Trends and Transients

Over­view

Each year there are more new tech­no­lo­gies to keep track of, more ways to organ­ise your life and your company’s inform­a­tion, more ways to com­mu­nic­ate. This ses­sion will intro­duce you to new tech­no­lo­gies, dis­cuss older, under-appreciated tech­no­lo­gies, and enter­tain you at the same time. Our expert speak­ers will debate cur­rent issues and tech­no­lo­gies, giv­ing you the bene­fit of their wide exper­i­ence and dif­fer­ing points of view, so you can decide for your­self which tech­no­lo­gies will meet your needs and which are a waste of your time.

Speak­ers for 2012 include Dr. Marc Had­ley, Eric van der Vlist, and Tom Scott, as well as Fac­ulty Board mem­ber Priscilla Walms­ley.

Classes for 2012

Should you write XML Schemas by hand?

Taught by Eric van der Vlist

XML Schema lan­guages are won­der­ful tools to val­id­ate XML doc­u­ments but they are dif­fi­cult to read for most people and they are even more dif­fi­cult to write.

A sig­ni­fic­ant num­ber of my cus­tom­ers are people who know what they want to express fairly well but they do not know XML schema lan­guages well enough to be able to express their data mod­els using any of these lan­guages. They usu­ally come to me with some kind of descrip­tion of their mod­els (doc­u­ment­a­tion, Excel spread­sheets, XML samples, UML mod­els, …) and instead of writ­ing the schemas for them or giv­ing them XML Schema train­ing, we usu­ally choose to gen­er­ate schemas from these descriptions.

In this talk, I’ll give sev­eral examples and describe the bene­fits of using cus­tom high-level descrip­tions to gen­er­ate XML schemas.

JavaS­cript, JSON and “Big Data” Analytics

Taught by Dr. Marc Hadley

From its early days as a means to script Web pages, JavaS­cript has emerged as a “ser­i­ous” pro­gram­ming lan­guage used for a wide vari­ety of server-side tasks. An off­shoot data seri­al­iz­a­tion format, JSON, is now often used instead of XML and recently we’ve seen new lan­guages com­ing to the fore that com­pile down to JavaScript.

In this talk we’ll present a case study of a hybrid XML and JSON based pro­ject that uses a poly­glot pro­gram­ming approach to solve some big data ana­lyt­ics prob­lems. Along the way we’ll explain the map-reduce algorithm, talk about JSON vs XML tradeoffs, and explore some poly­glot pro­gram­ming techniques.

Aca­demic pub­lish­ing — becom­ing digit­ally native

Taught by Tom Scott.

While the Web has cre­ated new busi­ness mod­els and sig­ni­fic­antly increased the num­ber and range of art­icles that can be pub­lished, sci­entific pub­lish­ing has remained largely unchanged. Sci­entific dis­cov­er­ies are still pub­lished in journal art­icles where the art­icle is a review, a piece of metadata if you will, of the sci­ent­ists’ research.

Con­tent might be dis­trib­uted over http but what is dis­trib­uted is still, in essence, a print journal over the Web. Little has changed since 1665 – the primary objects, the things a pub­lisher pub­lishes remain the art­icle, issue and journal.

In this talk I will dis­cuss how refram­ing the prob­lem — look­ing at what sci­entific com­mu­nic­a­tion is try­ing to achieve and what is pos­sible with Web tech­no­lo­gies — presents us with an altern­at­ive approach. I will also present a case study demon­strat­ing some of what is pos­sible when we con­sider aca­demic pub­lish­ing from this perspective.

Canon­ical Mod­el­ing: NIEM and beyond

Taught by Priscilla Walms­ley.

Canon­ical mod­els are stand­ard­ized vocab­u­lar­ies for XML inter­change that are designed by an organ­iz­a­tion or group of organ­iz­a­tions to ease integ­ra­tion and pro­mote reuse and inter­op­er­ab­il­ity. The U.S. National Inform­a­tion Exchange Model (NIEM) provides an example of a canon­ical model: a 6000-element XML vocab­u­lary that is used by U.S. gov­ern­ment entit­ies and their inform­a­tion shar­ing part­ners at the fed­eral, state and local levels.

Design­ing XML vocab­u­lar­ies with mod­u­lar, reusable com­pon­ents is of course noth­ing new. But new tech­niques and tools are emer­ging to allow organ­iz­a­tions to form­al­ize and man­age large canon­ical mod­els expressed in XML.

Using NIEM and other examples, this ses­sion will describe the bene­fits of devel­op­ing canon­ical mod­els and the tech­niques used to deploy them. It will also provide guid­ance on some of the asso­ci­ated chal­lenges, includ­ing allow­ing for cus­tom­iz­a­tions, inter­op­er­ab­il­ity among dif­fer­ing sub­sets, and versioning.