Anyway, I'm sorry if some people are reading this hastily written blog entry to seem like the archive is taking credit for the process of discovery being done by people. The phrase likely does not mean definitely and perhaps I should have used a different word when I wrote the entry. But the fact remains that the wayback machine is the only place you can see the image in the context of its original website, and that is only happening because the archive is doing such a general crawl. That's all I wanted to get across, all hail the wayback machine, have a great day.

It wasn't needed this time but it would be great to have a fulltext (and binary!) and reverse image search of the internet archive. Often you know something about what you are searchng fore but have no clue about the location.

E.g. I'm interested of finding old installers of a particular type (identifyable by certain byte sequences) but they could have been hosted anywhere.

I guess this is not feasible with the current resources of the IA and if it did happen it would probably just result in them getting hammered with DMCA requests and other legal demands for content that the "owners" didn't even know was on the archive.