W3C home > Mailing lists > Public > public-owl-comments@w3.org > February 2009

Re: [LC response] To Jim Hendler (was Re: Fwd: Question re: HasKey entailments)

From: Jim Hendler <hendler@cs.rpi.edu>
Date: Thu, 19 Feb 2009 15:13:24 -0500
Cc: public-owl-comments@w3.org, "Ralph R. Swick" <swick@w3.org>, W3C OWL Working Group <public-owl-wg@w3.org>
Message-Id: <26B9DBC8-1B15-4A50-9B6F-3EB8C4BF6B3D@cs.rpi.edu>
To: Bijan Parsia <bparsia@cs.man.ac.uk>
OK, seems to me that the response was signed off on by the WG, and  
since it said "The working group was unsure" I assumed it was approved  
by the WG - nevertheless, I see where the misunderstanding was.  If  
the WG approves the response with the offending sentence removed, and  
submits is as an updated LC comment, I will be happy to respond as to  
whether I accept the technical substance or not...
  JH


On Feb 19, 2009, at 11:59 AM, Bijan Parsia wrote:

> On 19 Feb 2009, at 15:02, Jim Hendler wrote:
>
>> Bijan-
>> I understand clearly that you (personally) would not mean it in an  
>> insulting way, but coming from a WG, in an official form, implying  
>> somehow that if I knew more (or maybe less) or came from a  
>> different background I'd have "gotten it" seems to me to be very  
>> out of place in this sort of comment.
>
> As the person who deliberately put it in, I don't understand why.  
> You introduced the idea that your *sort* of expertise was relevant:
> 	"<http://www.w3.org/mid/3D9CD8CA-8994-4703-93EF-2E0753B2BF8E@cs.rpi.edu 
> >
> """"However,  given two of us who had PhDs in AI and long experience  
> with DBs took a while to work through the semantics, and didn't get  
> the idea of these examples without the emails from you folks,  I do  
> think documenting it will be important..."""
>
> I certainly didn't mean to imply that you had too *little*  
> expertise, but just that your *confusion* may have stemmed from  
> thinking "too hard" about it, as someone who is a designer of  
> ontology languages. It's not clear to me that we should optimize  
> specs for *that* case, rather than a more general audience for whom,  
> perhaps, the question would not even arise.
>
> I see that my terseness about it in this message could be read  
> differently.
>
>> My earlier response to you, which you quoted, was meant to be  
>> focused on the specific sentence under discussion at that time, and  
>> I wanted to indicate it would be a good addition.
>
> Sure, but I hope you understand why I read it otherwise. This is why  
> I would like to stress that the working group, as a whole or the  
> rest of the participants, is not at fault. I insistent that you were  
> fine with the draft text as written because I thought, based on that  
> email, that you were.
>
>> I still feel that the comment is inappropriate - it somehow implies  
>> that if I had a different background I'd have seen this from the  
>> original documents.
>
> Or not noticed it as a problem. Yes, that was what implied because I  
> wasn't clear that that wasn't the case and it was a relevant aspect  
> of my design of the fix.
>
> I reiterate that I am solely responsible for the *group's* being  
> unsure (i.e., it was unsure because *I* was unsure). The working  
> group deferred to me because I said, forcefully, that you were ok  
> with this response.
>
>>  I feel that I have a pretty strong background with respect to the  
>> semantic web and to knowledge representation, and having helped  
>> chair the first Web Ontology Working Group, I think I have a pretty  
>> good background in OWL (in fact, I even coauthored what is now the  
>> best selling book in the area).  I would also point out that in  
>> email threads going back to the beginning of this WG there was  
>> discussion of Easy Keys essentially being a way to deal with the  
>> issue of inverseFunctional datatype properties, and thus there was  
>> an expectation that doesn't seem to me to be from some "other  
>> perspective" that, like inverseFunctional, the key property would  
>> imply a domain restriction
>
> InverseFunctionality does not imply a domain restriction. For example:
> 	<http://www.w3.org/TR/2008/WD-owl2-syntax-20081202/#Inverse-Functional_Object_Properties 
> >
>
> """InverseFunctionalProperty( a:fatherOf ) Each object can have at  
> most one father."""
>
> There's no domain implication that I can see.
>
>> - and if you read my original response, you'll see that while I was  
>> asking why this was the case, I was also pointing the editorial  
>> problem of recognizing that this was not the case from the  
>> documents (required fairly careful analysis of multiple sections)  
>> which doesn't seem to me to have anything to do with my expertise.
>
> That all seems reasonable. The reason I even thought about it this  
> way was because of your final comment in your thread. I was asking  
> myself whether it wasn't the fact that you were *too deep* into OWL,  
> databases, etc. that led you this way.
>
>> Once the advantage of the solution in the WG was pointed out, I  
>> certainly agreed that the design was okay, and my comment was  
>> turned into an LC comment since it was felt at that point to be the  
>> documentation issue that was in discussion.
>> I know some on the WG will feel my response was an over-reaction,
>
> Regardless, your ire should be directed to me.
>
>> but I did get email from a colleague on the fact that this public  
>> comment seemed to be a polite pejorative, and that LC responses are  
>> part of the official record available to the public and to the AC  
>> for review during the PR decision, and thus I felt it needed to be  
>> pointed out in a manner that would not only remove the passage from  
>> the response in question, but also remind the WG that these are not  
>> casual emails, but public responses.
>
> This is, again, a criticism of me. I wrote the response and wrote it  
> this way. I championed it. I championed this specific bit of it  
> because I thought (wrongly) that it was helpful. I insisted that  
> there was no problem with it because I thought I had your buy in.
>
>> I appreciate your personal response, but I still do not feel it  
>> appropriate to respond technically until  I see both a new response  
>> without the comment in the LC comments and a a publicly readable  
>> apology from the WG that I could point colleagues to if, like the  
>> one who pointed this out to me, they are wondering why the original  
>> comment was included.
>
> Well, I think the better solution would have been to respond  
> privately and get the mail deleted from the archives. The second  
> response could have included a note apologizing for a "draft" having  
> been sent. I guess we're past that now.
>
>> Sorry to ask for WG time for something seemingly so trivial, but  
>> perhaps it will help remind people that we are not in an academic  
>> forum where debate and personalization is par for the course, but  
>> rather in an industrial standardization effort trying to create a  
>> durable specification that will be of interest to the more than 400  
>> companies in the W3C consortium, and if we are successful,  
>> eventually to the 1.4 billion people (more than 20% of humanity)  
>> who use the Web.
>
> Well, first, it wasn't a personal comment, technically speaking.  
> That is, my intent wasn't speak about *you* in particular (any more  
> than your intent in mentioning your experience above was to be about  
> *you* but about a larger class of people, i.e., if people like you  
> had trouble, then so too would people with less expertise).  
> Similarly, someone else, with a strong DB background, complained  
> about keys not being functional, like database keys. It's not clear  
> to me that their finding that disconcerting is likely in the less  
> expert audience for this document.
>
> Regardless, if someone else read it as dissing you, then it is  
> unfortunate and would have been better not included. If you are  
> concerned about this sort of thing, then I suggest you suggest to  
> the chairs that I not be charged with writing responses as it seems  
> that I am prone to this sort of thing.
>
> Cheers,
> Bijan.
>

"If we knew what we were doing, it wouldn't be called research, would  
it?." - Albert Einstein

Prof James Hendler				http://www.cs.rpi.edu/~hendler
Tetherless World Constellation Chair
Computer Science Dept
Rensselaer Polytechnic Institute, Troy NY 12180
Received on Thursday, 19 February 2009 20:14:17 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 19 February 2009 20:14:17 GMT