Google Reader will now generate feeds for Web pages that do not already have them. You can plug in a URL and it will figure out a way to track changes to the page and notify you of them. From a technology standpoint, I’m fascinated. About 10 years ago, I worked for a company called Alerts.com that tried to do this sort of change detection. This was before feeds really took off, Morbus Iff’s AmphetaDesk was probably the leading news reader at that time.
The company built custom scrapers for any Web site we monitored, and our strategy was to seek out deals with content sites to scrape their sites and generate news alerts for them. Let me be blunt: this was a stupid strategy, and I knew it. RSS was starting to take off, and any company with a real CMS can keep track of the new stuff that’s being published without any help from a third party. Even so, we had a number of large content sites as customers for these keyword-based alerts.
I was hired to work on the consumer-facing site, and my idea was to do the sort of thing Google is launching now — automatically generate feeds and news alerts for any site, not just ones that were our partners. Unfortunately we didn’t really have the resources to pursue that strategy, and after a failed acquisition by LifeMinders, things steadily went downhill until everybody got laid off.
It’s funny to see Google doing now what I thought would have been a good idea a decade ago.