W3C home > Mailing lists > Public > public-owl-wg@w3.org > April 2009

Re: review of Document Overview

From: Sandro Hawke <sandro@w3.org>
Date: Fri, 03 Apr 2009 14:31:45 -0400
To: Bijan Parsia <bparsia@cs.man.ac.uk>
cc: Ian Horrocks <ian.horrocks@comlab.ox.ac.uk>, "Peter F. Patel-Schneider" <pfps@research.bell-labs.com>, public-owl-wg@w3.org
Message-ID: <26965.1238783505@ubehebe>

> > * Full > DL > RL,QL,EL
> > * RL, QL and EL have a non-empty intersection
> 
> At some point, we need to be more secure.

Maybe, after we get to Rec?

> We've made these  
> relationships more sane, which is the *key thing*. We don't need to  
> explain them to death. Leave something for the freelance writers!
> 
> [snip]
> >> Am I the only one who thinks OWL 1 users will want to know, in no
> >> uncertain terms, whether OWL 2 breaks their stuff, BEFORE they accept
> >> OWL 2?
> 
> I don't even know what this means.

*shrug*

I'm not sure how to be more clear.

Everyone out there who cares about OWL has to make a decision about
whether to move forward to OWL 2 or stick with OWL 1.   Some of these
people, as they make their decisions, will affect the market.  Others
will affect us more directly, through their public comments and votes as
AC Reps.

We in the Working Group are pretty sure the choice is easy: OWL 2 is
much better, and there's no significant drawback to using it.

I just think we need to communicate that better.

I think it's important to say (as you suggest) "it's easy -- no
worries", but I also think some users will want to know the gorey
details, and if we make them read carefully through hundreds of pages of
spec to find it, they'll (rightly) be annoyed and even suspicious.  I
should reply to Ian separately, but it seemed to me his language in
Overview, saying OWL 2 is "almost" compatible with OWL 1 is like blood
in water is to sharks.   "What do you mean 'almost'...???"

That said, they're not really sharks, and maybe everyone will go along
fine, and I'm being needlessly dramatic.  Still, I don't see what harm
it does to list clearly, in one place, all the possible gotchas in the
transition. 

> Plenty of people who think they are using OWL 1 have been using OWL  
> 1.1 tools for years. And for many months, they've been using OWL 2  
> tools. As things have been changing even!
> 
> Let's not buy into the puerile confusion and FUD that have been  
> pretty constant since before the first OWLED workshop. Point of fact:  
> Migrating from OWL 1 to OWL 2 is piss easy for almost everyone with a  
> lot of benefits. And you can do it without wading through a bunch of  
> explanations, for the most part.
> 
> Seriously. Download protege4.
> 
> Worries about adoption and market confusion have been *seriously*  
> overstated. Worries about sufficient documentation have also been  
> seriously overstated. Similarly on how much support the transition  
> needs.
> 
> > This is *exactly* what I am trying to achieve here.
> >
> >> Figuring that out by sifting through our entire spec seems a bit
> >> much to ask.
> >
> > That isn't the intention. The intention is to tell them that  
> > everything is fine, *which it is*. I
> 
> YES!
> 
> When I moved from XSLT 1 to XSLT 2, I certainly didn't give a flying  
> fuck about the differences bit...and I'm a spec geek!
> 
> I just used the new document and new tools and was fine.

Sure, but you weren't tied up in the XSLT market in complex business
and/or personal ways.  If you had been, I think you might have been
looking at the list of incompatilities pretty closely as XSLT 2
approached Rec, and might have been arguing for some changes.

> > MHO this message, we should *not* provide irrelevant information  
> > about the corner cases and "bug-fixes" in OWL 2 that prevent us  
> > from simply saying that it is *completely* backwards compatible  
> > with OWL 1 -- this does need to be documented somewhere, but not  
> > here (not sure if it should be normative and/or in NF&R, but this  
> > is a different issue that I will address in another email).
> 
> Actually, I don't think this matters either. Tool builders need to  
> know this, but they should read the specs anyway. If this is a real  
> problem, it'll show up *mighty* quickly and documentation and  
> migration tools will emerge.
> 
> Why do we think OWL users are so much more fragile than the rest of  
> the world? They aren't even (in total --- I'm not sure about per  
> capita) whinier.

I don't think they're more fragile, I just think some parts of any user
community are sensitive to change, often for good reason.  In this case,
I think it's our job to address their concerns, and it looks like that
will be pretty easy here.

      -- Sandro

> Cheers,
> Bijan.
Received on Friday, 3 April 2009 18:31:53 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 3 April 2009 18:31:54 GMT