Applying XSLT in the real world with specific attention to how certain existing and well-known use-cases become more accessible and easier to manage with features introduced in XSLT 3.0. We will look in particular about processing a Twitter feed, a technique that can be used with any feed, be it Facebook messages, weather forecasts or a news feed. For this we'll explore new techniques and a couple of best-practices that have come to fruition during my recent work in the XSLT fields and from feedback I have received from customers. We will see how a streaming approach is easier than a traditional approach in set-up, configuration and management, how the Ten Rules of Thumb (TODO: reference to earlier paper) help us write the stylesheet efficiently and quickly and how an uninterrupted stream can be processed by a single stylesheet, gaining performance and stability in comparison to the traditional approach.

The regulatory burden for financial institutions makes it a hard environment to innovate, yet the unfavourable market conditions mean that banks must adopt latest technologies to cut costs and keep up with the market demands.

This paper aims to show a technical proof of concept of how to combine the seemingly opposite goals using a combination of tried and tested XML technologies such as XSLT and Schematron in conjunction with experimental distributed ledger technology to drive both regulatory compliance and implement innovative features such as inter-bank trade settlement using blockchain technology.

Most of the publishing world has long embraced the value of structuring content using SGML/XML in the form of DocBook, DITA or any number of vertical, standardized publishing markup schemes. Yet those who publish magazines have stubbornly remained in a world where page-layout-based workflows predominate. While this remains true today, for the first time magazine publishers are flirting with a change to their traditional publishing workflow that would employ XML-first content creation. In this session you will gain an understanding of the two worlds of publishing, why magazine publishers have been so reluctant to embrace XML technologies and the emerging trends that may bring magazine publishers into the XML publishing domain.

CALS tables are used in many technical documentation standards. There are OASIS specifications for CALS tables which include a number of semantic rules to ensure table validity. This paper reports on some of our experiences with CALS table processing and validation. We implemented the majority of our validation code in XSLT and have needed to carefully consider performance when handling large tables of several thousand rows.

We have experimented with a number of new XSLT features when addressing performance issues and will report on our experiences. In addition to processing tables we wished to improve the quality of CALS tables that we would meet and which our users/customers would produce (we wished to rid the world of bad tables!). For this we have tried to use schematron to check and report the validity of tables in a user friendly way. We met a number of obstacles and will report on these and our solutions ('work-arounds') in this paper.

Looking at version control systems, XML data updates, or simply code generation, merging XML based documents has become an omnipresent challenge these days. Each problem domain comes with its own facets of merge algorithms, like three-way merges, two-way merges, and patch applications. In this paper, we will focus on the problem domain of code generation applying patch semantics for generated XML document patches to already existing XML documents. As XML Schemas and DTDs just cover syntax of XML documents, patch application become even harder when language semantics are taken into account. Additionally, due to the fact, that one XML document can be based on multiple languages separated by namespaces, an appropriate merge mechanism has to focus on each language description separately rather than just covering XML syntax. For this need, we developed a XML-based descriptive language called MergeSchema. It is designed for controlling a two-way XML merge to also cover semantical constraints of a XML-based language during merge processing. As part of the MergeSchema, the specification of the equality of nodes is partially based on XPath.

Linked Data Templates define the syntax and the semantics of a Linked Data processor which publishes and consumes RDF data over HTTP. The processor responds to Linked Data requests by interpreting a sitemap ontology as instructions to indicate how the request metadata maps to an operation on SPARQL service, and how to generate response body. The LDT vocabulary also provides capabilities to define hypermedia controls, container resources with paginated access, resource constructor templates, validation constraints and skolemization templates.

Big Data tools and techniques are starting to make significant contributions in clinical research and studies. We explore the use of XML for holding data in an electronic health record, where the primary data storage is an open source XML database of clinical documents. We evaluate the feasibility of using such a data store for Big Data and describe the techniques used to extend to the massive data sets required for meaningful clinical studies.

Using an open source Electronic Health Records system we have loaded the database with a set of patient records and measured the size of the database from 1 to [500,000] patients, together with the response time of a typical query to retrieve and combine data across a cohort of patients.

We then describe the implementation of a federated data store, whereby we can scale to millions of patient records.

This paper proposes a method how to optimize the set of DSDL validation methods for an arbitrary industry vocabulary. The research is based mainly on a practical case study of creating such an optimized set of DSDL validation artefacts for XLIFF 2, a complex industry vocabulary. Available schema languages have advanced functionality, enhanced expressivity and can be used in concert if needed. This advantage, on the other hand, makes the creation of a stable and robust set of validation artefacts hard, because there would usually be more than one way to describe the same FD (Functional Dependencies) or IC (Integrity Constraints) and various validation tasks can be solved by more than one schema language.

XML is often treated as a neutral format from which to generate other outputs: HTML, JSON, other flavours of XML. In some cases, it can make sense to use it as a means to auto-generate other XML- based assets which themselves act on XML inputs, such as XSLT or Schematron schemas.

This paper will present a study of how XML and related technologies helped a publisher to streamline the production process for one of its products, enabling better consistency of data capture and an enhanced customer experience. It will describe how a legacy document-centric format was refined to allow publication processes to run more smoothly, and how an abstraction of the capturing format allowed other key assets in the workflow to be generated automatically, reducing development costs and delivering ahead of schedule.

DemoJam (bring out your demos)

Followed by Social Dinner

We have created an application that enables searching the UK statute book with reference to the structure of legislative documents. Users can target individual legislative provisions and receive direct links to the matching document components. For example, users can search for chapters that contain certain phrases or for sections that have certain headings. In this paper, we describe the XML format used to represent UK legislation and the technologies used to fulfill the queries. We have developed a simple, domain-specific query language to express user requests, and we use MarkLogic to store and index the documents. We parse user requests in XQuery and translate them into native MarkLogic query directives, but because some of the searches we support cannot fully be expressed as MarkLogic queries, we sometimes construct over-inclusive queries and filter the results at the application level. The application is currently in beta, and a working demonstration will be provided.

XProc proves to be a very efficient language when it comes to apply complex chains of operations on sequences of XML documents. In this paper we will evaluate the interoperability of XProc pipelines, i.e. the possibility to migrate a complex pipeline system developed for one XProc processor to another. We take interoperability in this sense to be an indicator for the maturity of XProc and its usability, which in turn is relevant for technology decision makers, pipeline authors and users and the XProc community as a whole.

Although XProc as a W3C recommendation is the result of a complex process of discussions and testing, the question of interoperability in our sense might come up, because for a long time there was only one XProc implementation publicly available and actively maintained: XMLCalabash developed by Norman Walsh, who is also chair of W3C’s XProc working group. Therefore authors did not even have the chance to test their pipelines in different environments. This does also hold for transpect, a framework for checking and converting XML documents and XML-based data, such as .docx, IDML, and EPUB, developed by le-tex, a premedia services and software development company based in Leipzig, Germany. transpect is based on open standards, in particular XProc 1.0, XSLT 2.0, Relax NG, and Schematron. It has been adopted by many publishers and standardization bodies and is considered as the largest XProc application worldwide.

Our real world scenario to assess interoperability of XProc pipelines is the migration of transpect to MorganaXProc, a new XProc processor developed by <xml-project/>. As it does also conform to the W3C test suite, there is a good chance to find some answers to the question of real world interoperability of XProc pipelines.

Exploring XSLT 3.0

This paper details the numismatic thesaurus, Nomisma.org, and its associated front-end and back-end features. The application's architecture is grounded in XML technologies and SPARQL, with XForms underpinning the creation, editing, and publication of RDF. Its public user interface is driven by the XML Pipeline Language in Orbeon, enabling transformation of RDF/XML and SPARQL XML responses into a wide array of alternative serializations, driving geographic visualizations and quantitative analyses in other digital numismatic projects.

STANLEY Black and Decker Innovations have a requirement to produce and maintain DocBook-based documentation, which is translated into up to 10 languages. Documents are built by transclusion from several source files, some of which may be stored remotely. Each document may contain SVG illustrations which also needed translation.

We selected XLIFF as a translation file format. To keep maintenance effort to a minimum, we needed tools that enabled dynamic translations, i.e. translations made at publication time without skeleton files. We also needed tools that could find the correct translations for each translatable element after all the source files (including remote source files) had been transcluded into a single document.

This paper describes the solutions we developed. These included markup (mostly ITS 2.0) in the source documentation, linked data (using RDF/XML) to identify the translation resources, and a set of XSLT stylesheets to handle the transformations.

Invisible XML, ixml for short, is a generic technique for treating any parsable format as if it were XML, and thus allowing any parsable object to be injected into an XML pipeline. Based on the observation that XML can just be seen as the description of a parse-tree, any document can be parsed, and then serialised as XML. The parsing can also be undone, thus allowing roundtripping.

This paper discusses issues around grammar design, and in particular parsing algorithms used to recognise any document, and converting the resultant parse-tree into XML, and gives a new perspective on a classic algorithm.

Giving thanks and closing of the conference.