3

I often notice that when I use the SE cross-site search engine (which is powered by Google), I get just a small subset of the results obtained by searching Google with the operator: site:stackexchange.com.

In many cases the additional results found by Google are meaningful appearances of the search term, which are not duplicates of the appearances found by the SE engine.
Here are a few examples for the number of search results obtained by each method, at the time of writing of this question:

╔══════════════════╦════╦════════╗
║   Search term    ║ SE ║ Google ║
╠══════════════════╬════╬════════╣
║ football manager ║  6 ║   1120 ║
║ roomba           ║  3 ║   1490 ║
║ papercut         ║  8 ║    180 ║
║ treehouse        ║  2 ║    552 ║
╚══════════════════╩════╩════════╝

Is there a good reason for this discrepancy?
Is this a bug?

Improve this question
3
  • Well, one possible reason is that the custom site search we build (with Google) excludes per site metas. It is a list of just the main site urls as well, nothing else, no blogs, etc. So site:stackexchange.com is pretty broad considering all the stuff that is under that domain.
    – Jeff Atwood
    Jul 15, 2012 at 16:51
  • That's a possible reason, but it doesn't account for all the missing results. Here is one example of a question about a treehouse, which is a question on a main site and not a site meta, but which doesn't appear in the search results of the SE engine. A quick scan of the search results for the terms in the above table will reveal many such examples.
    – Joe
    Jul 15, 2012 at 19:55
  • Sure, that's fair, just pointing out some obvious differences.
    – Jeff Atwood
    Jul 16, 2012 at 2:27

0

Reset to default

You must log in to answer this question.

Browse other questions tagged .