Re: WordNet 2.0 data files in MySQL format available for download
From: Robert Oschler (no-mail-please_at_nospam.com)
Date: 10/05/04
- Next message: projectshave_at_yahoo.com: "Persistant graphs"
- Previous message: reader: "Re: Basic Prolog problem"
- In reply to: Jan Wielemaker: "Re: WordNet 2.0 data files in MySQL format available for download"
- Next in thread: Paul Tarau: "Re: WordNet 2.0 data files in MySQL format available for download"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Tue, 5 Oct 2004 15:22:06 -0400
"Jan Wielemaker" <jan@ct.xs4all.nl> wrote in message
news:slrncm5nos.ni8.jan@ct.xs4all.nl...
> In article <e8udnQ3S7LwuXf_cRVn-hA@adelphia.com>, Robert Oschler wrote:
>
> Robert,
>
> Thanks for the kind words. Luckily for me more people are contributing
> these days. I couldn't have done constraints and coroutining without the
> support from Bart Demoen and Tom Schrijvers who not only contributed a
> lot of code, but also saved me a lot of time by answering questions on
> `how things are done' in areas I knew little about.
>
> For the argument though, I think using a database to make WordNet
available
> from Prolog only makes sense if we are talking about a toy Prolog system.
> Any today Prolog system aiming at serious applications should -on todays
> desktop hardware- handle WordNet without a hitch and the access will be
> *much* faster than using an RDMS in client/server mode.
>
> This doesn't really imply there are no cases where you wouldn't use
> MySQL. If you have a setup involving various applications that require
> WordNet, Prolog being just one of them, the Prolog process lives shortly
and
> the required access is infrequent it may be a good option. This is an
> extreme case though.
>
> Cheers --- Jan
Actually Jan that is precisely what happened to us. We needed shared access
to the WordNet files from Prolog, Delphi (Object Pascal), and PHP script
(browser access). Access in all three cases is sporadic, unlike a inference
engine crunching full time on multiple documents, which would be better
serviced by native Prolog access because of speed isssues (as you pointed
out). So in essence I am in full agreement with your logic.
Thanks,
Robert.
- Next message: projectshave_at_yahoo.com: "Persistant graphs"
- Previous message: reader: "Re: Basic Prolog problem"
- In reply to: Jan Wielemaker: "Re: WordNet 2.0 data files in MySQL format available for download"
- Next in thread: Paul Tarau: "Re: WordNet 2.0 data files in MySQL format available for download"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|