Google's Blogger service had performance problems Thursday night, which were noted on TechCrunch and ParisLemon, which noticed odd error codes where blogs should be. Here's the official update from the Blogger Status page:
This afternoon, many blogs showed the error bX-uxu3fu for about two to three hours. This was caused by a bug in our latest release that affected highly-customized Layouts templates. We noticed the problem as soon as the release finished deploying, and worked quickly to find and fix the bug. We apologize to the bloggers and readers of the affected blogs.
Time was when a three-hour outage at Blogger would have the blogosphere in a tizzy. Apparently TechCrunch learned about the outage when it was mentioned on Twitter.
Speaking of which, Andrew Baron and Dave Winer are leading a conversation about ways to make Twitter more reliable so it can be useful as a disaster communciations tool. Larry Dignan asks the key question: "Will people pay $20 a month for an industrial Twitter?" The data center industry exists to deliver scalability and reliability, but it doesn't give them away for free. Services with working business models can afford to invest in infrastructure (or at least convince their venture capital backers to pony up en route to profitability).