site:example.com and Site:example.com show different search results?

[Future]

I was amazed to find this today by giving:
keyword site.example.com
shows all urls for example.com

wereas
keyword Site.example.com
shows all urls for example.com + related websites.

Anyone find the exact difference here ?

[Robert Charlton]

Future - Your post and your title here differ in their syntax, but I'm assuming that you're asking about the Google site operator, followed by a colon, and what happens if you capitalize the S in "site", as in....

keyword site:example.com
keyword Site:example.com

The site: operator is case sensitive, and when the S is capitalized, Google treats the query not as a site: operator query, but rather as a search that includes the word "site" as one of the search terms.

I've never found official Google documentation to confirm that the site operator is case sensitive, incidentally.

I should also add that, on my machine anyway, if I do one of the above searches right after the other, the new Google ajax interface isn't always responding to a case change immediately and I sometimes need to refresh the page after doing a search.

[svepo]

Yes, results were different.
Very interesting observation. This is the first time I'm noticing it!

[phranque]

in the very small sample i ran [Site:example.com] results were similar to [Site example.com] with a small increase in reported results for the latter.
what really surprises me is the difference in reported results between [Site example.com] and [site example.com] in a few of the cases i tested.
i may try some more testing here later to see if i can discern any pattern or explanation.

[Lame_Wolf]

I thought I would do it to my site.
My results :

site:www.example.com : 5680
Site:www.example.com : 1,920,000

Results for images :

site:www.example.com (moderate): 6,720
Site:www.example.com (moderate) : 1,190,000

The good thing is, Site:www.example.com shows other sites that are using your images. Just like the old days. Makes it easier to find people stealing your work.

[Robert Charlton]

in the very small sample i ran [Site:example.com] results were similar to [Site example.com] with a small increase in reported results for the latter.

Theoretically possible, of course, but in the tests I ran, results are generally more like what Lame_Wolf reports. You want to try this with a domain that's likely to be mentioned on the web beyond its own pages. If you try a few domains of different sizes, chances are you'll be seeing the differences.

Also, to assure yourself that there's nothing mystical about [site example.com], try searching [example.com site]. Number of pages returned should be close, but because Google estimates this number, it won't always be exactly the same.