While I have done my fair share of ranting about the HTML5 rendering deficiencies of Google Android, I must admit that I’ve “gone public” about it because of my interest in trying to make it better. I have no illusion that I’m just one mobile developer, out in the weeds, when millions of Android units are shipping / adopted during this holiday season. However, right now trying to work with HTML5 on Android only makes the iPhone Safari look so much the superior target.
Dianne Hackborn rocked my world
I’ve been involved with HTML5 for a while, but heading into my last major project, I was on a team that was convinced that the mobile-verse was at a point where an HTML5 project could be implemented as a cross-platform approach for an app targeting both the iPhone Safari and the Android Webkit browser. Not that much of a “far out” line of thinking considering this is 2011 and smartphones are quite the pieces of hardware. Our surprise that this wasn’t the case, quickly turned to fear upon the grim discovery that while Android Webkit supported the HTML5 spec, the actual performance of its implementation is a totally different story.
Most of the Android developer community seems to be fully aware of the various performance problems of Android, but it feels like the “slowly simmering” feeling has started to reach a boiling point.
Dianne Hackborn penned a very descriptive G+ post as an attempt to separate the myths and truths behind Android’s rendering issues. However, instead of clearing the air, what ensued in the (highly interesting) comments is the voice of the Android developer community pushing for some real changes to the very architecture of the OS.
While Dianne makes several detailed points, the basic premise of her viewpoint is that “Android as-is is A-OK”. I can understand defending your work because you’ve been so close to it for so long, but I find this attitude is a little bit troubling.
As an aside to this interesting (and extremely technical) post, Dianne is a true heroine. Seriously.
Andrew Munn for the followup
To throw some kerosene into the mix, Andrew Munn, an ex-Google intern who is now working at Microsoft (huh?), publicly disagreed with a few of Dianne’s points and put together his own G+ post on the issue to explain his own view why the rendering performance on Android is laggy based upon his own Google experience.
Perhaps Andrew didn’t realize his post would go so viral within the community.
The original post of his is still up, but the community (along with Dianne herself) has responded with enough clarification to dispute several of Andrew’s explanations. While it was welcome to read his attitude of trying to “acknowledge” some deficiencies with the Android core design, and trying to spur discussion within the Android community, it really can’t be ignored that he’s moving to Microsoft. Probably to work on Windows Phone 7.5 and beyond.
Again, all very interesting and all pointing to realization that the current Android design has enough red flags to cause a lot of concern to its core developer supporters.
The Real Story – The community wants to help but is demanding change
Better and newer hardware equipped with Android 4.0 (aka Ice Cream Sandwich) has only begun to highlight the performance flaws instead of suppressing them. Perhaps some devices in the very near future will actually be amazing. Perhaps more CPU cores and / or memory will sweep under the rug the core architecture problems behind Android OS.
Perhaps not.
THEN there’s the problem with every Carrier who won’t provide OS updates to customers…but that’s another post.
About wazoo
Thank you very much for doing such a great job summing up with “Dianne Hackborn vs Andrew Munn vs Dianne Hackborn” G+ posts!!!
Regardless of the cause of the performance (scrolling stutter/lag) issues on Android, I believe it will take its toll on Android handset sales. Google keep boasting about their “700,000 Android handset activations per day”, but I believe that many of these are activations of cheap Android handsets (sub £100 in the UK) or users activating multiple times (after installing ROMs for example). Companies are already dumping Android tablets because they can’t sell them (http://news.yahoo.com/android-tablets-not-selling-well-claimed-145600449.html).
Even with Google’s marketing muscle, mainstream non-technical consumers are becoming aware of the Google ‘lag’ (stutter/choppiness) issue, and they will (if they’re not already) vote with their wallets.
Hi Andy,
Thanks for stopping by and sharing your thoughts. I agree, I think the numbers we’re constantly seeing in the media are skewed to form the appearance that the planet has gone Android.
While it’s been a simmering issue for a few years (since Honeycomb probably), I’d venture that 2012 will be the year when the Android Fragmentation issue will go nuclear.