Feeds and Other Blogs
Writely Blog Archives
9.22.2005  - Permalink
Found some interesting posts today. This over on OK/Cancel there is a very good discussion of whether and why to build Ajax apps. A quote I like:
What's the difference between Oddpost and Gmail? One followed desktop interaction conventions, required a particular browser and a particular operating system, and gained a cult following. The other came along four years later, followed Web interaction conventions, worked across all modern browsers -- and transformed its entire category. Some might argue that Oddpost is the more sophisticated solution. But Gmail is part of the larger Web in a way that Oddpost never could have been.
The rest of the discussion here is about whether or not Ajax will take off. I don't think users care, they care about the solution...and nice, light, pretty, no-install apps in the browser are going to win this fight. And Ajax is really the only technology that can hit all those points (Java is a download, and usually a JVM version hassle, Flash is pretty but kind of sluggish, and lives more or less in a walled garden on the page unless you work very hard, etc).
Another interesting thing popped up, Ajaxian covered Bindows. Personally, I think this is a very nice toolkit, but exactly the wrong direction. I think it's wrong to just try to replicate the desktop experience - this is the exact mistake many early windows apps made. Remember DOS apps ported into Windows? And then being eaten by folks who get it. That's this, I think. One of the reasons the web is so nice is that the page UI is simple...a few things at a time, a very easy metaphor, etc. It passes the "mom" test - I can usually just tell my mom to go to a site, and she usually can figure it out. I can't remember the last time I could do that with a desktop app. So, even though the windows desktop is "richer", it's not neccesarily better.
There was a discussion long ago about how sometimes "worse" is actually better, if it's simpler, has broader reach, etc. I think the two above stories are examples of exactly this.
I think you're right: ajax apps will win the day. In fact, this is the hardest time right now for any ajax-based service - some people just don't "get it" yet, and you have to find plenty of customers who do. At some point, ajax will go mainstream, but I don't think that's happened yet.
I think that people are finding there way, and the like of Gmail are showing them that just because you can take your "rich" app and put it on the web looking the same, doesn't mean you should.
There will be a lot of abuse before we get good :)
I was at a Silicon Valley party last night where lots of folks were showing off their new shiny "Web 2.0" apps...which looked a lot like Web 1.0 apps with some dynamic menus and such. Not interesting. I also saw quite a few exact copies, pixel per pixel, of desktop apps. Also not what users want.
He says: "On several key issues, Google's engineers have decreed that familiar email practices are no longer useful, and have substituted approaches they prefer..." Where he takes that (read the column) is quite opposite of what I believe is the difference between these two approaches to webmail:
Gmail is an experiment in new UI paradigms, the new Y! mail is a webified copy of current standards. Google's approach is risky, Yahoo!'s is safe. In risk - even if their approach is not correct - Google is doing more to move the state of the art forward than Yahoo! is. (Unless you consider the "state of the art" what you can do with DOM/JS.)
The reason we're stuck with the desktop interfaces we have today is the fear of radical experimentation and the safety in copying what was done before.
I think there's some truth to what WM is saying - you do have to listen carefully and respect what users really need, and give as much choice as you can. But I also think it's worth experimenting, and this is a great place to do it.
FWIW, I use gMail all the time, and I love it. labels are just enough organization for me, and being able to just dump stuff in the archive for further search is really useful. It might be nice to get rid of ads, but I don't really notice them.
I note - there are some facts wrong in the review - you can select and delete multiple emails, you can open multiple emails when composing (shift key) for example, in gmail.
I think there's probably room for both styles. I was talking about folks who blindly copy the desktop...I don't see the point. Why not make it better, if you can, while you have the chance? Clearly Google and Yahoo both agree with this.
What the hell is oddpost?
In a similar vein, the influence of Greasemonkey and similar user-driven scripts is also changing the definition of work by placing more control in the hands of consistent users of a web-service.