I think it is because in 2008, we expect a certain level of uptime for web services. It's a good indicator of how the standards for web applications have increa

I think it is because in 2008, we expect a certain level of uptime for web services. It's a good indicator of how the standards for web applications have increased over the years. I remember about a decade ago when I first signed up for services like Hotmail or eBay, they would be down fairly often and for long periods of time, but it was understood and even accepted, since they were usually free services that did not have much direct competition. "Internet application" usually carried implications of slow, buggy, and unreliable service. However, in the last couple of years, it seems that we have to come to expect a near-constant uptime for the websites we use. It points to a general up trend in application quality and consumer expectation, which I think is healthy for both developers and consumers. In the end, I think Twitter speaks to Ev Williams's brilliance in building an extremely sticky product, since people continue to Twitter despite the downtime and the deluge of blog posts from people criticizing them at every turn. It reminds me of how most people continued to blog on Blogspot even with its notorious bad pre-Google uptime.