Beyond Ajax: Software Development, Two Years from Now
by Volker Weber
Esther has been working on this article (or, rather, a set of five articles) for the last month. She took a very different angle on "what's coming next"; instead of asking some pundit "where do you think the Web is headed?" she asked the guys who build the tools you'll use to create the next round of Web apps: ScottGu, DavidI, Brian Goldfarb, Tim Bray etc. And then she asked the guys in charge of IE and Firefox how those expectations affect the evolution of the Web browser.
Also on Slashdot >
Comments
Because it fits thematically: Joel Spolsky: Strategy Letter VI. Be warned - it's a rather long read, but worthwhile (and funny) all the way.
Interestingly, all of the people she talked to have something to sell.
My guess is that the tools we'll use two years from now will be open source, not commercial, and are likely to be built by someone nobody even knows yet.
Nope, Stefan -- that's not so. Not everyone on that list has something to sell. The Dojo Toolkit is open source. Eclipse RAP is open. Lazslo has OpenLaszlo as well as a commercial version, so you're half-right on that one.
But while you may use solely open source development tools, that isn't common today and there's no real indication that it'll be common two years from now. Yes, open source has great stuff and I'm firmly a believer (in fact, poking at CIO.com you'll see I've written a fair amount about it, such as a guide for enterprises who have open source committers on the payroll). But most developers today use Visual Studio, or some other proprietary environment -- at least as their primary IDE.
The reason I approached these particular people was because (a) they, individually and in corporate terms, have a history of making tools that people use and (b) they're building the software tools that other people are most likely to use. People do follow the path of least resistance, in building software as in anything else. If there's a wizard to make it easy for you to do such-and-so, by golly you're going to see people doing a lot of such-and-so. If you have to do something from scratch, you'll do it only when you must. So: what will the next round of development tools make it easy to do? That's going to depend on what people like ScottGu and Tim Bray and DavidI think you're going to want to do.
Sure -- someone may come along and do something entirely unexpected. But I can't interview her because she hasn't come along yet.
Open source is great for what its great for -- but thinking that it will be the core of the industry in "n" years is foolish. It will be part of the core, as it is now.
I think the only thing holding the next major wave of technology change back now is interface size. Anything too small to have a keyboard and easily read screen is limited. The right next paradigm will come along and change that. Maybe its about heads up optics that write to your eyes from the corner of your glasses (I doubt it for now) and maybe there's some kind of in-air hand waving magic that will replace keyboards. If I knew, I'd be a wealthy man.
Still, it will be something.
I agree - Ajax is yesterday's news. Look at Adobe Flex and MS Silverlight to see where interface design is going. Flex is maturing rapidly and MS... we all know the third release of Silverlight will be great. Sadly, I see IBM still trying to catch up with Ajax instead of looking at the new RIA technologies. They were an early adopter of Lazslo, but didn't pursue it and are now behind the curve.
I'm afraid that all but the most ambitious Domino developers will continue to be years behind the competition in UI design. As any developer knows, the UI "is" the application as far as the end user is concerned. The upcoming Ajax tools will be a welcome step forward, but are about two years late.
