Conversations with the presenters

Due to the relatively small size of the conference (~400 people), and the friendliness of the various speakers, I was able to have 1-on-1 chats with several of the folks I was most interested in talking to.

Molly Holzschlag spoke about “Why Web Standards Aren’t”, and how the original vision of the Web Standards Project (“anyone, anywhere, any platform, any user agent”) has run aground. She argues that what we should be aiming for isn’t this kind of universal standard, but interoperability, and that what we are really talking about are Best Practices, not Standards. I spoke with her about IE’s non-cooperation with such practices, and Microsoft’s relationship to the web dev community. Molly has already discussed the politics of these issues on her blog (how sweet is her URL: molly.com!). I asked her whether she thought that we as an industry have the opportunity/responsibility to educate users why they are better off using a standards-based browser, rather than pleading with Microsoft (in vain) to get their products up to date with their competitors in this area. The most tedious aspect of web development is browser compatibility testing. My experience is that catering to IE6 takes up hours and hours on every project, not to mention that it bloats your CSS and markup. She didn’t have a clear answer (nobody does), but after chatting for a while she suggested that a set of real tests to demonstrate the loss of time (and consequently, profit) caused by Internet Explorer’s lack of standards support would go a long way toward convincing institutions and companies to make the switch. After all, I’d be willing to bet that the greater proportion of those 1/2 billion IE installs are on corporate or institutional intranets.

Liz Danzico, of Happy Cog and A Brief Message, gave an interesting talk about “The Framework Age”, and how we can help users to ‘perform’ the interfaces we build by helping them detect patterns and organize them into frameworks for interaction and improvisation. I actually thought I was going to a talk on development frameworks (should have read the programme more carefully!), but it turned out to be quite an interesting discussion anyway. Liz definitely has the experience to back up her arguments that we need to focus more on use, interpretation, and evolution of interfaces rather than the traditional paradigm which puts all the emphasis on their (initial) design and composition. During her presentation, Liz used a few examples of the soon-to-be-launched redesign of the WordPress Admin interface, which she worked alongside Matt Mullenweg and Automattic to develop. After the failed Shuttle Project, Happy Cog stepped in to assist with the redesign. I liked the look of the redesign which Liz briefly displayed, but as an avid WordPress user and open source developer, I was curious to what extent the WP and open source community was involved with the redesign.

I asked her about this during the after party. She explained that the community had been involved to some extent throughout the process, but that it had been very tricky to balance the needs of all parties. She related that there had been a serious backlash when Matt delivered a prototype of the redesign in a nightly build without explaining the changes. WordPress 2.5 (scheduled to be released in the coming weeks/months) will have the new design embedded, and will be included with both the hosted and non-hosted versions of WordPress. I discussed the surrounding issues with another developer I met (François from Catalyst), and I am as yet unresolved on the question which the WP situation raises for me: can open-source communities (in their current form primarily made up of programmers) successfully deliver high-quality, progressive design solutions to match the quality of their code? François argued they could, given the right tools, and he may be right. I believe strongly in the open source philosophy, but I wonder if the same structures can support good design. Is design (and here I am more concerned with simple aesthetics than the tangled question of usability) an intrinsically different beast than code? Would open source designs suffer from the dreaded flaws of design-by-committee? I need to develop these ideas more and get others’ input to work through the connected concerns.

I also spoke briefly with Craig Nevill-Manning, director of Google New York, and Chris DiBona, Open Source Program Manager at Google. Craig gave a fascinating talk on “Why scale makes things interesting and useful”, about the opportunities offered by the staggering scale of data which Google has access to. For example, the Google spellchecker which notices your mistyped queries and asks “Did you mean: _____?”, is incredibly accurate because it relies not on the Oxford English Dictionary or some other top-down, centralized source, but on the 100,000 other people who have misspelt or mistyped the same query into that omnipresent search box. And thanks to this mass of data, it also knows that different misspellings indicate a search for different content (see photo above).

Chris DiBona compressed his existing “History of Open Source” talk from its 3 hours down to a speedy 46 minutes, covering the beginnings of UNIX, GNU, Linux, and everything else along the way right up to today’s open-source web frameworks and tools. It was interesting to hear Chris’ take, as he has been involved with these projects for the entirety of his professional (and amateur) career. After his presentation, I asked him about the potential for open source communities to contribute meaningfully to humanitarian projects such as One Laptop Per Child and the social aspects of the Ubuntu project. He said that he believes that the idealism embodied in open source is a natural match for social projects, but that it’s really hard for open source devs alone to make dramatic change. He pointed to Google’s investment in OLPC and other projects as a good combination of big business and open source know-how.

Microsoft Resentment

Despite being one of the primary sponsors (along with Provoke and Google), and having a well designed booth in the lobby (running a 24″ iMac, nonetheless), the resentment toward Microsoft amongst the presenters and attendees was tangible and vocal. I can’t count the number of speakers who made insults toward or bitter jokes about MS during their presentations. It’s true that they’re an easy target, but the anger at the company’s lack of regard for an open dialogue with the people using their products to develop and deliver web content was evident throughout the conference.

Sigurd Magnusson, an organizer of Webstock and Chief Marketing Officer of SilverStripe (a Wellington-based company building an impressive open-source CMS and accompanying web dev framework called Sapphire), indicated the frustration at this lack of dialogue in his recent blog post, The Internet Explorer 8 Doctype Smackdown. A List Apart also recently released a discussion of the issue, criticizing Microsoft for their heavy-handed approach to browser standards: The Shoot Browsers, Don’t They?. An excerpt:

The proposed default behavior for version targeting in Internet Explorer solves the problem of “breaking the web” in much the same way that decapitation solves the problem of headaches. In its current state, version targeting is a cure that will kill the patient. Version targeting could have been an opportunity for Microsoft to demonstrate innovation. Instead, the proposed default behavior demonstrates a fundamental misunderstanding of the World Wide Web, a place that according to its creator, Sir Tim Berners-Lee, will always be “a little bit broken.”

The rest

The conference was absolutely packed with activity and information. A high-resolution experience if there ever was one! Simon Willison gave great presentations on OpenID and Django, Tom Coates shared his insights on ‘designing for a web of data’ (as opposed to a web of pages), Damian Conway had the audience in stitches during his sarcastic “26 Ways to keep the Morlocks from using your website” (hint: use tiny type, inconsistent navigation, bright clashing colours, and lots of JavaScript and Flash), and Kathy Sierra gave a really useful talk on creating passionate users (I’m very glad she’s back doing speaking engagements after her ordeal last year). And of course, I met and had interesting conversations with dozens of other attendees, and got to know a good number of people in the Wellington (and NZ) web industry, including the great guys at SilverStripe. I’m skipping over so much, but have exhausted the limits of my tiny brain to regurgitate everything it was fed. In short, a fantastic experience, and one I will repeat at every opportunity!

My responsibility was to work alongside Matt Bouchard with the design team, consisting of Milena Radzikowska, Piotr Michura and Stan Ruecker to design some JavaScript screen effects and animations to realize the responsive interface they were imagining for the Workbench. An overview and links to this work can be found on the Monk Wiki, and in the Monk client area. I also contributed the (unsolicited) logo for the event, choosing one of the many memorable catchphrases as our slogan (though “It’s more than just a red box” was a close runner-up)!

The event was a lot of fun. Thanks to the culinary wizardry of Maryanne Wynne, we ate like kings (and queens) at every meal (and several times in between). We were also well looked-after by Gary and the rest of the staff at Kramer Pond Lodge. Highlights for me included my introduction to extreme/pair programming at the suggestion of Stéfan and the steak-tearing jaws of Matt Bouchard, the hot tubs (at their varying states of ‘hotness’), and playing baseball and ultimate on the lawn. Oh, and the work was cool!

I’ve posted photos from the event on my Flickr page, which you can view as a set or slideshow. Here’s to next time!