More on how Google works

I guess that my impression that Tesugen.com had fallen out of the main index was perhaps only due to my being connected to a different server with different stuff cached (as in locally cached on that server, as opposed to the cache you can view by clicking the “Cached” link for a search hit). Because now I have “Similar pages” links for hits from my site.

I thought some more about the motivation for a temporary index, and I guess that the point is to act as a filter for pages that only happen to be in the “crawl path” once, but never again. Probably, Google doesn’t want to index those pages permanently, since they probably aren’t of interest to many people (or else the would have been linked by many peope).

If you think about it, Google probably want to keep the index and cache as small as possible and still provide high quality search results. A temporary index asserts that new pages (max. 24 hours old) are likely to be in the search results, but prevents the main index from getting littered with pages that are only updated once and referred to only by few people.

The local caches of the different servers (there’s ten of them according to Netcraft’s What’s That Site Running) confuses the one trying to figure out how Google works from observing it as a black box. (-: