[Termtools] second stage (5 minutes) overwrites previous results?

Claude Marche Claude.Marche at lri.fr
Wed Jun 21 13:01:51 CEST 2006


When proposing to have a second round with a larger timeout, I wanted
to see if this will rise interesting questions. I'm glad to see it
happens. 

>>>>> "Johannes" == Johannes Waldmann <waldmann at imn.htwk-leipzig.de> writes:

    Johannes> Dear all,
    Johannes> Dieter and I were looking at the results
    Johannes> from the second stage of the competition,
    Johannes> and we noticed one strange thing about SRS/Trafo - un17:

    Johannes> TPA can now prove it in 117 seconds,
    Johannes> but Torpa cannot prove it any longer
    Johannes> It could, in the first stage,
    Johannes> BUT Torpa's first-stage proof is gone
    Johannes> (it is overwritten by the DONTKNOW from the second
    Johannes> stage)?

Yes, another bug in my engine... sorry about that. The file containing
the proof has been overridden. 

    Johannes> Note that also some of Jambox's loops have vanished.

What do you mean ?

    Johannes> Normally, this problem would not occur
    Johannes> because the second stage should not be executed
    Johannes> for problems that already had an answer.

But I decided it could be interesting to relax the condition, and I
think I was right, because otherwise the Trafo - un17 would not have
been rerun. 

    Johannes> -- 
    Johannes> -- Johannes Waldmann -- Tel/Fax (0341) 3076 6479/80 --
    Johannes> ---- http://www.imn.htwk-leipzig.de/~waldmann/ -------

    Johannes> _______________________________________________
    Johannes> Termtools mailing list
    Johannes> Termtools at lists.lri.fr
    Johannes> http://lists.lri.fr/mailman/listinfo/termtools

-- 
| Claude Marché           | mailto:Claude.Marche at lri.fr |
| LRI - Bât. 490          | http://www.lri.fr/~marche/  |
| Université de Paris-Sud | phoneto: +33 1 69 15 64 85  |
| F-91405 ORSAY Cedex     | faxto: +33 1 69 15 65 86    |
_______________________________________________
Termtools mailing list
Termtools at lists.lri.fr
http://lists.lri.fr/mailman/listinfo/termtools



More information about the Termtools mailing list