[Web4lib] Google limit of 1,000 results

[Roy Tennant]

And let's not fall into the pitfall of thinking that seeing the "most  
relevant" 1,000 hits that Google returns is all that anyone wants to  
do. That only serves one particular need -- certainly an important  
one, but only one. There are many other needs. Sometimes I need to  
see the hits that logically would show up at the bottom of Google's  
ranking, but I have no option to do so. I can hardly wait until  
Google decides that serving up 1,000 hits is way too much anyway,  
that 100 is more than enough. Then we'll be left with most hits  
having bought their way to the top -- either from direct sponsored  
placement or the kind of under-the-table Google fooling that  
everybody knows goes on all the time. The clearer we are about what  
Google is and is not the better able we will be to advise our  
clienteles.
Roy


On Jul 12, 2005, at 7:35 PM, Richard Wiggins wrote:

> I think this thread is important.
>  I think for serious researchers it's probably more helpful and  
> more useful
> for Lexis-Nexis to punt than for Google to do the best it can at  
> the top of
> the hit list.
>  This involves the myth of the exhaustive search. Suppose you click  
> on every
> link in the hit list from a 1000 item Google search. Have you  
> covered the
> world? No, you have not. You've covered the part of the world that  
> Google's
> algorithms choose to present to you.
>  The hit list is not an exhaustive list of "the" answers.
>  /rich
>
>
>  On 7/12/05, Karen Coyle <kcoyle at kcoyle.net> wrote:
>
>>
>> Another interesting issue, which may not apply to Google but does to
>> some other databases: when the search is cut off at 1,000 documents,
>> WHICH 1,000 documents are in the result set? The most recent by date?
>> Last in-first out? Is order applied after the results are  
>> retrieved? (So
>> it looks like you've got the records in publication date order,  
>> but you
>> don't have all of the documents with the publication dates in the  
>> result
>> set)?
>>
>> kc
>>
>> Jenne Heise wrote:
>>
>>
>>>
>>>
>>>> It's intriguing that Attila the Search Engine makes that cut-off so
>>>> low. I'm
>>>> assuming we're one of the few groups that care about this  
>>>> (though as
>>>> groups
>>>> go, we should be). One wonders how much of the Google world is  
>>>> smoke
>>>> and
>>>> mirrors (or two twelve-year-olds in a garage).
>>>>
>>>>
>>> Does this phrase sound familiar?
>>> "*This search has been interrupted because it will return more than
>>> 1,000 documents."*
>>> Lexis-Nexis.
>>> I'm beginning to see a trend:
>>> - missing data
>>> - wierd search and relevance algorigthms
>>> - database coverage subject to the Schroedinger uncertainty  
>>> principle
>>> - drives librarians nuts
>>> - users become addicted and have to be forcibly changed over to
>>> another database
>>> - costs a lot of money and requires a lot of training... oh, wait,
>>> that's just Lexis/Nexis...
>>>
>>> -- Jenne Heise
>>> _______________________________________________
>>> Web4lib mailing list
>>> Web4lib at webjunction.org
>>> http://lists.webjunction.org/web4lib/
>>>
>>>
>>>
>>
>> --
>> -----------------------------------
>> Karen Coyle / Digital Library Consultant
>> kcoyle at kcoyle.net http://www.kcoyle.net
>> ph.: 510-540-7596
>> fx.: 510-848-3913
>> mo.: 510-435-8234
>> ------------------------------------
>>
>> _______________________________________________
>> Web4lib mailing list
>> Web4lib at webjunction.org
>> http://lists.webjunction.org/web4lib/
>>
>>
> _______________________________________________
> Web4lib mailing list
> Web4lib at webjunction.org
> http://lists.webjunction.org/web4lib/
>
>
>