Connecting with Koranteng

IBM insider Koranteng Ofosu-Amaah has posted a marvelously detailed account of what's been done with the work that Halfbrain started: On rich web applications, AlphaBlox and Oddpost, in response to my previous posting. He has an intimate knowledge of — and admiration for — what the Halfbrain/AlphaBlox/Oddpost crew achieved, having been assigned the task of porting it to Mozilla last year for IBM's WebSphere Portal product (it appears that IBM licensed the Halfbrain technology from AlphaBlox as the basis for the "Simple Browser Productivity Components" — the spreadsheet, presentation and rich text editors — that now ship with the portal).

He outlines the history of the Halfbrain products and writes in knowledgeable detail about both how it was done then and how it can be done now. Interestingly, he says it is no longer necessary to write to Win/IE:

"Mozilla is now the best platform for doing such development. It has the best standards support, is cross-platform and has the best developer tools: DOM Inspector and Venkman, and perhaps even mindshare. In fact (market share be damned), it makes sense to first write your rich web app for Mozilla. Your code will be cleaner to start off, will be better structured and will most likely work in other browsers."

He then goes on to add some indispensible advice for anyone planning to write rich web applications to the browser, recommending a neat design pattern: "Database <--> XML (Optional) <--> Javascript Object Bindings <--> UI Bindings (HTML) + UI management code" along with a reminder that, "It's the Latency, stupid. When dealing with distributed applications, it's the issue of latency that will determine which applications will rule." The other things that need to be tackled, as he notes in concluding, are offline synchronization and related security considerations — but as Adam Bosworth has indicated [link updated 31 July 2004], these are matters that BEA is already working on in its Project Alchemy.

Koranteng begins his post by observing that Loosely Coupled's "comment entry fields don't show up in Mozilla or Firefox," and thus the only way he was able to post a comment was to fire up Internet Explorer (fortunately, having begun, he continued on his own blog, where he's been able to include much more detail). I haven't yet had a chance to check out the problem, but if true, this state of affairs is somewhat ironic given the tone of recent postings here about browser platforms. The irony increases when you consider that the comment system is itself a Javascript construct, which is hosted on a separate system called Xara Online, and served to these pages as a Javascript include. So the most likely reason why such a fascinating account of client-side scripting development couldn't be posted here was a flaw in the Javascript that delivers the comment system, which presumably is tripping over something in the newer browsers. Like Halfbrain/Blox, the Xara Online service is another survivor from those heady days when browser-based computing was about to take over the world. Maybe it too will have a renaissance. Certainly Koranteng's posting carries a clear message that the browser is continuing to mature — and has the potential to mature — as a rich-client platform at a much faster rate than most people realize.

UPDATE [added July 17]: In the latest update to his musings on my recent posts (scroll down to the end of the link) Olivier Travers has raised the possibility that IBM may decide to buy BEA. This looks quite a credible outcome now. It would not only help to consolidate the app server market but also bring in some important technologies and skills in Adam Bosworth's developer team — not just the Crossgain alumni but also the Westsiders they later acquired (see my February posting, Only Connect). We'll see what transpires.

Separately, an interesting aside from John Dowdell — "it's just striking to see the emphasis put on the actual user experience" — reminded me that it's worth reiterating just why the user experience is moving to center stage. Here's how I put it a couple of years back in a posting called Reaching out to users:

"... while some programmers get on with the glamorous work of building the standards, platforms and infrastructure of web services, others are quietly perfoming the hard, heroic slog of sweating the user interface. Their work will have its reward in the end, because convenience is something people will pay for. It deserves recognition too, because the web services grid is only as powerful as the number of nodes that can productively connect to it. Once the infrastructure is in place, the next great challenge will be to make all that connectivity and automation effortlessly accessible to the great mass of users."

posted by Phil Wainewright 10:22 AM (GMT) | comments | link