W3C home > Mailing lists > Public > public-swbp-wg@w3.org > January to March 2004

Re: ALL: philosophy of SWBPD (was Re: [OPEN] and/or [PORT] : a practical question)

From: Christopher Welty <welty@us.ibm.com>
Date: Thu, 25 Mar 2004 08:48:22 -0500
To: Jim Hendler <hendler@cs.umd.edu>
Cc: Bernard Vatant <bernard.vatant@mondeca.com>, Ian Horrocks <horrocks@cs.man.ac.uk>, Jeremy Carroll <jjc@hplb.hpl.hp.com>, SWBPD <public-swbp-wg@w3.org>, public-swbp-wg-request@w3.org
Message-ID: <OF4A1D5D5A.1272309B-ON85256E62.0048764D-85256E62.004BD707@us.ibm.com>
Jim wrote on 03/24/2004 05:34:06 PM:

> See, it's this kind of converse that makes me nervous -- somehow the 
idea that the 
> people who prefer separating class from instance (as Ian is quoted by 
Jeremy) are 
> right and those who prefer to use metamodeling (like Guus as quoted to 
WOWG. I 
> don't have time to dig up the mail) are somehow mismodeling.  This is 
nonsense -- 

I agree.  This statement is nonsense.  This kind of modeling is well 
understood by people who understand logical modeling, it is available in 
many systems and it is more heavily used than some OWL DL features, like 
arbitray number restrictions (cardinalities other that 0 and 1).

However, it is the case that people get it wrong, or make poor modeling 
choices that people with experience can say, e.g. "I tried it that way 
once and this is what happened...." or "that is wrong because ..." 

Given that happens, *I* believe part of our mandate IS to show people good 
ways to model the world.  And, sorry if you don't like it, but that 
includes pointing out bad or plainly wrong ways to model the world.  In my 
view, that there are "worst" practices is implicit in the name "Best" 
practices. 

So I am saying using "classes as instances" as a heuristic to determine 
that someone made a modeling mistake would be bad, but I do consider this 
to be a fairly important issue for us to explain and show both good and 
bad (i.e. pitfalls) examples. 

I know there is a widespread fear that we will pick on people - it is 
certainly the case that a lot of the ontologies in the DAML+OIL repository 
are chock full of these pitfalls, as are many of the OWL examples often 
used by people who participated in Webont.  But we won't be naming names. 
And like everything else in this semi-anarchy of the web, you are free to 
ignore it, as many people will.

If you study something, e.g. representation and reasoning algorithms, for 
a long time, you will become an expert in that and be able to understand 
and perceive things that people without that experience won't.  I'm not 
sure why there is this pervasive belief that this is not true for ontology 
design.  We have fairly good representation in this group from the 
community that has been studying that for the last 10-20 years, and <big 
surprise> we have learned some things.  And <big surprise> you're going to 
hear about it.

>  My big fear for this WG is that we're going to somehow "endorse" 
certain kinds of 
> representation and say other folks are somehow making errors

Your fear is my hope, though I'm not interested in the "who".  Just 
explaining what's good and what's bad.

> - yet on the web, 
> different people with different opinions about representation will all 
need to use 
> the languages, we must be careful not to be like the "soup nazi"s in the 
Seinfeld 
> show [1] who get to dictate who gets their soup and who doesn't based on 
some set 
> of rules that no one else understands...

Certainly this idea of desribing best and worst practices could be carried 
too far.  I agree we must be careful.  But you have been saying this so 
much it's starting to sound like "don't do it at all".

>  Seriously, I think the BPD will do a great service if we explain the 
issues and 
> the advantages and disadvantages of various representations - but if we 
start to 
> dictate one way or the other as "correct" then we will be doing a 
disservice to the
> community and will not be helping to deploy the semantic web.

I suspect you mean to emphasize the "one" there.  And I agree with that, 
absolutely.  There are often many correct ways to model the same thing, 
and there are many incorrect ways.

-Chris
Received on Thursday, 25 March 2004 08:51:24 EST

This archive was generated by hypermail pre-2.1.9 : Thursday, 25 March 2004 08:51:36 EST