eriksmartt.com>Selected Archives

Comments on getting to the mobile web...

Russell has a nice post today titled "Getting to the Mobile Web" where he discusses some of the challenges people face, and the bad decisions they make when building mobile websites (or more accurately, mobile versions of existing websites.) If you're interested in this kind of thing, it's a good read; However, I think there's a bit more to it.

No doubt everyone in mobile has heard the "WAP is crap" phase, but it's worth mentioning because I think it really did hurt the mobile web initiative. I haven't built WAP sites in five or six years, but back then, WAP was a pain to work with. It wasn't just a matter of tweaking a few templates to change page layout -- to do it well, you really had to rework the information architecture and navigation style. That's an expensive proposition for very little measurable gain, and it meant that keeping both sites up-to-date was a challenge (the parallel development model Russell mentions.)

But at the time, parallel development was really the only way that made sense. All the great things you can do with CSS and XHTML today weren't available back then, so layout was created with nested tables and browser hacks. That meant there was no way for your desktop browser layout code to be reusable for your mobile site. And it took enough effort just to get IE and Netscape to agree on a layout that reworking the whole thing for a dozen different mobile browsers was just ridiculous.

Browser issues aside, you then had a challenge finding an audience to support all this extra development. Americans are behind the curve in having the latest mobile technologies available to them, but the one thing they do have is relatively inexpensive net access. But it still costs extra and not everyone pays for it. And even if you had net access back then, it was slow and viewed through small, low-res, non-color screens.

Jump forward to today and web development is a whole different ball game. All that time spent discovering and conquering subtle browser glitches is no longer useful. Today it's all about standard XHTML and CSS, and the modern browsers support a separation of content from presentation that typically make layout degrade better on less capable browsers.

Back in the day if you were building IE-only web sites, I would have suggested that you think about a career change. But today it's even more so. It's SO easy to make nice, cross-platform sites that companies shouldn't even be thinking about targeting specific rendering engines. The browsers have caught up with the vision and requirements of web developers such that browser-specific layouts should be a relic of the past. Mobile will do the same, and when it happens, the mobile web experience will follow.

That means my perspective on the whole picture is different then Russell's. I'm in the camp that thinks the browser should be smarter rather then forcing developers to do special tricks for each device. Hacking-in support for every joe-web-browser out there is a losing proposition. We've seen the light using Opera on Series 60 and there's no going back. It's possible for a mobile device to handle traditional web sites and I think the mobile web will be stuck in a funk until mass market devices start shipping with capable web browsers.