Dissecting the new whitehouse.gov

There’s a cool new guy in the White House, and naturally, along with the keys to the nuclear arsenal and the unenviable task of cleaning up the garbage from his predecessor, he gets a new website.

The official whitehouse.gov has been given a serious overhaul, and against all reasonable expectations for a government website, it’s actually pretty good.

Under the hood, it’s running ASP.NET 2.0. Not exactly the revolutionary change we were promised, but the markup almost validates, which for a government website in 2009 feels suspiciously close to progress.

There’s a blog. There are RSS feeds. There’s a Twitter feed. And there are almost 2 400 lines in robots.txt, because apparently even the White House has things it would rather Google didn’t poke around in.

Somebody involved in this thing clearly knew what they were doing. Third-party content is Creative Commons licensed. GZIP compression is enabled. They’re actually using jQuery. There are even CSS sprites, using the classic technique popularized by Dave Shea. Somewhere, a frontend developer is quietly weeping with patriotic pride.

The real White House in Washington D.C.

Of course, nothing built with ASP.NET in this era escapes completely unscathed. ViewState is lurking in there, fat and comfortable, like a lobbyist who knows exactly which senator to buy lunch. There’s plenty of unnecessary whitespace, HTML comments scattered around the markup, and a handful of hidden links providing convenient little backlinks for the developers.

So no, it isn’t flawless. Few websites are, and government websites traditionally set the bar somewhere around “loads eventually in Internet Explorer”. But compared with the old whitehouse.gov, this is a considerable improvement. Cleaner, smarter, more open and built by people who appear to have heard of the modern web.

Change has come to Washington. Or at least to its HTML.