But beneath the hood? Within a few minutes of reading Taby’s post I read this one by Farhad Manjoo, which explains how Google implemented automatic updating of Chrome in the background. No “Do you want to install this update?” dialogue boxes. It just works. Manjoo writes,
Chrome's painless update process is the product of a lot of sophisticated software engineering. Rather than reinstalling the whole program, Chrome's designers have figured out a way to update the ones and zeroes in the program on your computer just where it has changed. This approach dramatically reduces the size of the files they send to your computer during an update. "It is an anathema to us to push out a whole new 10MB update to give you a ten-line security fix," Stephen Adams, a Chrome software engineer, wrote in 2009. Instead, the update system compares the version of Chrome you have on your computer to the new one sitting on Google's server, and then it sends you only the key differences in code. A 10-megabyte update can be reduced to a tiny, 78K download. In a recent post praising Chrome's update system, Coding Horror's Jeff Atwood argued that Chrome had "transcended" the very notion of "software versioning." Chrome updates so quickly that its version number might as well be infinity.
This is why I have increasingly been trying to leverage Google’s engineering without having to encounter Google’s butt-ugly user interfaces. And Google has done a lot of work over the past few years to make it possible for me to do this, especially by introducing a (wonky but usable) version of IMAP for Gmail and by employing Microsoft’s Exchange technology with its calendars. I can now use Google without ever having to look at Google. Best of both worlds. At least until I can escape Google's clutches altogether. . . .