On Jun 22, 2010, at 6:33 PM, Roman Chyla wrote:

> I may speak about Lucene, because it was used - but in fact it is not
> exactly Lucene-ity vs DB-ity. It is more IR-ity vs SQL-ity. SQL-ity
> may rule out IR-ity if you believe that SQL can handle all the cases
> and that would be pitty for INSPIRE - it can already be shown that
> IR-ity does very well for certain cases where DB faired poorly (at
> least according to some test that were reportedly done in the past)
> 
This is fine, and I'm not casting aspersions at Lucene, merely laying out some 
functionality that we can't live without.  I believe such uses are fundamental 
pieces of IR, so it shouldn't be surprising that IR tools can do this, however, 
we have to actually make something SQL or something IR do it in finite 
time...(both developer time and wallclock time...)


> 
> Tibor's use case is made for a db query. Lucene is good for some
> tasks, Mysql is better in others, if I somehow suggested that
> everything should be handled by Lucene, it must have been some error
> on my side, I apologise. Never meant anything similar.

No no, I didn't interpret it that way, I may have been unclear.   I just want 
to make sure do eventually build things that provide what Tibor described.   
But additionally I didn't want to skirt around Tibor's example by saying that 
it might overload the DB.   I really would like to know whether that use could 
be delivered in Lucene in acceptable speed, because it seems like its a nice, 
but practical and relevant, test of what an IR tool might be doing in our 
context.   I'd still like to know whether this case is better with Lucene or 
MySQL, as that helps me understand where the boundaries should naturally live.

Best,
Travis

> 
> Cheers,
> 
>  roman
> 
>> 
>> Best
>> Travis
>> 
>> 
>>> 
>>> Best,
>>> 
>>> roman
>>> 
>>>> accordingly (e.g. CERN would come before University of Geneva).  This
>>>> example is what Marko was working on via KBDs.
>>>> 
>>>> Best regards
>>>> --
>>>> Tibor Simko
>>>> 
>> 
>> Travis C. Brooks
>> Manager of Information Systems & SPIRES/INSPIRE
>> SLAC National Accelerator Laboratory Library
>> http://www.slac.stanford.edu/spires/
>> 
>> 
>> 
>> 
>> 

Travis C. Brooks
Manager of Information Systems & SPIRES/INSPIRE
SLAC National Accelerator Laboratory Library
http://www.slac.stanford.edu/spires/




Reply via email to