I found an article on Macworld this morning that raised an issue with me as someone who does mobile development. This gist of this article is that a group of “wireless experts” are saying that HTML5 is problematic at the moment with mobile development.

I just finished a paper for AAM this year on using HTML5 for museum tours so obviously I have a stand that web applications are actually more sensible to build for certain applications. I was really curious to see what was being said in this article.

Apparently this has been a debate at Mobile Northwest this year in Seattle. This brief article features quotes from app developers which is arguably biased. Also frustrating is that no examples or specifics are given to back up the opinions being stated. I’m not complaining about the difference of opinion, but I would be very interested in understanding experiences that other developers have had making designing apps a better solution.

“We’ve run into the limits of what HTML5 can and can’t do and it’s painful,” Woodbury said. He expects the standard to improve over time, though.

This statement is sort of accurate. HTML5 is an open standard that isn’t official yet, so yes there are support issues. But I wouldn’t say that it has “reached a limit”. This is a technology that is expanding not stopping. In fact, I’ve found that the HTML5 is actually more adapted to mobile browsers than desktops (thanks in part to the flavors of IE floating around).

Starbucks initiated a project to use HTML5 in hopes of making its Website work well across the many mobile browsers. “It worked but it taught us it was too hard,” said K.C. MacLaren, director of mobile and emerging technologies at Starbucks. Starbucks’ Website is relatively “high functioning,” with a lot of Javascript, making it hard to replicate, he said. It took too much work to transition the site to HTML5 and make it run well on the many browsers that run on mobile phones, he said.

So what was “too hard”?

It sounds like the issues were related to Javascript, not HTML (Javascript is problematic with phone processors, memory and browsers). I’d really be curious as to what is meant my “high functioning”. Starbucks has a wonderful website, but honestly there’s not much that’s unconventional from a technology standpoint. That’s not an insult, just an observation. If you are replicating UI features like the complex navigation my argument here is that this is probably not a good idea. Due to the fairly obvious screen size difference and the behavioral differences of using a mouse vs one’s fingers — replication of a desktop design is going to be a mess. Interpretation is completely different. Pulling up Starbucks’ website on my iPhone was a flat experience. It looks like the main site with features disabled.

Most of the experts think that HTML5 will advance enough to become widely used and when that does happen, it might create some significant changes in the market. Woodbury predicts that in 18 months the shift to HTML5 will boost Web application stores, which will then “take a serious bite out of the OS-driven app stores,” he said.

18 months is not a long time. I think this is an extremely interesting time to be developing in the web space. But what’s really interesting to me is the statement about web application stores. I think this is an interesting idea, but I’m not convinced that the model used by Apple and Android to sell applications will work with web based apps. I’d love to be wrong about this, but I just don’t see it this way. Google has an "app store" for its chrome browser and I have some frustrations here. Most of the apps are just websites. Why do I need to see it as an “app”? And why can’t I just go to the website either directly or from a bookmark? But look at the HTML5 Angry Birds and tell me that’s not impressive…

Interesting conversation though. Would love to have been at the conference to meet and talk to some of these folks.