[Termtools] Duplicate systems in category CSR

Peter Schneider-Kamp psk at informatik.rwth-aachen.de
Fri Oct 24 09:35:22 CEST 2008


Hi Simon,

I agree that eliminating duplicates is not a big priority.

But having a directory called "tmp" that was not in the
TPDB 4.0 seems more like a mistake than the usual duplication stuff.

Best regards,
Peter

Simon Bailey wrote:
> hi,
> 
> On Oct 23, 2008, at 8:49 PM, Carsten Fuhs wrote:
>> I have just had a look at the CSR category of the TPDB for the rerun  
>> of
>> the competition of 2007:
>>
>> http://colo5-c703.uibk.ac.at:8080/termcomp/competition/categoryDetail.seam?categoryId=12629&cid=350
>>
>> There seem to be several duplicate examples in this category:
>>
>> I checked "tpdb-4.0/TRS/CSR/ExIntrod_Zan97.trs" (#49)
>> and "tpdb-4.0/TRS/CSR/tmp/ExIntrod_Zan97.trs" (#130), and it looks  
>> like
>> the two CSRs are identical. I presume that all CSRs with "tmp/" in
>> their path are duplicates of the corresponding examples of the same
>> filenames without "tmp/" in the path (in total 43 examples).
>>
>> The same issue arises for "tpdb-4.0/TRS/CSR/ 
>> ExSec11_1_Luc02a.trs" (#51)
>> and "tpdb-4.0/TRS/TRCSR/ExSec11_1_Luc02a.trs" (#132).
>>
>> What is the intended policy for these 44 examples at the upcoming
>> competition?
> 
> it is a known fact that there are duplicate systems in the tpdb. when  
> this was discussed at the meeting in leipzig, the general consensus  
> was to leave these duplicate systems in the tpdb, iirc.
> 
> as far as i know, rené has written a tool that tests all of the  
> systems for equality modulo variable names, i think he came up with  
> more than 44 examples which are duplicates.
> 
> regards,
> sb
> _______________________________________________
> Termtools mailing list
> Termtools at lists.lri.fr
> http://lists.lri.fr/mailman/listinfo/termtools



-- 
Peter Schneider-Kamp   mailto:psk at informatik.rwth-aachen.de
LuFG Informatik 2      http://verify.rwth-aachen.de/psk
RWTH Aachen            phone: +49 241 80-21211


More information about the Termtools mailing list