardly customizable 64 4 Software for XML markup because they are built on desktop software and are strictly related to a spe cific XML language Most recent legal markup editors are based on Akoma Ntoso and this give to them a certain portability and customizability but their can not be used with other markup languages maybe not related to the legal context My challenge is to create a language independent editor that can be used both in the legal context and in other contexts where XML can bring inno vation In the next chapter I will explain how starting by the experience gained in the legal context we created a parametric editor that is suitable for the creation of structured documents marked up with any XML dialect Chapte
92. n order to be independent from a specific markup language 4 1 WYSIWYG editors WYSIWYG What You See Is What You Get describes document edi tors in which the document displayed during the editing mode is quite similar to the one that will be eventually printed on paper The first software that incorporated WYSIWYG features was the BRAVO editor developed in 1974 at Xerox PARC It was intended to be used with a specific monitor in which users could see a full page of text Doing so users can adjust the document s layout while they are editing the page and then they can print it out obtaining a similar result that they saw Xerox PARC is a research and development company located in Palo Alto California that is the mind behind a lot of innovative products such as graphical user interfaces laser printing and object oriented programming 53 54 4 Software for XML markup on the screen The phrase WYSIWYG was later coined in 1982 by Larry Sinclair an engineer at Triple I P to express the idea that what users see on the screen is what they finally get on paper In the XML context the WYSIWYG paradigm is used in two different ways The first type includes the ones that allow users to create XML doc uments by directly editing XML elements and attributes Other editors markup editors do not allow users to directly manipulate XML code but they allow users both to describe the logical structure of the document ac cording t
93. neric markup process that can be used for the markup of all the textual based resources 3 1 Different dialects for different contexts Currently there are a lot of XML dialects used in a large variety of contexts Many of them are used for the markup of text based resources There At the time of writing it is possible to count two hundred twenty two languages in the wikipidia s page that lists the most known XML dialects 27 28 3 Structural differences in XML dialects are XML used for the markup of medical resources that enable for example predictive medicine TDK 99 there are XML dialects used for markup recipes that are useful for the creation of internationalized cookbooks Raa03 and there are even languages used for the markup of theological texts that should be suitable for online sharing of ser mons Some of these languages are really specific for the context they belong to and for this reason are composed by elements and attributes that de scribe exactly the items of their framework Other languages need to be more generic because even if they are intended for the markup of textual resources residing in a clear cut area they have to face up a big set of differ ences due for instance to the diversity of the traditions of the documents creators or to the laws that regulate that kind of documents in the country of the documents writers In the next sections I will show som
94. nes in order to create homogeneous XML documents Thirdly LIME parameters are specified in JSON configuration files To create the configuration for a new XML language a language plugin must be writ ten A language plugin is a set of configuration files packaged in a specific way that LIME can understand This means that LIME configuration files are independent from the programming language used to develop LIME and it is possible to create a LIME language plugin without any skill in code developing Lastly LIME stores files in a native XML database and as signs them URI according to the Functional Requirements for Bibliographic Records FRBR In this way produced documents are immediately pub lished on the web and can be accessed by humans and machines by using the Representational State Transfer REST As a markup editor LIME drives users to create XML files even if they do not have any knowledge of XML and moreover even if they do not have any knowledge of the specif XML language they are using to markup docu ments Users select the parts of the text they want to markup and the editor displays the elements that they can use to markup these parts For example when users are marking up an act in Akoma Ntoso language and they select a text inside an article the editor allows to markup it as a section and does not allow to create another article inside it By creating LIME configuration files it is also possible to define the markup workf
95. next sections I will describe three legal markup editors created for the Ital ian parliament the African parliaments and the United States parliaments respectively 4 4 59 4 4 1 Norma editor The Norma editor was developed starting by 2002 as a compo nent of the Norma System project PB02 It is built on Microsoft Word and allows the creation of XML documents compliant to the NIR XML stan dard The software was intended to be used by the Italian parliament s legal drafters and by legal drafters from many Italian public offices such as the ones from the Italian Supreme Court of Cassation The Norma editor can acquire all the unstructured formats that Microsoft Word supports as well as documents already marked up in Norma format Users can then markup the document and insert all the elements of the Ital ian laws structure For example they can insert the basic information of the laws such as the opening formula the closing formula and the main body of the law and then they can refine the markup by adding more articles and sections to the laws body The editor supports either a manual markup of the law and a semi automatic markup Through the manual markup users select the fragment of the text they need to markup and then they use a toolbar to assign to it the correct label and semantic meaning They repeat these actions until the document is completely marked up and valid against the NIR schema The semi automatic ma
96. nguage as I will explain in the next sections this per mitted the creation of the Hyper Text Markup Language and later evolved into the eXtensible Markup Language 2 3 1 The Standard Generalized Markup Language The Standard Generalized Markup Language henceforth referred to as SGML is an ISO standard technology P whose purpose is to define gen eralized markup languages for documents It is important to underline that 8A more detailed reading about the differences between the procedural and the descrip tive markup languages can be found on the TEI s manual page ISO 8879 1986 Information processing Text and office systems Standard Gener alized Markup Language SGML 2 3 19 the ISO standard also introduced for the first time a definition of generalized markup According to the standard a generalized markup should be declar ative because it must describe the document s structure rather than specify the processing to be performed on it and it should be rigorous in order to allow computer programs to process the documents in the right way SGML descends from the Generalized Markup Language that was devel oped in the 1960s by Charles Goldfarb Edward Mosher and Raymond Lorie Gol91 It was originally designed to mark up large set of documents in a machine readable format These documents should be preserved in time and should be easily shared between public offices and industries The standard defines that
97. nine tasks were considered com pleted if users respected the time limit The other tasks did not have a time limit They were considered completed if users properly finished them inde pendently from the time it took Tasks were considered not finished if users abandoned them before their completion Table 6 6 lists the completed and not completed tasks 6 4 109 Table 6 6 The summary of the tasks performed by users 1 indicates completed tasks 0 indicates not completed tasks pl p2 p3 p4 p5 p6 p7 p8 p9 avg Tib 0 1 1 0 1 1 1 1 1 78 T2b 1 1 1 0 1 1 1 1 1 89 T3b 1 1 1 1 1 1 1 1 1 100 T4b 1 1 1 1 0 1 1 1 1 89 T5b 0 0 1 0 1 0 1 0 0 33 T6b 1 1 0 1 1 1 1 1 0 78 T7b 1 1 1 1 1 1 1 1 1 100 T8b 1 1 1 1 1 1 1 0 1 89 T9b 1 1 1 1 1 1 1 1 1 100 T10b 1 1 1 1 0 1 1 1 1 89 T11b 1 1 1 0 1 1 1 1 1 89 T12b 0 1 1 1 1 1 1 1 1 89 T13b 1 0 1 1 1 0 1 1 1 78 T14b 1 1 1 1 0 1 1 0 1 78 T15b 0 1 1 1 1 1 1 1 1 89 T16b 1 1 1 1 1 0 1 1 0 78 T17b 1 0 1 1 1 1 1 1 1 89 T18b 0 1 1 1 1 1 1 1 1 89 T19b 1 1 1 1 1 1 0 1 0 78 As table shows I took the time each user needed to complete each task and the average time needed by all users 110 6 Evaluation of LIME s user experience The average completion time and the task completion rate allow us to measure the efficiency of the editor T
98. nowledge needed in order to accomplish the simple test s tasks This session lasted thirty minutes and was performed in a lab of the University of Bologna After the presentation during the foresee session users were asked to compile a questionnaire in which they foresaw the effort that they presumed to make to accomplish all the test s tasks I gave them a copy of the question naire and I asked them to sign it with the ID they had previously received This session had no time limit In the last session I asked the users to individually perform some tasks while I was observing and timing them For this session I produced my per sonal sheet in which I reported the information related to users while they were performing the task I divided the tasks into two groups The first group contained the tasks developed to examine the navigation system Tasks in this group was labeled with a NS ID and was treated as binary data TA08 for this reason I assigned them a time limit If users finished the task in time I rated it as accomplished otherwise I rated it as not accomplished The second group contained the tasks used to examine LIME s efficiency 96 6 Evaluation of LIME s user experience Tasks in this group were labeled with a MU ID and were treated as time tasks TA08 For items in this group I simply reported the time users spent to accomplish the task After the completion of each task I asked the users to report the degre
99. o the XML dialect which the document must be compliant to and to format the document in the exact way they want to present it In the next section I will describe an example of generic XML editors belonging to the first category and in section 4 4 I will describe the most common editors that are used in the legal context and that belong to the second category 4 2 Generic XML editors Generic XML editors are independent from the XML dialects used for document markup They let users create any kind of XML document and they can to check the well formedness and the validity of the document if a document type declaration is set While using this kind of editors users are focused on the quality of the created XML and not on the final presentation of the document For this reason here the WYSIWYG ability indicates that users can directly ma nipulate the XML code and that the final result will be exactly the XML document that they created These editors assume that users have a perfect knowledge of XML and of the XML dialect they are using but they can supply some kind of tools to help users to create a correct markup For example parsing the schema of the XML language editors are able to give hints about the elements that Triple i is a computer company located in Los Angeles California 4 2 55 can be inserted in a specific position of the document or they can give the list of the attributes allowed in each element Gen
100. ocuments The second section aims to briefly navigate through the history of the markup exploring the different styles of markup The third section is about the de scriptive markup languages and explores examples of them Finally in the fourth section I will introduce some notions about the markup of text based resources and the currently most used languages to markup them 2 1 Structured documents A structured document is an electronic document which embeds other information in addition to its content This supplementary information is used to give semantic meaning to the whole document or to specific parts of it and is fitted in the document using some kind of embedded coding such as markup l Extracted from the wikipedia s definition of structured document 13 14 2 The markup of structured documents Typically a document contains at least three different conceptual struc tures P The first one is the logical structure that is a description of the objects that compose the document and the relations among them Then there is the layout structure that specifies the document s layout i e the pages format the elements that must be underlined or the text s parts that must be centered Last but not least is the physical structure that describes for example the fact that a book must be divided into pages and that a page must be divided into sections P In any type of document including paper ones it is possib
101. oding Initiative The Text Encoding Initiative IV95 hereafter TEI is a community cre ated in the 1980s that aims to support the digitization of texts chiefly in the humanities social sciences and linguistics The result of the collaboration is an XML standard and a set of guidelines for the creation of digital text The first version of the TEI guidelines was released in 1990 but the third version of the guidelines P3 released in 1994 was the first to be widely used XML was introduced in 2001 with the release of the fourth version of the standard P5 The latest version P5 was published in 2001 The task of flexibility is achieved by TEI standard because it does not specify a fixed set of rules but it is intended to be customized both for the users that need to select subsets of TEI s elements and for users that have to add elements for the particular features that they need in order to markup their texts TEI permits local variation of usage and it is not different human languages it supports idiomatic usage dialects and local usage TEI is extensively used to markup detailed descriptive information about ancient manuscripts because it supplies a specific module that is generic enough to permit to describe any kind of handwritten resource A The stan dard also permits to create electronic documents for different purposes for 3See the PHP s documentation page to have an idea of how programming language s documentation are
102. of other complex software and this means that any attempt to change the business logic or the interface of the software would be very expensive Secondly they are built to comply very specific markup workflows that derive from the legal tradition of the country where they are used Last but not least they rely on a specific XML language and it is often difficult to adapt them to other XML standards Starting from the analysis of legal markup editors it is possible to design and implement a markup editor whose architecture revolves on parameters in order to ensure the customizability and the independence from any XML markup language LIME Language Independent Markup Editor is a parametric and lan guage independent markup editor aimed to be used with a wide variety of different XML languages It relies on a set of parameters that are used to 1 0 describe the XML languages that must be used and the markup workflow related to the languages It solves many of the problems of current legal markup editor and aims to be used in any context in which XML can bring innovation LIME is an open source stand alone web application and its architecture is based on a set of technologies that ensure its independence from any XML language Firstly and most importantly LIME relies on XML patterns In order to be used in LIME XML languages must be analyzed and a pattern must be assigned to each content model Secondly LIME follows specific XML guideli
103. ologna Bologna quanti bei giorni che mi hai regalato A cominciare da quella prima casa al Lunetta Gamberini con le sue partite di calcio in corridoio il mio compagno di stanza e la luce rossa e la ciabatta il biondo e la 500 e sempre la luce rossa mentre dormiva E quelle amiche delle Marche e una in particolare con le birre il venerd e far mattina al sabato E dopo un po arrivata anche la laurea triennale E con essa il lavoro all universit e quei docenti che mi hanno insegnato a fare lo scienziato anche quando scienziato certificato non ero E sono arrivati quei colleghi e amici che mi hanno ricordato che in fondo se sei nerd non conta se fai il dottorato in Olanda o la popstar Nerd si nasce Ed io lo nacqui Come anche mio fratello che mi ha regalato una giappognata e un nippotino ma sempre nerd rimane E intanto arrivata un amica toscana la Toscana l olio d oliva e i la voretti in campagna Ed arrivata Arianna E anche grazie a lei se questa dissertazione scritta in un inglese spero perfetto Mi ha insegnato tanto e non solo della lingua inglese Per ora la fine questa Ma si sa un punto fa morale e sono abbas tanza sicuro che la storia continuer Ma intanto cos altro potrei dirvi per ringraziarvi tutti di cuore I m not crazy my mother had me tested Sheldon Cooper
104. omething hap pens to the document the editor fires events that are handled by the other components of the application By doing so even if the editor is currently based on TinyMCE and is the most important component of the application it can be easily substituted by other third parts WYSIWYG editors The explorer is the component of LIME that has in charge to display the hierarchical structure of marked up documents When a part of the docu ment is marked using the marking menu the editor also asks the explorer to update itself The marking menu is the component that creates the buttons used to markup documents When users click buttons in marking menu it asks the editor to modify the displayed document and the editor replies updating the marking menu The main toolbar aims to supply all the common operations of a WYSI WYG editor such as the operations to load and save files and the operations to change users preferences The JSON configuration files are the client side components of LIME that allow to specify its parameters I will describe them later in section 5 6 2 Figure shows how the client side components interact among them selves 80 5 LIME a parametric editor for structured documents JSON Main configuration files toolbar language plugin Instantjates Instantiates Instantiates ExtJs 4 1 App Marking menu Explorer Instantiates Ask to upd
105. one A multi layer digital library for mediaeval legal manuscripts In Digital Libraries and Archives pages 81 92 Springer 2013 Monica Palmirani Giuseppe Contissa and Rossella Ru bino Fill the gap in the legal knowledge modelling In Guido Governatori John Hall and Adrian Paschke editors Rule Interchange and Applications volume 5858 of Lecture Notes in Computer Science pages 305 314 Springer Berlin Heidelberg 2009 Christopher Piro and Eugene Letuchy Functional program ming at facebook In Commercial Users of Functional Pro gramming Conference 2009 7 0 129 Pla98 Prz09 Raa03 RCDS Ros04 RR08 Polia Rus67 Sha91 SMB 94 Sof03 Marie France Plassard Functional requirements for bib liographic records Final report IFLA Study Group on the Functional Requirements for Bibliographic Records KG Saur Verlag GmbH amp Co KG Mtinchen 1998 Adam Przepi rkowski Tei p5 as an xml standard for tree bank encoding In Proceedings of the Eighth International Workshop on Treebanks and Linguistic Theories TLT8 pages 149 160 2009 Sebastian Raaphorst Cookbook A usability study 2003 J Robie D Chamberlin M Dyck and J Snel son Xquery 3 0 An xml query language 2011 Availab le http www w8 org TR 2011 WD xquery 30 20111213 visited on 01 11 2012 Lawrence Rosen Open source licensing Prentice Hall PTR 2004 Leonard Richardson and Sam Ruby RESTfu
106. ord processor software and for this reason it is focused on the legal drafters needs On the other hand LegisPro is not an open source software and as said in section this is a big problem if the editor aims to be used in Eu ropean parliaments Another criticism about LegisPro is that even if it is based on Akoma Ntoso it does not exploit the patterns specified by Akoma Notoso and as I will explain in the next section this could be the keystone to create a language independent editor 1 is a Californian vendor of legislative products and services to governmental bodies at all levels 4 5 63 LegisPro is currently used for the bill drafting and publishing system in the State of California and many American and world s parliaments are considering using the software for their drafting purposes LegisPro Web 4 http legisproweb com editor Index html cfg standard BE c Qr Good ee M HH Doworato di ricercay Apple Yahoo Google Maps YouTube Wikipedia Notizie 703 _1 pi conosciuti SE gt ai Pat cacao AZ LegisPro 112th CongressPublic Law61 NOV 29 2011 125 STAT 753 Public Law112 61 112th Congress AnActTo facilitate the hosting in the United States of the 34th Anerica s Cup by authorizing certain eligible vessels to participate in activities related to the competition and for other purposes notel NOTE Nov 29 2011 H R 3321 Be it enacted by the senate and House of Rep
107. p toolbar of LIME that is showed in figure 74 5 LIME a parametric editor for structured documents Lime File Edit Preferences Window i New document E 1 3 pr yl Open document 3 5 Save ig Save as a Import document Export as b Figure 5 2 A screenshoot of the LIME top bar In order to enable documents markup and to write new documents LIME supplies an editor displayed in the center of the application s main window figure 5 3 This also supplies buttons for cross language markup features like buttons for bold text inline text and so on The document that users see in this editor is formatted in HTML5 because it is currently the best suitable format for the visualization of pages on browsers When users save documents they are immediately translated to the markup language users are using In this part of the application is it also possible if enabled in configuration files to see the preview of the final XML result and the PDF preview of the document As said before LIME does not simply allow users to markup documents but also drives them to create a correct markup even if they do not have a 5 4 75 deep knowledge of the XML language they are using Document editor Akoma ntoso preview PDFpreview B Z S iz i Es By o BW Wupoa HM DOCUMENT NUMBER 281 1 DOCUMENT NUM
108. pe hierarchicalStructure preface Type prefacaapra preamble Type BIRNRSETCA body Type bodyType conclusions Type basicopt 5 ALL i e Figure 3 2 The content model of the act element in Akoma Ntoso 3 4 2 Hierarchical containers A hierarchical container is a set of titled and numbered nested sections Each section can contain other sections or a container if it is the last in the hierarchy Like the containers no text is allowed directly in the hierarchy The diagrams in figure shows a part of the content model of article elements in Akoma Ntoso and the main hierarchical elements allowed inside them 3 4 47 J hierarchy Base Type basehierarchy I basehierarchy extension base num Type ining heading ENS Type inlinerea subheading Type inline attributes corereq intro Type blocksreq tal P Type srcType YP Yp A E hierElements fa ANhier Ll clause Type hierarchy E section Type hierarchy A paragraph Type hierarchy chapter Type hierarchy article Type hierarchy Type hierarchy a Figure 3 3 The content model of the article elements in Akoma Ntoso 3 4 3 Markers Markers are elements with empty content model that are meaningful for their name their attributes or their position in the text They are used to insert metadata
109. r 5 LIME a parametric editor for structured documents In this chapter I will describe LIME a parametric editor for structured document that is independent from the markup language In the first section I will describe what I mean for parametric The second section is aimed to de scribe the technologies that can be used for the development of a parametric editor Then in section I will deeply describe LIME and its architec ture The last section of this chapter makes a brief introduction about the methodology that should be used to evaluate the usability of markup editors 5 1 The importance of being parametric A parametric software is a software that relies on a set of parameters to ensure customization For example a software aimed to forecast the sales of a videogame should be built around specific parameters like the videogame type the area in which it must be sold the gamers attitudes of that area and so on In this way the software can be easily adapted to be used for any videogame In the same way a parametric markup editors must rely on a set of pa rameters that abstract the structure of XML dialects Doing so it is possible 65 66 5 LIME a parametric editor for structured documents to use the same editor with all XML languages without changing its code but simply modifying its parameters There are a lot of technologies suitable for the creation of a language inde pendent markup editor In the next s
110. rder to highlight its logical structure and to make this structure under standable both by software and if the markup language is human readable by human beings Gol81 The term markup is derived from the publishing practice of marking up a manuscript CRD87 according to which a reviewer adds some instruc tions usually with a blue pencil on the margins and on the text of a paper manuscript to indicate what typeface style and size should be applied to each part of the text In electronic markup the blue pencil was replaced by some kind of instructions that are embedded in the text in binary format or surround the text like labels There are three categories of electronic markup the presentational markup the procedural markup and the descriptive markup CRD87 The presentational markup is the markup used by traditional words pro cessing systems such as Microsoft Word s versions prior to 2002 P These kinds of software embed the markup in the text in binary code Such markup is intended to be hidden from the author and the editors With the procedural markup the markup is embedded in the text and provides information for programs that need to process the text The text is marked up in a visible way and directly manipulated by the author and usually during the compiling phase the software runs through the text from the beginning to the end following the instruction as encountered First ex amples in the history of procedural mar
111. related to the difficulty of marking up specific legal ele ments can probably be fixed by observing large sets of legal experts during the markup workflow Next usability tests can be divided in usability tests with legal drafting experts and usability tests without them In this way I will be able to understand what are the issues strictly related to the legal context The hackaton during which I proposed the usability test was also the first time in which LIME was massively used This highlighted many bugs indeed over thirty bugs were reported by users This was expected because the software is in its alpha version but of course I cataloged each bug and rated it in a severity scale and scheduled their fix Several parliaments and many political and apolitical institutions are cur rently waiting the first official LIME release to markup their legal and legisla tive documents Also some history scholars are waiting a first stable version of LIME to markup ancient manuscripts descriptions and transcriptions In the future I will create language plugins for many common XML languages and I hope that LIME will be used by business and common people to markup all kind of XML documents 122 7 Conclusions Bibliography AH08 BB 99 BFST03 BHV 08 BL89 BLB BLC95 Andy Austin and Christopher Harris Chapter 9 Case studies Library Technology Reports 44 4 31 36 2008 Jon Bosak Tim Bray et al Xml and t
112. require workarounds The content of a HTML document is usually formed by a head element and a body element The head contains information abut where to retrieve See the W3C s page for HTML 4 01 specification 12The World Wide Web Consortium W3C is an international community that develops open standards to ensure the long term growth of the Web 13ISO IEC 15445 2000 Information technology Document description and processing languages HyperText Markup Language HTML MATML 5 1 Nightly A vocabulary and associated APIs for HTML and XHTML Edi tor s Draft 23 September 2013 22 2 The markup of structured documents the files related to the HTML document like script files and style files and contains important metadata about the whole document such as its title and its keywords The body contains the actual content of the document As is it possible to see in listing 2 5 HTML 4 01 is not a pure descrip tive markup language due to the fact that it mixes descriptive markup and presentational markup Elements like A1 are used to specify that the inside text represents a heading in this case a first level heading and they do not denote a specific markup style even if most web browsers have a default style in order to format these kind of elements On the other side of the coin ele ments such as b and u are used to indicate that the devices should render the text in bold face or underlined respectively But the u
113. resentatives of the United States of America in Congress assembled SECTION 1 SHORT TITLE note2 America s Cup Act of 2011 This Act may be cited as the America s Cup Act of 2011 SEC 2 DEFINITIONS In this Act 1 34th america s cup The term 34th America s Cup A means the sailing competitions commencing in 2011 to be held in the United States in response to the challenge to the defending tean from the United States in accordance with the terms of the America s Cup governing Deed of Gift dated October 24 1887 and B if a United States yacht club successfully defends the America s Cup includes additional sailing competitions conducted by America s Cup Race Management during the l year period beginning on the last date of such defense 2 America s cup race management The term America s Cup Race Management means the entity established to provide for independent professional and neutral race management of the America s Cup sailing competitions 3 Eligibility certification The term Eligibility Certification means a certification issued under section 4 4 Eligible vessel The term eligible vessel means a competing vessel or supporting vessel of any registry that A is recognized by America s Cup Race Management as an official Competing vessel or supporting vessel of the 34th America s Cup as Figure 4 5 A screenshot of the LegisPro Web editor 4 5 Open issues i
114. rkup tries to parse the document in order to un derstand its structure This is possible because all the Italian laws contain some keywords in specific parts of the document For example articles start with the word articolo and some pointed lists start with the words 7 seguenti punti After the parser finished their job the drafter can integrate or modify the inferred markup The Norma editor lacks in modularity and portability because it is widely based on Microsoft Word macros and for the same reason it can not be dis tributed as an open source software u Another problem of the Norma editor is that it does not manage documents using directly the NIR XML Doc The Italian law nr 4 of 2004 January 9th also known as Legge Stanca and the European parliament encourage Italian public offices to use open source software 60 4 Software for XML markup uments are edited and saved in the Microsoft Word format and only in a second step they are translated to NIR However this editor was the first noteworthy legal markup editor It is still used in some Italian public offices and some of its ideas like the use of an interface the users are comfortable with and the legal document parsing lay the foundations for more recent legal editors f lak_act_20060926_31_eng Microsoft Word File Modifica Visualizza Inserisci Formato Strumenti Tabela Finestra Norma Editor Document
115. rmatica giuridica sono stati introdotti i markup editor software WYSIWYG che assistono l utente nella creazione di documenti corretti Questi editor possono essere utilizzati anche da persone che non conoscono a fondo XML ma d altra parte sono solitamente basati su uno specifico linguaggio XML Questo significa che sono necessarie molte risorse in termini di programmazione per poterli adattare ad altri linguaggi XML o ad altri contesti Basando l architettura degli editor di markup su parametri possibile 0 Introduzione progettare e sviluppare software che non dipendono da uno specifico linguag gio XML e che possono essere personalizzati al fine di utilizzarli in svariati contesti A tale scopo necessario innanzitutto caratterizzare i documenti strut turati I documenti strutturati sono documenti che contengono altre infor mazioni in aggiunta al loro contenuto Queste informazioni supplementari servono per descrivere la struttura logica del documento ossia gli oggetti che lo compongono e le relazioni fra questi oggetti Esistono diverse tipologie di linguaggi di markup per la marcatura dei documenti strutturati ma i linguaggi pi appropriati a questo scopo sono i linguaggi descrittivi e XML attualmente il pi completo in questo genere Negli ultimi dieci anni sono stati sviluppati svariati dialetti XML per la mar catura di documenti testuali appartenenti a molti contesti Questi linguaggi devono essere specifici
116. ronous Javascript and XML is a programming paradigm according to which a group of interrelated technologies are used to create web applications By using javascript and HTML5 with the Ajax paradigm it is possible to create dynamic in browser applications Ajax application can send data to or retrieve data from the server without interfering with the behavior of the page displayed on the browser HTML5 supplies some interesting features that are useful for creating of in browser markup editors For example HTML5 allows web applications users to directly modify elements of web pages displayed in the browser and provides some technologies for in browser storage Moreover HTML5 docu ments are already structured documents compliant to some XML patterns Javascript is the most used script language for the developing of dynamic in browser applications but because it is an interpreted language different browsers can execute it in different ways For this reason it is important to use javascript cross browser frameworks that ensures the same application s behavior in the most known browsers 5 2 2 Frameworks for cross browser software Javascript syntax can be very verbose and difficult to understand More over since it is an interpreted language and each browser has its own inter preter sometimes a javascript code that works on a browser could not work properly on another browser Javascript frameworks supply objects and methods that abstr
117. rya Nabavi Ron Kikinis W Eric L Grimson Noby Hata Peter Everett Ferenc Jolesz and William M Wells An integrated visualization system for surgical planning and guidance using image fusion and in terventional imaging In Medical Image Computing and Computer Assisted Intervention MICCAI 99 pages 809 819 Springer 1999 Seth Godin Getting Real 37signals 2006 C F Goldfarb A generalized approach to document markup In Proceedings of the ACM SIGPLAN SIGOA symposium on Text manipulation pages 68 73 New York NY USA 1981 ACM Charles F Goldfarb The sgml handbook 1991 126 BIBLIOGRAPHY HH10 HM88 HPH10 IV95 Jel01 KP 88 Lam86 LBM 96 Lew98 David Hyatt and Ian Hickson Html 5 World Wide Web Consortium WD WD html5 20100304 2010 Susan Hockey and Jeremy Martin Oxford Concordance Program User s Manual Version 2 Oxford University Computing Service 1988 Dracine Hodges Cyndi Preston and Marsha J Hamilton Resolving the challenge of e books Collection Management 35 3 4 196 200 2010 Nancy M Ide and Jean V ronis Text encoding initiative Background and contexts volume 29 Springer 1995 Rick Jelliffe The schematron An xml structure validation language using patterns in trees URL http aml asce net resource schematron schematron html 2001 Glenn E Krasner Stephen T Pope et al A description of the model view controller user interface para
118. s Speech communicator OFFICIAL REPORT Wednesday 13th July 2011 e House met at thirty minutes past Two o clock f pi Cha PAPERS LAID The following Papers were laid on the Table The Report of the Commonwealth Parliamentar Swaziland from 7th A 004 to 15th Aug ciation CPA Kenya branch delegation to the 25th CPA A amp ica Region conference held in Lobamba 4 Annual Report and Ac T fhereo iditor of Jomo Kenyatta University College of Agnculture and Technology for the year ended 30th June 1990 and the Certificate orations i Accounts of Higher Education Loans Fund for the year ended 30th June 1983 and the Certificate thereon by the Auditor General Page1 t Defaut Engish USA INSRT STD commenta BE o e 100 Figure 4 4 A screenshot of the Bungeni editor 4 4 3 LegisPro Web Editor LegisPro WebAuthor is a in browser markup editor developed by Xec ntial It is based on LegisPro Author an XML editing tool used to markup the authoring and amending of local state and federal government legislation It has been developed using HTMLS5 designed to work natively with Akoma Ntoso and it can run in the most recent web browsers On the one hand LegisPro was the first legal markup editor completely in browser and resolved the issues of the Norma and the Bungeni editors re lated to the easiness of customization and portability Also its interface was not derived from the one of a w
119. s web desktop applications web applications and desktop words processors The majority of them have a good experience in computer science or law and use quite often online word processors The minority of them are experts of legal informatics and use desktop XML editors I failed to recruit enough people skilled in other existent legal markup editors This would have been useful to compare the LIME usability to the one of the other editor However as already said this was intended to be a summative usability test and not a comparative test and for this reason the testers were enough reliable for my purposes 104 6 Evaluation of LIME s user experience Table 6 2 The summary of the testers Question pl p2 p3 p4 p5 p6 p7 p8 p9 Average age 37 38 29 30 39 26 25 38 37 33 knoweledge in computer 5 4 4 5 2 7 6 4 7 4 9 science knoweledge in law and 6 E Wer 1 7 1 1 7T 3 44 jurisprudence knoweledge in legal 3 14 5 1 5 1 5 6 14 3 8 informatics f computers fe ie e ESES e e aes ee use of web 7 7 6 7 7 7 7 7 7 J69 use of desktop applications 7 7 6 7 I6 7 7 7 7 J68 use of online applications 7 3 15 I7 I6 7 7 7 qoa J61 use of desktop word 7 7 qe 7 I7 5 7 6 64 processors use of online wor 5 7 13 12 6 5 7 J6 3 149 processors use of desktop XML editors 6 3 2 7 2 7 I1 1 3 13 6 use of M gislative MEME We oleae ll I Re ica m
120. s are used to abstract XML languages and to al low LIME to markup many of them When a document is marked up on LIME a pattern is assigned to each element and its behavior is described by parameters stored in JSON configurations files 5 6 1 XML guidelines and patterns used by LIME The LIME editor relies on six patterns the same used by Akoma Ntoso When a LIME configuration for a specific language is created the contents model of the language are matched with one of the following patterns inline block hierarchical container container marker or subflow Sometimes a content model can not be abstracted using one of the above patterns In this situation the elements using that content model are labeled as patternless elements LIME also follows XML guidelines for the creation of elements unique identifiers and elements classes 82 5 LIME a parametric editor for structured documents 5 6 1 1 Patternless the patterns wildcard A patternless element is an element that does not belong to any known markup pattern Patternless elements are really complicated to manage but are also useful to abstract XML languages that have not a patternized archi tecture An example of patternless elements can be seen in the listing In the example the element called patternlessErample is obviously a patternless element because it contains text a block and a hierarchical container and for this reason it does not belong to any of th
121. s flexible as possible in order to hit all the peculiarities of the dif ferent areas in the same context For example a novel has a quite different structure from a scientific paper One of the most challenging context is the legal one Even if legal and legislative documents rely on recurrent structures they have to meet the re quirements of a large set of different traditions rules and users Moreover different kinds of legislative documents of a single country can have different structures For instance Italian bills can be quite different from Italian acts In the last ten years many effective XML languages for the legal context were developed Some of them were designed to meet the requirements of specific countries like Norme In Rete that was designed for the Italian leg islative documents Other languages like CEN Metaler and Akoma Ntoso have a more flexible architecture and can describe documents coming from a variety of countries These languages try to abstract documents structure by using XML patterns Like programming languages patterns XML patterns describe recur rent markup situations Akoma Ntoso architecture relies on six XML pat terns containers hierarchical containers markers blocks inline and sub flow These patterns are the most descriptive and the most common ones and even if other languages do not have a patternized architecture it is possible to abstract them and to assign one of these patterns to th
122. s json a file that contains the specific configura tion for each element and the hierarchy to be used for the buttons in the markup menu e custom_buttons json a file containing custom style rules regarding buttons and elements marked by it It also contains optional rules regarding the structure of the document e custom_patterns json a file that allows to specify custom patterns and to customize the ones already existing All these files can be inserted in nested directories to specify different con figurations for different sub types of documents For example Akoma Ntoso allows to create bills acts judgments and other document types The config uration for these documents types is different from one another and for this reason it is possible to create a different folder for each document type in side the interface directory This folder must contain all the above described configuration files System administrators that want to enable LIME to markup other XML languages have simply to write these configuration files and to package them properly inside a language plugin folder The configuration files are really simple to read and modify but at the same time are very powerful In listing 5 4 it is possible to see a fragment of a configuration file used in the Akoma Ntoso language plugin With few lines of human readable code the fragment describes a button called act that has some children buttons When these buttons are cl
123. s task Very difficult O O O O O O O Very easy Can you change the editor s language to your preferred one How difficult was to complete this task Very difficult O O O O O O O Very easy Can you import a Microsoft word s file from your desktop How difficult was to complete this task Very difficult O O O O O O O Very easy Can you open one of the examples supplied by the editor How difficult was to complete this task Very difficult O O O O O O O Very easy Can you save the file under a different name How difficult was to complete this task Mery Cutie o e D aN COSY Can you open the XML preview of the example you are looking at How difficult was to complete this task Very Cunt ei Can you save the XML version of the document on your desktop How difficult was to complete this task Very difficult 0 _ _ Very easy Can you log out How difficult was to complete this task Very difficult O O O Very easy Can you set the preface of this document How difficult was to complete this task Very difficult O O O O O O O Very easy Can you set the document date How difficult was to complete this task Very difficult O O O O O O O Very easy Can you set the main body of the document 6 3 101 12c 13b 13c 14b 14c 15b 15c 16b 16c 17b 17c
124. s task and they produced four generations of legal documents XML standards PCR09 For the aim of this dissertation I consider two of these generations The second generation of legal XML languages i e Norme In Rete de scribed in the next section is focused on the document s modelling and on the description of the text but it does not contemplate a previous abstrac tion of the classes of the elements in the legal documents The third generation of legal XML standards includes languages like 3 3 35 CEN Metalex and Akoma Ntoso described in section and re spectively that are based on patterns This means that all the elements that belong to the standard are grouped in classes henceforth referred to as patterns Each of these patterns has its own semantic meaning behavior and hierarchy in respect to the other classes The XML dialects that belong to the third generation are the perfect test case for the purpose of this research Indeed a legal XML standard that hopes to be widely used in the world must be very flexible and extensible to hit all the countries needs and should also be able to abstract a generic logical structure of the laws This inspires the design of a generic markup process for all the laws and legal resources 3 3 1 Norme In Rete Norma In Rete hereafter NIR is an XML standard created in 1999 financed by the Italian Authority for Information Technology in the Public Administration and
125. se of presentational markup was discouraged in all the HTML versions it is deprecated in the current versions of HTML and XHTML and it is illegal in HTML 5 HTML and chiefly HTML 5 are good descriptive markup languages but as pointed out in the introduction of this paragraph they are intended to be used to create web pages and not to markup documents that must have their specific in context semantically relevant elements for example a cook ing recipe or a law In these cases as I will explain in the next section we find solace in the eXtensible Markup Language that consents to create specialized dialects for different contexts Listing 2 5 An sample HTML 4 01 document DOCTYPE HTML PUBLIC W3C DTD HTML 4 01 EN http www w3 org TR html4 strict dtd html head lt title gt HTML example document lt title gt lt meta name keywords content HTML example paragraph bold gt lt head gt lt body gt lt h1 gt Hello wolrd I m an HTML document lt h1 gt lt p gt This is my content I have a paragraph 2 3 23 and a lt b gt bold face lt b gt text inside lt p gt lt body gt lt html gt 2 3 3 The eXtensible Markup Language The eXtensible Markup Language henceforth referred to as XML is a simple and flexible descriptive markup language derived from SGML BBt99 It defines a set of rules aimed to producing human readable and machine readable documents and it
126. so gt 3 4 5 Inline Inline elements are similar to blocks elements because they can contain other inline text and markers Unlike block elements they do not split the text but are used to add presentational markup to the document or to give specific semantic meaning to specific parts of text The docTitle element and the b element in the listing 3 8 are examples of inline elements 3 4 6 Subflow Subflow elements are containers appearing in the middle of sentences but containing full structures They are used when a foreign structured text with its own markup rules must be inserted in the middle of the text of the document The fragment of markup in listing 3 9 shows the usage of the Akoma Ntoso mod element Listing 3 9 A fragment of an Akoma Notoso document showing the usage of the mod element lt xml version 1 0 encoding UTF 8 akomaNtoso xmlns http docs oasis open org legaldocml ns akn 3 0 CSD05 xmlns html http www w3 0rg 1999 xhtml gt s OMISSIS a lt body gt lt article id art1 gt lt num gt Article 1 lt num gt omissis lt content gt lt p gt by inserting the following new subsection immediately after 3 Structural differences in XML dialects lt ref id ref5 href ak act 2010 08 27 1 main sec47A sub3 gt subsection 3 lt ref gt mod id sec2 lstl itmb mod1 gt lt quotedStructure id sec2 1lstl itmb modl qgtdl gt
127. soft Word and Open Office The editor offers the most common formatting tools like bold italic underline lists and so on and can be configured in order to display only subsets of these tools TiniMCE is designed to be easily integrated in content management sys tems but it is possible to integrate it in ExtJS by developing a new compo nent This component preserves all the functionality of TinyMCE editor but the effects of them are intercepted by the core of the ExtJS application 5 2 4 REST style communication REST stands for Representational State Transfer and is the foundation of the RESTful architecture Fie00 It emphasizes the abstraction of data and services as resources that can be requested by clients by using the resources name and address specified as a Uniform Resource Locator URL It revolves around five fundamental notions a resource e g a document 5 2 69 or image the representation of a resource synchronous request response interaction over HTTP to obtain or modify such representations a web page as an instance of the application state and engines e g browser crawler to move from one state to the next REST specifies a client stateless server architecture in which each request is independent from the previous ones inducing the property of scalability For example the following request DELETE photos 17 will be mapped to the photo whose ID is seventeen and will perform the desir
128. surely improved but must not be considered as problematic These are the ones involved in those tasks that got an efficacy rating barely above fifty percent The test meets my requirements and demonstrates that overall LIME is an efficient markup editor specially in the legal context 6 4 4 Analysis of the users satisfaction In order to measure the users satisfaction I asked them to foresee the effort needed to complete the nineteen tasks Then after they completed each task I asked them how difficult it had been to complete the task Table lists the average expectation and experience ratings for each task 114 6 Evaluation of LIME s user experience Table 6 9 the average expectation rating and the average experience rating Task Expectation rating avg Experience rating avg T1b 6 4 3 4 T2b 6 7 6 3 T3b 6 7 6 6 T4b 6 1 3 T5b 6 4 3 T6b 6 6 4 7 T7b 6 3 6 6 T8b 6 8 6 0 T9b 5 2 6 9 T10b 3 2 6 0 T11b 2 7 5 4 T12b 2 8 5 7 T13b 3 3 3 5 T14b 5 7 6 9 T15b 4 7 6 8 T16b 5 4 5 2 T17b 5 8 6 3 T18b 5 4 6 7 T19b 2 7 2 8 After this phase I inserted the average expectation and average experi ence ratings in the scatterplot as shown in figure 6 3 6 4 115 rience Rating rage Expe AJ N Av Average Expectation Rating Figure 6 3 Average Expectation and Experience Ratings per Task The diagram in the fig
129. t lt capo id 1 gt lt num gt CAPO I Disposizioni urgenti per il perseguimento di obiettivi di razionalizzazione della spesa nelle pubbliche amministrazioni nell societa partecipate lt num gt lt articolo id 1 gt num Art 1 lt num gt comma id artl coml gt lt num gt 1 lt num gt lt corpo gt lt h p h style text align center gt IL PRESIDENTE DELLA R T PUBBLICA lt h p gt h br h p h style padding left 2px gt Visti gli articoli 77 e 87 della Costituzione h p lt h p h style padding left 2px gt Ritenuta la straordinaria necessita ed urgenza di emanare disposizioni in materia di pubblico impiego al fine di razionalizzare e ottimizzare i meccanismi assunzionali e di favorire la mobilita nonche di garantire gli standard operativi e i livelli di efficienza ed efficacia dell attivita svolta dal Corpo nazionale dei vigili del fuoco e in altri settori della pubblica amministrazione h p omissis lt comma gt omissis lt articolo gt omissis lt capo gt omissis lt articolato gt lt Legge gt 3 Structural differences in XML dialects lt NIR gt 3 3 2 CEN Metalex The CEN Metalex BHV 08 standard is a standard proposed in 2006 during the CEN workshop on an Open XML interchange format for legal and legislative resources The first version o
130. t have unique identifiers that are composed by three letters of the element s name and by a sequential number In this way documents parser can analyze the unique identifiers of the element and can immediately understand what is the type of the element and the numbers of the same elements in the document The other guideline that LIME follows is the one that specifies that ele ments must have unique identifiers that explicate their nesting This allow to understand the nesting of the document and the relations among the ele ments in the documents In listing it is possible to see examples of the unique identifiers as signed to elements belonging to an Akoma Ntoso document in editing mode 5 6 1 3 Guidelines for elements classes LIME also relies on guidelines for elements classes Classes in LIME documents contain the name of the element that users want to create and the pattern it belongs to This is useful because in this way it is possible to create generic XSLT stylesheets to translate the document displayed in edit mode into the final 84 5 LIME a parametric editor for structured documents one independently from the language to which the final document must be compliant to In listing 5 3 it is possible to see how LIME assigns classes to documents elements Listing 5 3 A sample of the elements unique identifiers and elements classes assigned to a LIME document lt div class akomaNtoso conta
131. t the dish inside it and set the oven to one hundred seventy degrees We will also check if he expected the oven handle on the right side and his feelings during the whole process for example we check if he feels frustrated because he can not figure out how to set the alarm clock of the oven In the same way the aim of a user experience s test on a markup editor is generally speaking to check three different things Firstly we need to examine the actions that the user performs in order to open save and close files secondly his behavior in documents markup and last but not least his feelings during the overall process This test must then be refined for specific editors in order to identify the issues related to their specific context In the next section I will explain the goals of the user experience s test created to evaluate LIME editor 6 2 Goals of the test The aim of this test is to evaluate the usability of LIME The test is structured as a summative usability test TA08 Therefore it is created to examine how well the editor and its functionality meet their objectives In order to do this the efficacy and the efficiency of the editor and the satis faction of the users will be evaluated Sha91 Efficacy and efficiency are about what users actually do when they inter act with the editor trying to accomplish a task In the sections and 6 2 2 I will explain the tasks that where chosen to examine these character
132. ters llle 104 6 3 The tasks used to evaluate LIME eficacy 105 6 4 The summary of the tasks completed by users 1 indicates completed tasks 0 indicates not completed task 106 6 5 The tasks used to evaluate the LIME efficiency 108 6 6 The summary of the tasks performed by users 1 indicates completed tasks 0 indicates not completed tasks 109 6 7 the average time in seconds for completion of tasks NC indi cates the tasks that were not completed 110 6 8 The evaluation of LIME efficiency 111 6 9 the average expectation rating and the average experience rating 14 135 136 LIST OF TABLES Listings eda a 16 TETTES 17 POEET EAE 20 fara ea Ed i 20 25 An sample HTML 4 01 document 22 3 1 A TEI document describing the source in figure3 1 30 Pu 33 36 RIEN 39 pr 42 PETITE EE 42 Da S pd EA 43 RW vai es di rl c es eee a 48 rre 49 5 1 An example of patternless element 82 5 2 A translated patternless element ss 82 137 138 LISTINGS 5 3 A sample of the elements unique identifiers and elements classes assigned to a LIME document 84 5 4 Am example of a LIME configuration file 2 87 Special thanks I apologize to those who do not speak Italian But what I have to say here must be said in my mother tongue Anyway thanks for rea
133. the XML documents marked up through the editor This database is hosted by a Tomcat Application Server and resides on a separate machine from the one that hosts the server side script of the applications 5 LIME a parametric editor for structured documents The server side components are hosted by an Apache web server and are responsible for the parsing of the documents and act as proxy for the two database of the software The client components have in charge the business logic of the system They draw the user interface intercept the interaction of the user with it supply the features for the markup of documents and interact with the server side components to manage documents and users Figure 5 6 shows the LIME architecture In the next sections I will deeply explain the roles of the client side components and of the server side compo nents Parse Documents Get Docments Save Documents Get User Into Save User Info ume Tomcat Web Servgr Exist XML DB Exist Web ID Services Figure 5 6 The LIME editor architecture 5 5 79 5 5 1 Client side components The core of the application is constituted by a small set of ExtJS com ponents When the application starts the main viewport is loaded by the application and the editor the marking menu the explorer and the main toolbar are instantiated inside it The editor is the central part of the application but when s
134. ts a SGML document that is tag valid and type valid according to the document type declaration in listing 2 3 SGML is still used for small scale and general purpose applications One of its derivative is the HyperText Markup Language that is described in the next section Listing 2 3 A fragment of a document type declaration lt ELEMENT anthology poem gt lt ELEMENT poem title stanzat gt lt ELEMENT title O PCDATA gt lt ELEMENT stanza O linet gt lt ELEMENT line O O PCDATA gt Listing 2 4 A type and tag valid SGML document lt DOCTYPE anthology SYSTEM anthology dtd gt lt anthology gt lt poem gt lt title gt The SICK ROSE lt title gt lt stanza gt lt line gt O Rose thou art sick lt line gt lt line gt The invisible worm lt line gt lt line gt That flies in the night lt line gt lt line gt In the howling storm lt line gt lt stanza gt lt stanza gt lt line gt Has found out thy bed lt line gt lt line gt Of crimson joy lt line gt lt line gt And his dark secret love lt line gt lt line gt Does thy life destroy lt line gt lt stanza gt lt poem gt lt more poems go here gt lt anthology gt 10Courtesy of the stanford edu site 2 3 21 2 3 2 The HyperText Markup Language The HyperText Markup Language BLC95 hereafter HTML is a World Wid
135. ure above must be divided in four sectors in order to be read properly In the lower right sectors there are the tasks that users thought would be easy but actually turned out to be difficult These are the problematic tasks already highlighted by the efficacy and efficiency measurements and precisely the tasks where users were asked to sign up to the system to open examples and to save documents under a different name The fact that they are in the lower right sectors means that they need to be fixed as soon as possible In the upper left sectors there are the tasks that users thought would be difficult and were actually easy In these tasks users were asked to markup legal parts of documents such as the preface the documents data and their main body This is a very useful data because even these tasks do not have a very high efficiency score fifty nine percent sixty three percent and sixty 116 6 Evaluation of LIME s user experience three percent respectively the fact that they belong to the upper left sector of the scatterplot means that they must be promoted and maybe improved Very difficult tasks to analyze are the ones belonging to the lower left sector of the diagram In this sector there are the tasks where users had to markup an article its number and its heading and the one where users were asked to save documents as new expressions Users thought that these tasks would be difficult and indeed they were so This d
136. workflow of markup On the one hand these are powerful tools that can be used even by people that have no knowledge of the XML syntax and XML dialects But on the other hand these software are strictly dependent from the XML language they use to format documents and from the context they belong to This means that in order to use these software in other contexts customization is needed and this can be expensive if the editor was not de 5 1 Introduction signed adequately in the first place By relying the architecture of markup editors on parameters it is possible to create markup editors that are independent from any specific XML lan guage and that can be adapted to several contexts with the minimum effort in terms of coding In order to create a parametric editor for structured documents it is im portant to understand what structured documents are Structured documents are documents that contain other information in addition to their content This information adds semantic meaning to documents or to fragments of them and describe documents logical structure The logical structure is a description of the objects that compose a docu ment and of the relations among them It is ascribed to documents by using appropriate and standardized markup languages Different kinds of markup languages were developed in the course of time but descriptive markup lan guages are the most suitable ones for the description of documents logic
137. written 4See the TEI s module s documentation page to read more about this module and its usage 30 3 Structural differences in XML dialects example it is suitable both for those projects that may simply wish to trans late an existing catalogue into a format that can be displayed on the web and for the projects whose aim is to create detailed databases of highly struc tured information that are useful for the quantitative codicology A TEI P5 document contains a msDesc root element in which it is possi ble to specify information like its cataloging with the ms dentifier element its content using msContents elements and its history marked up whit the physDesc element Listing 1 shows the markup of the sample manuscript s description in figure 3 1 E The TEI standard is approved by a lot of organizations like the National Endowment for the Humanities and the Modern Language Association and should be used in the future to create a big standardized online digital library 28843 In Latin on parchment written in more than one hand of the 13th cent in England 7 x sf in i 5 leaves in double columns with a few coloured capitals Hic incipit Bruitus Anglie the De origine et gestis Regum Angliae of Geoffrey of Monmouth Galfridus Monumetensis deg Cum mecum multa amp de multis On fol 54 very faint is Iste liber est fratris guillelmi de buria de Roberti ordinis fratrum Pred
138. you expect it will be to markup a bold text inside it Very difficult O O O O O O O Very easy 15a Having a marked up article how difficult do you expect it will be to markup a table inside it Very difficult O O O O O O O Very easy 16a Having a marked up article how difficult do you expect it will be to markup a quoted structure inside it Very difficult O O O O O O O Very easy 17a How difficult do you expect it will be to find the subsection 2 of section 3 in a complete marked up document Very difficult O O O O O O O Very easy 18a Having a complete marked up document how difficult do you expect it will be to un mark the preface and all the elements inside it Very difficult O O O O O O O Very easy 19a Having a complete marked up document how difficult do The task submitted to users and the questionnaire the compiled you expect it will be to save it as a new expression in the same work Very difficult O O O O O O O Very easy after the tasks completion 1b lc 2b Can you sign up to the system How difficult was to complete this task Very difficult O O O O O O O Very easy Can you log in with your account 100 6 Evaluation of LIME s user experience 2c 3b 3c Ab Ac 5b 5c 6b 6c Tb 7c 8b 8c 9b 9c 10b 10c 11b 11c 12b How difficult was to complete thi
139. you save this document as a new expression in the same work In order to evaluate the efficiency it is important to check how much time users need to complete these tasks and the actions that they perform in order to do that It is important to underline that a failure in these tasks is admissible because it should not result in user s frustration but leads the 6 2 93 user to find a different way to markup the document For example if users are not able to markup bold text they can simply decide to skip it maybe because they think that the editor does not supply this feature or they can try other strategies to achieve a similar result for example using a generic inline and setting its class to bold This does not mean that failures must be ignored indeed a lot of failures can make the user to feel frustrated and abandon the system Therefore in this part of the test it is important to check the number of successes the number of failures and the users behavior while they perform the tasks 6 2 3 Examining the users satisfaction Satisfaction is evaluated collecting a set of users self reported data Self reported data are really important because they capture users actual feelings while they are using the system Indeed the data that come out of a summa tive usability test can be very different from the data that come out of users self reported data and in some situations the latter can be more signi

parametrics editors for structured documents

Contents

Download Pdf Manuals

Related Search

Related Contents