[Termtools] rules of competitition

[H. Zantema]

Let me react on Johannes email. I can not use my normal email program 
since then the message is bounced by a message that I am a non-member 
that wants to post to a members-only list. Therefore this detour.

For subscription you have to mention not only the categories in which you 
participate, but also the subcategories, like Juergen just mentioned too. 
>From last year we keep:

category TRS:  subcategories: standard
                              relative
                              innermost
                              modulo theory

category SRS:  subcategories: standard
                              relative

Here modulo theory is only modulo AC. This year we have a new category LP 
(just like in 2004) and a new TRS subcategory context-sensitive. For the 
format of these (sub)categories you best look at TPDB.

Many options of the format as it was defined a few years ago are not used 
at all. For instance, outermost is not used any more. Moreover, many 
unintended combinations like having both innermost and outermost are not 
explicitly forbidden. For me it is not clear what is meant if you mix up 
several sets of rules and several variable declarations. I propose not to 
start syntax wars on this, but simply follow the format as it is clear 
from TPDB how it is intended.

Every participant may submit 5 secret problems in the category it 
participates, not for every subcategory. Claude's software automatically 
puts every secret problem in the corresponding subcategory. It is 
encouraged to make secret problems for the standard subcategories, by 
which the largest numbers of tools will apply on them. Last year we had 
bounds on the sizes of secret problems. This year we have not: we do not 
want to force to modify a natural system of 11 rules or having a right 
hand side of 11 symbols. 

Johannes wrote:

> To make a concrete proposal, I plan to make a collection of TRS/SRS 
> problems (test cases) that show
> (a) all expected uses of the specification
>    (all sensible combinations of annotations to problems)
   
What do you mean by expected use? You can work out the expected use of 
modulo a set of strange equations, but as long there are no tools doing 
modulo theory other than modulo AC, I think this does not make sense.


-- 

		Best regards, Hans Zantema.

+--------------------------------------+-----------------------------+
|                                      |                             |
|   Dr Hans Zantema                    |   Hoofdgebouw   kamer 6.73  |
|   Faculteit Wiskunde en Informatica  |   tel (040)2472749          |
|   Technische Universiteit Eindhoven  |   fax (040)2468508          |
|   Postbus 513     5600 MB Eindhoven  |   e-mail  H.Zantema at tue.nl  |
|   The Netherlands                    |   www.win.tue.nl/~hzantema  |
|                                      |                             |
+--------------------------------------+-----------------------------+
_______________________________________________
Termtools mailing list
Termtools at lists.lri.fr
http://lists.lri.fr/mailman/listinfo/termtools