After a frenzy period between 2000 and 2008 where I have spoken at an impressive number of conferences, I temporally retired and hadn’t been at a conference since XTech 2008.
For me, XML Prague 2011 was the first opportunity to meet again face to face with the community of XML core techies and I was curious to find out what the evolution had been during the past three years.
Aside from all the technical food for thought, an image of the conference that I won’t forget is Murata Makoto expressing his grief for the victims of the earthquake in Japan with simple and sober terms.
The tag line of XML Prague 2011 was “XML as new lingua franca for the Web. Why did it never happen?”.
The actual content of the conference has been close to this tag line but was better summarized by Michael Sperberg-McQueen during his closing keynotes: “Let’s put XML in the browser, whether they want it there or not!”
The tone was given by Norman Walsh during the very first session: the convergence between HTML and XML will not happen.
XML has been trying hard to be an application neutral format for the web that could be used both for documents and data. It is fair to say that it has failed to reach this goal and that the preferred formats on the web are HTML for documents and JSON for data.
That doesn’t seem to bother that much the XML Prague attendees who are markup language addicts anyway: if the “mass of web developers” do not care about XML that’s their problem. The benefits of using XML is well known and that just means that we have to develop the XML tools we need on the server as well as on the browser.
Following this line, many sessions were about developing XML support on the browser and bridging the gaps between XML and HTML/JSON:
- Client-side XML Schema validation by Henry S. Thompson and Aleksejs Goremikins
- JSON for XForms by Alain Couthures
- XSLT on the browser by Michael Kay
- Efficient XML Processing in Browsers by Alex Milowski
- XQuery in the Browser reloaded by Peter Fischer
By contrast, server side tools have been less represented, maybe because the domain had been better covered in the past:
- A JSON Facade on MarkLogic Server by Jason Hunter
- CXAN: a case-study for Servlex, an XML web framework by Florent Georges
- Akara – Spicy Bean Fritters and XML Data Services by Uche Ogbuji
Of course, standard updates were also on the program:
- HTML+XML: The W3C HTML/XML Task Force (already mentioned) by Norman Walsh
- Standards update: XSLT 3.0 by Michael Kay
- Standards update: XML, XQuery, XML Processing Profiles, DSDL by Liam Quin, Henry S. Thompson, Jirka Kosek
We also had talks about XML applications:
- Configuring Network Devices with NETCONF and YANG by Ladislav Lhotka
- Advanced XML development – XML Projects by George Bina
- EPUB3: Global Language and Comic by Murata Makoto
- EPUB: Chapter and Verse by Tony Graham
- DITA NG – A Relax NG implementation of DITA by George Bina
Without forgetting a couple of implementation considerations:
- Translating SPARQL and SQL to XQuery by Martin Kaufmann
- Declarative XQuery Rewrites for Profit or Pleasure by John Snelson
And the traditional and always impressive closing keynote by Michael Sperberg-McQueen.
My own presentation, “XQuery injection”, was quite atypical and it took all the talent of Michael Sperberg-McQueen to kindly relate it to “XML on the web” by noticing that security would have to be taken more seriously to make it happen.
One of the things that had impressed me during XTech conferences was the shift in presentation styles, most speakers moving away from heavy bullet points stuffed traditional powerpoint presentations to lighter and better illustrated shows.
I had expected the move to continue and have been surprised to see that the movement doesn’t seem to have caught XML Prague presenters whom continued to do with traditional bullet points with only a couple of exceptions (John Snelson being a notable exception).
I had worked my presentation to use what I thought would be a common style. Using Slidy, I had created no less than 35 short pages to present in 25 minutes. Each page had a different high resolution picture as a background and contained only a few words.
The comments have been generally good even though some pictures chosen to represent injections seem to have hurt the feelings of some attendees.
Since my presentation is just standard HTML, I had been brave enough to use the shared computer. Unfortunately, the presentation loads 74 Megs of background pictures and that was a little bit high for the shared computer that took several seconds to change pages (note to self: next time, use your own laptop)!
Another interesting feature of this conference was the “twitter wall” that was projected in the room using a second video projector.
This wall has proven to be very handy to communicate during the sessions and it can be seen like a more modern incarnation of the IRC channels used in earlier conferences.
Unfortunately, twitter doesn’t allow to search in archives and while I am writing these words, I can no longer go back in the past and read the tweets of the first day of the conference.
Looking backward at the conference, I have mixed feelings about this gap that now seems to be widely accepted on both sides between the XML and the web developers communities.
The dream that XML could be accepted by the web community at large was a nice vision and we should not forget that XML has been designed to be “SGML on the web“.
Web developers have always been reluctant to accept the perceived additional complexity of XHTML and the gap has been there from the beginning and after XML missed the train of Web 2.0 it was too late to close it.
XML on the web will stay a niche and will be used by a minority but the creativity and dynamism of the community shown at Prague is inspiring and encouraging: there is still room for a lot of innovation and XML is more than ever the technology of choice to power web applications.