Sessions

Four formats wrestle with each other for web glory. By their acronyms we shall know them: HTML, XML, JSON, RDF. Sometimes they clash, and sometimes they merge, forming weird and wonderful hybrids. Is there any way for them to work together? I will talk about the problems of mixing models and describe how we are using these formats together in legislation.gov.uk.

Jeni Tennison

Jeni Tennison is an independent consultant and developer currently contracted to The Stationery Office in the UK. She specialises in XML and linked data with forays into client-side development. She trained as a knowledge engineer, gaining a PhD in collaborative ontology development. Jeni is the technical architect and lead developer for legislation.gov.uk and previously worked on the linked data aspects of data.gov.uk. She is author of several books including “Beginning XSLT 2.0” (Apress, 2005), was an invited expert on the XSL and XML Processing Working Groups at the W3C and was appointed to the W3C's Technical Architecture Group in 2011.

Revisiting the question that was the tag line of XML Prague last year: "XML as new lingua franca for the Web. Why did it never happen?", Eric tries to answer to some other questions such as: "where is XML going?" or "is XML declining, becoming an eX Markup Language?".

Eric van der Vlist

Eric is an independent consultant and trainer. His domain of expertise includes Web development and XML technologies.

He is the creator and main editor of XMLfr.org, the main site dedicated to XML technologies in French, the author of the O'Reilly animal books XML Schema and RELAX NG and a member or the ISO DSDL (http://dsdl.org) working group focused on XML schema languages.

He is based in Paris and you can reach him by mail (vdv@dyomedea.com) or meet him at one of the many conferences where he presents his projects.

W3C created a small Task Force to look at convergence paths between XML and HTML. One of the notions that it discussed was that the two technologies could perhaps not be aligned, but that they could cross-pollinate.

This talk will look at ways for this to happen. Can we use CSS Selectors in XPointer? Can we build something like XSLT using CSS + JavaScript and another syntax? Does HTML actually have some interesting approaches to distributed extensibility? Should SVG be in the HTML namespace and has it suffered from changing its syntax?

Norman Walsh

Norman Walsh is a Lead Engineer at MarkLogic Corporation where he works with the Application Services team. Norm is also an active participant in a number of standards efforts worldwide: he is chair of the XML Processing Model Working Group at the W3C where he is also co-chair of the XML Core Working Group. At OASIS, he is chair of the DocBook Technical Committee.

Robin Berjon

Robin has spent the better part of this decade working on standards for a variety of Web and XML technologies, mostly within the W3C. He has served as author or editor for more than a dozen W3C standards, has chaired multiple groups working on XML optimisation and Web APIs, and has recently been elected to the TAG. He also hacks on Javascript stuff a lot. In his copious spare time he reads, writes, and gets bitten by his cat.

This talk will take a brief look at what it takes to replace XML with a fully backwards compatible revision of XML that no longer halts processing at violations of well-formedness and why that might be worth doing.

Anne van Kesteren

Anne van Kesteren writes standards for the web platform from a smallish town in the Netherlands, Utrecht. He does so for Opera, maker of browsers.

Norman Walsh

Norman Walsh is a Lead Engineer at MarkLogic Corporation where he works with the Application Services team. Norm is also an active participant in a number of standards efforts worldwide: he is chair of the XML Processing Model Working Group at the W3C where he is also co-chair of the XML Core Working Group. At OASIS, he is chair of the DocBook Technical Committee.

Eric van der Vlist

Eric is an independent consultant and trainer. His domain of expertise includes Web development and XML technologies.

He is the creator and main editor of XMLfr.org, the main site dedicated to XML technologies in French, the author of the O'Reilly animal books XML Schema and RELAX NG and a member or the ISO DSDL (http://dsdl.org) working group focused on XML schema languages.

He is based in Paris and you can reach him by mail (vdv@dyomedea.com) or meet him at one of the many conferences where he presents his projects.

Jeni Tennison

Jeni Tennison is an independent consultant and developer currently contracted to The Stationery Office in the UK. She specialises in XML and linked data with forays into client-side development. She trained as a knowledge engineer, gaining a PhD in collaborative ontology development. Jeni is the technical architect and lead developer for legislation.gov.uk and previously worked on the linked data aspects of data.gov.uk. She is author of several books including “Beginning XSLT 2.0” (Apress, 2005), was an invited expert on the XSL and XML Processing Working Groups at the W3C and was appointed to the W3C's Technical Architecture Group in 2011.

Anne van Kesteren

Anne van Kesteren writes standards for the web platform from a smallish town in the Netherlands, Utrecht. He does so for Opera, maker of browsers.

Steven Pemberton

Steven Pemberton is a researcher at the CWI, Amsterdam, and chair of the W3C XForms working group. At the end of the 80's he designed and built a browser with extensible markup, stylesheets, vector graphics, and client-side scripting. He has been involved with the Web from the beginning, organising two workshops at the first Web conference in 1994, chairing the first W3C Style Sheets Workshop in 1995, and chairing the HTML working group for a decade. He is co-author of amongst others HTML 4, CSS, XHTML, XForms, RDFa, and XML Events.

Although primarily an XML processing language, XProc is increasingly being used in environments that involve processing of non-XML data. However, the limited support – or lack thereof – for non-XML media types in XProc poses real issues, both for pipeline authors as well as performance-wise. This article looks at some of these issues and explores the possibilities of extending XProc to support processing of both XML and non-XML data.

Vojtěch Toman

Vojtěch Toman is Consultant Software Engineer in the Information Intelligence Group of EMC Corporation where he is involved in the development of XML data management and processing technologies. He is a member of the W3C XML Processing Model Working Group.

NVDL stands for Namespace-based Validation and Dispatching Language. It is an ISO standard, like Relax NG and Schematron. NVDL allows to validate documents containing markup from different vocabularies without the need to change the schema of each vocabulary to know about the others, to use different schema languages to validate different parts of the document and to perform multiple validations.

In this presentation I will show an XSLT implementation of the NVDL dispatching plus an XProc orchestration of dipatching and validation tasks. This will allow to quickly understand how NVDL works, especially for people with an XSLT background.

I developed the XSLT implementation of NVDL dispatching initially as part of oNVDL (the open source implementation of NVDL now contributed to Jing) to help me understand how NVDL works. Now, having XProc available it is easy to put all the processing together, add also the validation steps and create a complete NVDL implementation based on XProc and XSLT.

George Bina

George Bina is one of the founders of Syncro Soft, the company that develops oXygen XML Editor. He has more than 12 years experience in working with XML and related technologies including XML related projects, oXygen XML Editor and participation in open source projects, the most notable being oNVDL - an open source implementation of the NVDL standard, project that is now merged into Jing.

XML and JSON have become the dominant formats for exchanging data on the Internet, and applications frequently need to send and receive data in many different JSON-based or XML-based formats. For XML data, a query language like XQuery can be used to query data, create or update data, transform it from one format to another, or route data. Adding JSON support to XQuery allows it to perform these tasks for both XML and JSON, combining data from multiple sources as needed. In addition, JSON support gives XQuery a lightweight, simple, and useful data structure that can often simplify queries.

JSONiq is a query language for JSON, based on XQuery. It is designed to allow an existing XQuery processor to be rewritten to support JSON with moderate effort. One profile of JSONiq removes everything directly related to XML, adding JSON constructors and navigation. Another profile of JSONiq includes the full XQuery language, with added JSON support, allowing queries to consume or produce JSON, XML, or HTML.

Jonathan Robie

Jonathan is an architect at EMC Corporation's Information Intelligence Division, and the lead editor of XQuery 3.0, XPath 3.0, and JSONiq. He was the program manager for DataDirect XQuery, and served on the architecture teams of XML repositories at Software AG, Texcel, and POET Software. He has also worked on AMQP messaging software at Red Hat.

Matthias Brantner

Matthias is the chief architect of 28msec. He studied Information Systems at the University of Mannheim in Germany from 1999 until 2004. In 2007, he acquired a PhD from the University of Mannheim, based on his research on rewriting declarative query languages. Matthias published several papers on XML query processing and co-developed one of the first research native XML databases.

Ghislain Fourny

Ghislain studied Computer Science at ETH Zurich from 2004 to 2007 with a focus on Software Engineering. He continued there as a Ph.D. student with Donald Kossmann in the area of XML-based Web applications, and passed his defense in December 2011. His background is also tinted with Mathematics and Physics. Ghislain has been working on XQuery and XML processing technologies since 2007, in projects such as XQuery in the Browser or the XML Time Machine. He is now a software architect at 28msec and a member of the W3C XML Query Working Group.

Till Westmann

Till is the vice president of engineering at 28msec. He has over 10 years of experience in the data management industry. He worked on XML processing since 2000 and on XQuery processing since 2002. Till holds a PhD in computer science from the University of Mannheim and has published a number of papers in academia and in industry.

This talk will be presented by Norman Walsh

What if you tried to provide the core value of MarkLogic Server to people who didn’t want to learn XQuery? The very idea might be blasphemous in XML crowds, but is it possible? How far could you get?

This paper explores that issue and shares our experience in designing and developing the open source "Corona" project. Corona provides a set of REST endpoints to store, retrieve, query, and analyze documents held inside MarkLogic. Corona exposes the vast majority of MarkLogic’s most popular XML-aware indexes and features, but in a way that lets them code in whatever language they’d like.

XForms 1.0 was an XML technology originally designed as a replacement for HTML Forms. In addressing certain shortcomings of XForms 1.0, the next version, XForms 1.1 became far more than a forms language, but a declarative application language where application production time could be reduced by an order of magnitude compared with traditional procedural programming.

Although XForms treats its data internally as if it is XML, using XPath both to address data and to calculate new values, it is not the intention that external data necessarily be only in XML.

An obvious data format widely in use on the web is JSON. There are several mappings defined in both directions between XML and JSON, but largely because JSON can only represent a subset of what XML can represent, many of the mappings are cumbersome, and make data-references both JSON-specific, and difficult to write.

Ideally, an XForm processing JSON data shouldn't have to know which data format has been used, so that JSON data can be selected with natural XPath selectors. Furthermore, XForms doesn't need the full generality of translating any XML to JSON, since the only need is to read and write data to and from existing JSON sources. In other words, it only needs to process existing JSON. This simplifies the mapping, and makes the selectors needed with minor exceptions opaque to the data format.

This paper presents the mapping proposed for XForms 2.0, the special cases that have had to be dealt with, and discusses generalisation to other formats, such as VCARD.

Steven Pemberton

Steven Pemberton is a researcher at the CWI, Amsterdam, and chair of the W3C XForms working group. At the end of the 80's he designed and built a browser with extensible markup, stylesheets, vector graphics, and client-side scripting. He has been involved with the Web from the beginning, organising two workshops at the first Web conference in 1994, chairing the first W3C Style Sheets Workshop in 1995, and chairing the HTML working group for a decade. He is co-author of amongst others HTML 4, CSS, XHTML, XForms, RDFa, and XML Events.

Michael Kay

Michael Kay is founder and director of Saxonica Limited, lead developer of the Saxon XSLT and XQuery processor, lead editor of the XSLT 2.0 and 3.0 specifications, and a member of the XSLT, XQuery, and XML Schema working groups.

Jonathan Robie

Jonathan is an architect at EMC Corporation's Information Intelligence Division, and the lead editor of XQuery 3.0, XPath 3.0, and JSONiq. He was the program manager for DataDirect XQuery, and served on the architecture teams of XML repositories at Software AG, Texcel, and POET Software. He has also worked on AMQP messaging software at Red Hat.

Whilst XQuery was originally envisaged and designed as a query language for XML, it has been adopted by many as a language for application development This, in turn, has encouraged additional and diverse extensions, many of which could not easily have been foreseen.

This paper examines how XQuery has been used for Web Application development, current implementation approaches for executing XQuery in a Web context, and subsequently presents a proposal for a standard approach to RESTful XQuery through the use of XQuery 3.0 Annotations.

Adam Retter

Adam Retter is an independent consultant, typically helping organisations with their information and data assets, by making them accessible, searchable and transformable. Adam is a long time contributor and core team member of the eXist-db Native XML Database open source project. Adam is an invited expert on the W3C XML Query Working Group and W3C XML Provenance Working Groups. Adam is also a member of the XML Guild, a group of leading independent XML consultants. With a professional background in Software Engineering and a passion for the Internet, he wants to share his belief that building advanced Web applications can be simple, by bringing XML Application Development to the mainstream.

There are different approaches for having XQuery in the browser. Developing a plugin is hazardous because of the number of versions of browsers. Performance is to be considered when porting an existing Java implementation with an engine such as GWT. Compiling XQuery code into Javascript instructions is another possibility and XML technics can facilitate this. XQueryX is an XML notation for XQuery code and converting into it is a first step to consider. Then, the resulting tree can be transformed with XSLT back to Javascript. XSLT 1.0 is powerful enough to write such a compiler, from text to tree and back to text (with the nodeset function), for both XQuery and XQuery Update Facility. This has been demonstrated by XSLTForms with its own XPath 1.0 engine: an XSLT 1.0 stylesheet to transform (with errors detection) XPath 1.0 expressions into Javascript objects and a collection of Javascript classes to effectively evaluate the XPath 1.0 expressions. YAPP is another interesting tool for generating a XSLT 1.0 parser from the grammar of a language expressed in BNF notation (XQuery Grammar is currently described in EBNF notation). After parsing, for more performance, instead of having the resulting Javascript instructions being the image of the tree, it is possible to generate the minimal number of effective instructions after optimization ([n] detection, maximal use of DOM API, ...), as compilers should always do. Finally, there are functions to be written so the generated instructions are limited to pertinent loops and calls.

Alain Couthures

Alain Couthures is the project leader for XSLTForms which is a client-side XForms implementation based on XSLT and Javascript. He is an Invited Expert in the W3C Forms Working Group.

The idea for Carrot, "an appetizing hybrid of XQuery and XSLT," was first presented at Balisage 2011. Since then, progress has been made on implementing Carrot by way of compilation to XSLT. This paper describes current progress on the Carrot parser and compiler, detailing certain fundamental aspects of the current architecture.

Evan Lenz

Evan Lenz has been a specialist in XML technologies since 1999, having served on the W3C XSL Working Group, written XML-related articles and books (including XSLT 1.0 Pocket Reference), and spoken at numerous conferences. He is currently working as Software Developer for the Community team at MarkLogic Corporation. When not at work, he enjoys piano, rowing, and being a dad.

It has long been held that one of the use cases for which XSLT excels over XQuery is document transformation. With its central rule based template engine and extensibility features, specifying and customising a transformation are straightforward. Although XQuery users have achieved transformation capabilities using recursive functions and type switch expressions, these have been found lacking in the expressiveness of the pattern syntax, and the extensibility of the resulting transformation code.

XQuery 3.0 provides many powerful new features which extend the boundaries of what can be accomplished in an XQuery program. This paper introduces Transform.xq, an XQuery 3.0 module which implements rule based transformations using higher order functions. It further extends this library with automatic construction of template "modes" using reflection extensions to examine the XQuery 3.0 annotations on functions designated for use as template rules.

John Snelson

John Snelson is a Lead Engineer at MarkLogic. He participates in the W3C XQuery and XSLT working groups, and maintains XQilla, an open source implementation of XQuery. He enjoys all things to do with computer languages and databases.

The Java ecosystem surrounding relational databases has enjoyed years of improvements such as a standard means to invoke stored SQL code from JDBC and time-saving object relational mapping (ORM) frameworks such as Hibernate.

For Java programmers working with XQuery and XML databases, there is the XQuery API for Java (XQJ); however it lacks the means to invoke stored XQuery code and currently provides no answer to the very successful ORM paradigm.

This paper introduces a novel, RPC style approach to calling XQuery from Java. Through the use of Java reflection, Java programmers are now able to invoke XQuery functions as if they were regular Java methods, with ease.

As for parameters and return types of both Java methods and XQuery functions, Java data types are mapped to XDM data types according to the rules described in the XQJ specification. Additionally, POJOs are mapped to XML which may save programmers a great deal of time, just as ORM technologies have done for SQL.

This paper proposes extensions to the XQJ APIs as well as discussing some initial implementations, namely MarkLogic, eXist and Sedna XQJ APIs.

Charles Foster

Charles Foster is a freelance XML Consultant and Software Engineer, who focuses on creating scalable, high performance XML and standards based solutions for difficult problems.

The community of users of a large data-driven web site may directly contribute to its management by feeding corrections and new additions, thus keeping “fresh” the information provided by the site. However, several issues may arise due to the fact that users may modify data in a more or less controlled way. Starting from a real-world scenario, we point out such issues and we present a simple and efficient framework. The proposed solution has been implemented in a XML-based prototype framework, that have been tested with large, real-world datasets.

Lorenzo Bossi

Lorenzo Bossi is a PhD candidate in Computer Science at the Insubria Univesity – Italy – since 2010. His main research interests are related to data management and data schema evolution.

Alberto Trombetta

Alberto Trombetta is assistant professor at Insubria University and works on novel approaches to data management - such as wiki-based DBs - as well as on security issues arising from such approaches.

Michael Sperberg-McQueen

C.M. Sperberg-McQueen is the founder of Black Mesa Technologies LLC, a consultancy specializing in information management using descriptive markup.

Before launching Black Mesa, he served for ten years as a member of the technical staff at the World Wide Web Consortium (W3C). There he chaired the XML Coordination Group and served on the XML, XML Schema, XSL, Service Modeling Language, and XML Processing Model working groups, and served as editor of the XML 1.0 and XML Schema 1.1 specifications. Before joining W3C he spent fourteen years working in the academic computer centers of Princeton University and the University of Illinois at Chicago, where he served as a consultant for humanities computing, developed database applications, ran documentation groups, and maintained library automation systems.

He was a founder of the international Text Encoding Initiative and served from 1988 to 1999 as the editor in chief of the TEI's Guidelines for Electronic Text Encoding and Interchange. He has also served as a co-coordinator, of the Model Editions Partnership, and has held adjunct positions at the University of Bergen (Norway) and the University of Illinois Graduate School of Library and Information Science.

He serves as a co-chair of Balisage: The Markup Conference, held annually each August in Montreal.