X-tech conference 2005 - day 1
I attended the X-tech conference this week here in Amsterdam. To quote the
website: "XTech 2005 is the premier European conference for developers and
managers working with XML and Web technologies, bringing together the worlds
of web development, open source, semantic web and open standards." This is a
summary of presentations I saw on the first day.
The X-tech conference 2005 was held in Amsterdam this year which is only an hours travel from Delft, where I currently live. X-tech broadend this year to include a more wider array of topics including browser technologies.
Keynote speaches
The first day started of with two keynote speaches one by Paula le Dieu, Director of Creative Commons International and the other by Mike Shaver, Project Coordinator, Mozilla Foundation.
The first talk was mostly about Paulas work for the BBC where she set up the Creative Archive. The archive plans to provide audio and video from the huge BBC archives but also from others like the open university and channel 4. Audio and video can all be used under a creative commons license which excludes commercial use but also excludes anyone from outside of the UK from using the archive which is rather a shame.
Mike Shaver talked about the evolution of the internet and how he expects there won’t be a big bang which turns everything upside down but a multitude of little bangs. Examples of this are AJAX. A simple combination of existing technologies ie xhtml,css,javascript, xmlHttpRequest combined to make richer web applications. Examples of this are Google suggest, gmail and Google maps.
Mike also showed a demo of the canvas element which is currently used in the Apple Dashboard and Safari and will be shipping in the upcoming release of Firefox 1.1. Canvas is a html element which can be used to programmaticaly render bitmap grahics in the browser. A popular example to rever to seems to be the apple clock shown on the right on which the hands are rendered using script.
Presentations
After the two keynotes and a coffee break the first of many presentations began. I’ll summarize them below:
Comparing XSLT and XQuery, Michael Kay
Michael Kay the main developer behind the Saxon XSLT and Xquery processor gave a very clear presentation of the advantages and disadvantages of both languages and the situations when you would choose either the one or the other. The simple conclusion is if you have large amounts of well structured data it’s best to use Xquery if your doing stuff with documents use XSLT.
XUL - Mozilla’s XML User Interface Language, Ben Goodger
Ben Goodger lead engineer for Mozilla Firefox showed of XUL. XUL (pronounced zuul) is an XML language used to markup user interfaces. Currently both the GUIs for Firefox and Thunderbird are completly written in XUL. CSS is used for markup and you can use scripting.
In both Firefox and thunderbird the GUI can easily be customized using extensions.
XML and Relational Storage - Best Practices Guide, George Lapis
Unfortunatly George Lapis isn’t a very good speaker and that’s a shame because the information in this presentation was excellent. George talked about the different types of XML databases (native xml, whole document and shredded) currently in use and the pros and cons of each.
All XML Databases are Equal, John Snelson
Unfortunatly this second talk in the Core Technologies track had a little to much overlap with the one from George Lapis and I didn’t find it all that interesting.
Browsing on small screens, Hakon Lie
This was a very interesting presentation by Hakon Lie CTS of Opera Software. Opera is a very popular browser on mobile phones (Nokia for instance) besides the normal desktop browser we know. Both use the same rendering engine and he showed how it reacts to varying screen sizes. Depending on the resolution of the screen the browser tries to fit as many columns into it when the width of a column becomes unreadable small the engine modifies the layout (instead of side-by-side elements are placed under eachother) and ressizes the images to fit. On the demo he showed us this looked really cool.
A second important part of his talk was about interoperability. He showed us examples of how to use css to apply different styles to varying screen sizes. He also made a point about webbrowsers (mainly focused on MSIE) not fully supporting current CSS standards which is a must for interoperability to work correctly. He also mentioned the ACID2 test which can be used to test css support in current browsers (currently no browsers passes the ACID2 test by the way)
RSS Syndication For A Worldwide Audience
The last speaker of this day was Ian Forrester from the BBC World Service. The BBC World Service currently has RSS feeds in 41 different languages and they’re planning even more. He talked about the problems of using RSS width these different languages. These mainly focus on languages who don’t use the roman alphabet (arabic, hebrew, japanese etc).
The problem at the moment is that no RSS standard (rss 0.91, 1.0, 2.0 or Atom) has any decent support for internationalization (i18n). The BBC World Service chose RSS 1.0 because it has the best support for character encodings but there are still numerous problems concerning languages which are displayed right-to-left. There is currently no real solution to solve these.