perspectives on OWL v.next and RDF

[Dan Connolly]

Bijan,

Thanks for your recent item on the relationship between OWL 1.1
and RDF...

http://clarkparsia.com/weblog/2006/11/13/not-so-alternate-a-view-of-owl-11/

I it addresses many of the concerns I have that Jim and Ora expressed
http://www.mindswap.org/blog/2006/11/11/an-alternative-view-for-owl-11/
http://www.lassila.org/blog/archive/2006/11/more_on_rdf.html

Between the 3 of those posts, I feel the urgency to finish my own
has gone down to the point where I can collect my thoughts
on OWL 1.1 without rushing too much. I'm still thinking about how
to reconcile punning, in particular...

Oh... and I have been doing some relevant work...

  Using ACL2 with Semantic Web Data
  http://dig.csail.mit.edu/2006/Papers/owl-acl2/doc20

I hope to look at the axiomatic semantics Jim mentions
in that light.

--
Dan Connolly, W3C http://www.w3.org/People/Connolly/
D3C2 887B 0F92 6005 C541  0875 0F91 96DE 6E52 C29E

[Jim Hendler]

Some parts of this message have been removed. Learn more about Nabble's security policy.

At 1:56 PM -0600 11/13/06, Dan Connolly wrote:

>Bijan,
>
>Thanks for your recent item on the relationship between OWL 1.1
>and RDF...
>
>http://clarkparsia.com/weblog/2006/11/13/not-so-alternate-a-view-of-owl-11/
>
>I it addresses many of the concerns I have that Jim and Ora expressed
>http://www.mindswap.org/blog/2006/11/11/an-alternative-view-for-owl-11/

>http://www.lassila.org/blog/archive/2006/11/more_on_rdf.html
>
>Between the 3 of those posts, I feel the urgency to finish my own
>has gone down to the point where I can collect my thoughts
>on OWL 1.1 without rushing too much. I'm still thinking about how
>to reconcile punning, in particular...
>
>Oh... and I have been doing some relevant work...
>
>  Using ACL2 with Semantic Web Data
>  http://dig.csail.mit.edu/2006/Papers/owl-acl2/doc20
>
>I hope to look at the axiomatic semantics Jim mentions
>in that light.
>
>--
>Dan Connolly, W3C http://www.w3.org/People/Connolly/

>D3C2 887B 0F92 6005 C541  0875 0F91 96DE 6E52 C29E

Bijan - in the interest of bringing things to the mailing lists instead of blog wars (which actually seem to have better Google karma, but never mind) - you put the following as deliverables for a potential WG

Deliverables

    * Syntax and Semantics (S&S)
    * Outreach material
          o update/diff//what the wg decides
          o New material (e.g., tractable fragments)

    * Test Suite

this reveals one difference - I am not interested in elucidating the many tractable fragments, that's not a WG activity in my opinion - I'd be interested in identifying one particular tractable fragment, naming it, and getting it into note or rec form

to show where we are at the moment - the attached (HTML table) reflects where some of us have been playing - comments welcome

 -JH

-- 

Prof James Hendler                              hendler@...
Dept of Computer Science                      http://www.cs.umd.edu/~hendler
AV Williams Bldg                          301-405-2696 (work)
Univ of Maryland                             301-405-6707 (Fax)
College Park, MD 20853 USA

Vocabulary Comparison Matrix	Proposal 1	Proposal 2	OWL Lite
RDF Schema Features:
* Class (Thing, Nothing)	yes	yes	yes
* rdfs:subClassOf	yes	yes	yes
* rdf:Property	yes	yes	yes
* rdfs:subPropertyOf	yes	yes	yes
* rdfs:domain	yes	yes	yes
* rdfs:range	yes	yes	yes
* Individual	yes	yes	yes
"Annotation" Properties:
* rdfs:label	yes	yes	yes
* rdfs:comment	yes	yes	yes
* rdfs:seeAlso	yes	yes	yes
* rdfs:isDefinedBy	yes	yes	yes
ADD FROM OWL
(In)Equality:
* equivalentClass	yes	yes	yes
* equivalentProperty	yes	yes	yes
* sameAs	yes	yes	yes
* differentFrom	yes	yes	yes
Property Characteristics:
* inverseOf	yes	yes	yes
* TransitiveProperty	yes	yes	yes
* SymmetricProperty	yes	yes	yes
* FunctionalProperty	yes	yes	yes
* InverseFunctionalProperty	yes	yes	yes
* AnnotationProperty	no	yes	yes
MAYBE:
* AllDifferent, distinctMembers	?	?	yes	Note: this is just syntatic sugar for "differentFrom"
* ObjectProperty	?	yes	yes
* DatatypeProperty	?	?	yes
* disjointWith	?	yes	no
Class/Restriction/MISC
* complementOf	no	yes*	no	Note: complementOf is doable because we don't have class intersection or union
* oneOf	no	no	no
* Restriction	no	no	yes
* allValuesFrom, someValuesFrom	no	no	yes
* minCardinality, maxCardinality	no	no	yes*	Note: cardinality can only be 0 or 1 for OWL LITE
* cardinality	no	no	yes*
* unionOf	no	no	no
* intersectionOf	no	no	yes
* hasValue	no	no	no
* DataRange	no	no	no
Note: OWL DL/FULL has everything.

[Danny Ayers]

On 13/11/06, Jim Hendler <hendler@...> wrote:

> Bijan - in the interest of bringing things to the mailing lists instead of
> blog wars

I'm here :-)

I don't have the expertise in OWL to offer much to the debate, but I
do care about the Web. So with regards to this:

> Deliverables
>
>      * Syntax and Semantics (S&S)
>      * Outreach material
>            o update/diff//what the wg decides
>            o New material (e.g., tractable fragments)
>     * Test Suite

I would suggest that the test suite covers compatibility testing, to
ensure that any proposal which is intended to build on the SW stack
doesn't conflict with the existing RDF and OWL layers. (I am not a
logician, but a first thought is that lossy round-tripping across any
mappings would be considered acceptable, round-tripping that might
lead to contradiction, not).

In addition (a point I've been struggling to get across in the
blogosphere), surely the primary consideration for a Web Ontology
Language should be the (Semantic) Web. Compatibility with deployed Web
languages should take precedence over the requirements of any
(effectively) offline users. The onus should be on the proposers of
any development to demonstrate there won't be any Web-breaking.

In my opinion, in this context the specific exchange syntax is of
secondary importance.

I personally have doubts about the wisdom of making a significant
increment to OWL at this point in time because of potential impact on
the adoption of RDF and OWL, but am happy to defer to Kendall, Bijan
and co. on the point that there is real demand for certain features.

Cheers,
Danny.

--

http://dannyayers.com

[Jim Hendler]

Danny Ayers wrote:
>I personally have doubts about the wisdom of making a significant
>increment to OWL at this point in time because of potential impact on
>the adoption of RDF and OWL, but am happy to defer to Kendall, Bijan
>and co. on the point that there is real demand for certain features.

I think he said that much better than I did!   I think simplification
is needed before pushing new functionality - I've said that many
times, but then I spent two years saying it to the WG and failed
there as well :-)   Seriously, I have no object to new design work
going on, but I think I'd be happier if  Googling "-xxxxxxx
filetype:owl" returned more hits before we began extending...
  -JH

--
Prof James Hendler hendler@...
Dept of Computer Science http://www.cs.umd.edu/~hendler
AV Williams Bldg 301-405-2696 (work)
Univ of Maryland 301-405-6707 (Fax)
College Park, MD 20853 USA

[Bijan Parsia]

Thanks Dan. Glad the post helped.

I'm still readjusting to being back in the UK and the email address I  
have subscribed to this list has been flaky. I do see some email from  
Jim and Danny which I hope to address later today (which may be early  
for those not in God's Own Time Zone :)).

Cheers,
Bijan.

[Kendall Clark]

On Nov 13, 2006, at 6:49 PM, Jim Hendler wrote:

>
> Danny Ayers wrote:
>> I personally have doubts about the wisdom of making a significant
>> increment to OWL at this point in time because of potential impact on
>> the adoption of RDF and OWL, but am happy to defer to Kendall, Bijan
>> and co. on the point that there is real demand for certain features.
>
> I think he said that much better than I did!   I think  
> simplification is needed before pushing new functionality - I've  
> said that many times, but then I spent two years saying it to the  
> WG and failed there as well :-)   Seriously, I have no object to  
> new design work going on, but I think I'd be happier if  Googling "-
> xxxxxxx filetype:owl" returned more hits before we began extending...

I consistently fail to understand the import of that kind of claim,  
as it seems to suggest that the public Semantic Web is the most  
important thing.

Doesn't that ignore the history of the Web itself, which, as we all  
know, got a significant boost -- several such boosts, actually --  
from enterprise intranet adoption. But now for the Semantic Web the  
equivalent of "intranet adoption" seems not only not to matter but  
seems to be a problem.

I don't get it.

Lots more people are using OWL than are using it "on the public Web".  
Isn't that a *good* thing? I think it is. :>

Cheers,
Kendall

[Jim Hendler]

At 3:05 PM -0500 11/14/06, Kendall Clark wrote:

>On Nov 13, 2006, at 6:49 PM, Jim Hendler wrote:
>
>>
>>  Danny Ayers wrote:
>>>  I personally have doubts about the wisdom of making a significant
>>>  increment to OWL at this point in time because of potential impact on
>>>  the adoption of RDF and OWL, but am happy to defer to Kendall, Bijan
>>>  and co. on the point that there is real demand for certain features.
>>
>>  I think he said that much better than I did!   I think
>>simplification is needed before pushing new functionality - I've
>>said that many times, but then I spent two years saying it to the
>>WG and failed there as well :-)   Seriously, I have no object to
>>new design work going on, but I think I'd be happier if  Googling
>>"-xxxxxxx filetype:owl" returned more hits before we began
>>extending...
>
>I consistently fail to understand the import of that kind of claim,
>as it seems to suggest that the public Semantic Web is the most
>important thing.
>
>Doesn't that ignore the history of the Web itself, which, as we all
>know, got a significant boost -- several such boosts, actually --
>from enterprise intranet adoption. But now for the Semantic Web the
>equivalent of "intranet adoption" seems not only not to matter but
>seems to be a problem.
>
>I don't get it.
>
>Lots more people are using OWL than are using it "on the public
>Web". Isn't that a *good* thing? I think it is. :>
>
>Cheers,
>Kendall

point taken, but one would expect the uptake on the public side to be
continuing while the other goes on - It is unclear to me why intranet
adoption would favor more expressivity, woudl assume it to be about
the same - a lot of the DOD projects I've been involved with are
using relatively low expressivity w/large ABOX as well
  again, I think the clarity of messaging and the development of a
simpler subset are both necessary to better OWL adoption - i don't
claim sufficiency, but if I have to prioritize (and I do, at least
w/respect to my time and that of my employees) then I would prefer to
see us fill the simplicity gap before chasing the expressivity end -
I think OWL DL/Full is expressive enough to hold most people for a
while...
  -JH
--
Prof James Hendler hendler@...
Dept of Computer Science http://www.cs.umd.edu/~hendler
AV Williams Bldg 301-405-2696 (work)
Univ of Maryland 301-405-6707 (Fax)
College Park, MD 20853 USA

[Danny Ayers]

On 14/11/06, Kendall Clark <kendall@...> wrote:

> I consistently fail to understand the import of that kind of claim,
> as it seems to suggest that the public Semantic Web is the most
> important thing.

The (Semantic) Web is a public entity. It seems a straightforward
distinction, by definition things that aren't connected to it aren't
part of the Web. It follows that systems that are connected to it (or
could potentially be connected to it) *are* the most important thing.

> Doesn't that ignore the history of the Web itself, which, as we all
> know, got a significant boost -- several such boosts, actually --
> from enterprise intranet adoption. But now for the Semantic Web the
> equivalent of "intranet adoption" seems not only not to matter but
> seems to be a problem.

I absolutely agree that "intranet adoption" and development is hugely
valuable. The point is only relevant when conflicts occur - in such
cases, (IMHO) priority should be given to solutions that show better
compatibility with the (Semantic) Web.

> I don't get it.
>
> Lots more people are using OWL than are using it "on the public Web".
> Isn't that a *good* thing? I think it is. :>

It is good to hear of significant off-Web adoption of OWL.
Incidentally I'm currently working on a medical records system that
uses RDF - it's data you really wouldn't want public on the Web. No
problem.

A problem only arises if the technologies developed in such an
environment are incompatible with the Web. The reassurances so far
with regard to mappings go a long way to alleviating such concerns.
All I'm actually asking is that care is taken to ensure any new stuff
won't break the Web. I don't think that's a lot to expect from a Web
Ontology Language.

Cheers,
Danny.

--

http://dannyayers.com

[Dan Connolly]

On Wed, 2006-11-15 at 12:13 +0100, Danny Ayers wrote:
> On 14/11/06, Kendall Clark <kendall@...> wrote:
>
> > I consistently fail to understand the import of that kind of claim,
> > as it seems to suggest that the public Semantic Web is the most
> > important thing.
>
> The (Semantic) Web is a public entity.

Oh? I don't think so. As I said to the XML 2005 crowd
and to several other audiences where I have used these
slides since then...

[[
Scaling
      * yes, scalability to 10^9 nodes and up is important
      * but so is scalability down to families and scout troops
]]
 -- slide "Getting into the Web: downhill steps"
 http://www.w3.org/2002/12/cal/mash/slides#(5)

See also
 The Fractal nature of the Web
 Tim Berners-Lee
 Date: 1998
 Status: personal view only. Editing status: first draft.
 http://www.w3.org/DesignIssues/Fractal


>  It seems a straightforward
> distinction, by definition things that aren't connected to it aren't
> part of the Web.

Intranets are often connected to the Web with links that go one
way but not the other.

--
Dan Connolly, W3C http://www.w3.org/People/Connolly/
D3C2 887B 0F92 6005 C541  0875 0F91 96DE 6E52 C29E

[Danny Ayers]

On 15/11/06, Dan Connolly <connolly@...> wrote:

> On Wed, 2006-11-15 at 12:13 +0100, Danny Ayers wrote:
> > On 14/11/06, Kendall Clark <kendall@...> wrote:
> >
> > > I consistently fail to understand the import of that kind of claim,
> > > as it seems to suggest that the public Semantic Web is the most
> > > important thing.
> >
> > The (Semantic) Web is a public entity.
>
> Oh? I don't think so.

Sorry, I forgot an important qualifier (which I'd included when I
previously commented on this question on Kendall's blog): - The Web is
essentially a public entity (even if areas are covered by significant
access controls).

Also:
[[
My definition of the Web is a universe of network-accessible information...
]]
http://www.w3.org/DesignIssues/Architecture.html

Cheers,
Danny.

--

http://dannyayers.com

[Bijan Parsia]

On Nov 15, 2006, at 1:10 AM, Jim Hendler wrote:
[snip]

> point taken, but one would expect the uptake on the public side to  
> be continuing while the other goes on - It is unclear to me why  
> intranet adoption would favor more expressivity, woudl assume it to  
> be about the same
>

While I share the belief that intranet adoption is a perfectly fine  
rationale for adding (or removing) something, I would like to point  
out that I certainly don't think that the extra expressivity is being  
driven by "intranet" adoption...at least, I sure didn't see it that  
way. The person who, afaik, introduced this claim was Danny Ayers.  
 From what I recall, he thinks that the kind of life sciences users  
such as NCI, Galen, Snomed, etc. that want things like qualified  
cardinality restrictions are "effectively offline users". Kendall's  
point was, even if you *grant* this (what I think to be false)  
premise, it doesn't invalidate their needs or make them less useful  
for driving forward the semantic web.

Of course, I see the life sciences folks as paradigm semantic web  
users, or could be users. Think of the NCI thesaurus! It's great  
thing that it's out in public (thanks to you) and in OWL. It's  
wonderful that the NCI people are migrating from a private, closed  
system with an idiosyncratic, well, everything, to a version of  
protege + open source reasoners. Just getting the stuff *on* the web  
in a reasonably interoperable format is very important.

All the features *added* are of this sort --- demanded by users at  
OWLED (and in other fora!). These are not *all* the users, obviously,  
but that's a reason why the increment is *very* small (basically, not  
too much more than DAML+OIL), but potentially high impact. That's  
also why we included more rational fragments.


> - a lot of the DOD projects I've been involved with are using  
> relatively low expressivity w/large ABOX as well
>

Sure. And there are large TBoxes from life sciences (for which the EL+
+ fragment was developed...but EL++ includes qualified number  
restrictions...so we sort of have to add them to the whole).

Also, consider more expressive datatypes. It's really nice for a  
number of applications (Web service policies come to mind!) to be  
able to do comparisons. Last year, in fact, surprisingly, datatypes  
came ahead of qualified cardinality restrictions.


>  again, I think the clarity of messaging and the development of a  
> simpler subset are both necessary to better OWL adoption - i don't  
> claim sufficiency, but if I have to prioritize (and I do, at least  
> w/respect to my time and that of my employees) then I would prefer  
> to see us fill the simplicity gap before chasing the expressivity  
> end - I think OWL DL/Full is expressive enough to hold most people  
> for a while...
>

Of course, none of us, at all, have done a formal market survey, so  
we're all giving impressions. Impressions are better at determining  
positives rather than negatives. We all hear, "rational subsets!" but  
perhaps you didn't hear, "Qualified number restrictions" (and, in EL+
+, they unite). I do think that catering to the life science  
ontologists is important (even if "offline")...one cool drug  
discovery facilitated by OWL is a great talking point.

One very interesting paper about the trade off is:
        http://owl-workshop.man.ac.uk/acceptedPosition/submission_19.pdf

""" Next, we used DL subsumption reasoning to classify
MED. In comparing the classified hierarchy for such defined concepts  
with the original
taxonomy in MED, we found 44 additional subsumptions for laboratory  
concepts. On
manual analysis of newly inferred subsumptions, 26 were correct  
subsumptions i.e. the
concepts actually had a subclass relationship, as confirmed by a  
domain expert.
Interestingly, the false positives revealed systematic modeling errors.

The important result here is not that we identified these modeling  
errors due to the
increased expressivity of DL.  More important is our finding that the  
missed subsumptions
could have cost the hospital many missing results in various decision  
support and
infection control systems that routinely use MED to screen patients.  
"""

So, I think we all would agree is that expressivity for expressivity  
sake isn't interesting. Results are what matter.

Finally, I would like to point out that the tractable fragments OWL  
1.1 document includes fragments that are user driven. EL++ was  
specifically done to address the very large medical ontologies. DL  
Lite was motivated by database integration (not integrating reasoners  
with databases, but integrating distinct database). Some variants of  
DL Lite cover foaf. Surprisingly many ontologies turn out to be  
hornSHIQ (as Boris found out).

Finally, I venture this explanation: one reason people with large  
scale problems haven't used more expressivity is that we've not  
explained well what expressivity they can "safely" use. OWL Lite  
definitely failed on that front. I would like to improve on that. I  
get good reactions from people when I explain these alternatives, FWIW.

Cheers,
Bijan.

[Bijan Parsia-3]

[Won't get caught up today :)]
On Nov 13, 2006, at 8:54 PM, Jim Hendler wrote:
[snip]

> Bijan - in the interest of bringing things to the mailing lists  
> instead of blog wars (which actually seem to have better Google  
> karma, but never mind) - you put the following as deliverables for  
> a potential WG
>
> Deliverables
>
>     * Syntax and Semantics (S&S)
>     * Outreach material
>           o update/diff//what the wg decides
>           o New material (e.g., tractable fragments)
>     * Test Suite
>
> this reveals one difference - I am not interested in elucidating  
> the many tractable fragments,

That document tried to capture all those "of interest", that is,  
those which we knew had implementations and user communities and  
seemed to have "naturally useful" expressivity. You might notice that  
we did not include any of the "classic" tractable DLs like FL0 or  
even Classic, for exactly these reasons. (Classic might be  
interesting, but my impression is that Classic users mostly migrated  
to more expressive logics, rather than staying in Classic. Or, now  
that I think about it, EL++ might capture a heck of a lot of classic  
uses.)

> that's not a WG activity in my opinion

Really? That seems a bit strong. I don't see why having a spread of  
interesting fragments, if they can attract markets, is not a WG  
activity. I would like to name them, and use the "species" framework  
that WebOnt set up to help guide users.

> - I'd be interested in identifying one particular tractable  
> fragment, naming it, and getting it into note or rec form

I don't see why just one if there are others vendors are willing to  
support directly. There is already an EL++ vendor (Arity something?  
they were at ISWC) and a hornSHIQ and DLP vendor. I think DL Lite  
will have some commercial activity soon. The nice thing is that these  
are all OWL! So let's call them OWL! I think there's a lot of room  
here and that people will use these fragments.

What I would expect the working group to do is examine things like  
the tractable fragments document and your document and others,  
solicit comment, and define a useful and rational subset that was  
helpful to the community.

I think there is room for more fine grained identification: the  
expressivity checker in Swoop is a much loved feature in my  
experience. But *that* seems outside the scope of a WG since there's  
nothing really to standardize and no reason to standardize it. (I  
guess we could give URIs to everything, but that's not so hard nor  
particularly helpful.) However, identifying subsets that are useful,  
easier to use, amenable to scaling or to certain implementation  
techniques *is* useful, because they can help users and vendors "meet  
up".

> to show where we are at the moment - the attached (HTML table)  
> reflects where some of us have been playing - comments welcome

Quick comment, I notice that the chart says that OWL Lite lacks  
complementOf and disjointWith...that's only true syntactically, not  
semantically (one of its problems).  Experience with other tractable  
fragments shows that "disjointWith" is much safer than  
"complementOf". I'd look at DLP or hornSHIQ. I'll point out that DL  
Lite has undergone some evolution...see the paper at OWLED.

Cheers,
Bijan.

[Danny Ayers]

On 18/11/06, Bijan Parsia <bparsia@...> wrote:

>
> On Nov 15, 2006, at 1:10 AM, Jim Hendler wrote:
> [snip]
>
> > point taken, but one would expect the uptake on the public side to
> > be continuing while the other goes on - It is unclear to me why
> > intranet adoption would favor more expressivity, woudl assume it to
> > be about the same
> >
>
> While I share the belief that intranet adoption is a perfectly fine
> rationale for adding (or removing) something, I would like to point
> out that I certainly don't think that the extra expressivity is being
> driven by "intranet" adoption...at least, I sure didn't see it that
> way. The person who, afaik, introduced this claim was Danny Ayers.

I asked some questions over of Bijan & Kendall's blog [1], variants on
"how many of the DL applications you've seen actually use the web?".

Kendall responded:
[[
I want to solve hard problems for my customers, and while they always
"use the Web" in that they use stuff like HTTP and XML and RDF and
SOAP, etc., none of them "use the Web" in the sense that you care
about.
]]

The sense that I care (most) about is that of timbl's definition of
the Web [2], "a universe of network-accessible information".

>  From what I recall, he thinks that the kind of life sciences users
> such as NCI, Galen, Snomed, etc. that want things like qualified
> cardinality restrictions are "effectively offline users". Kendall's
> point was, even if you *grant* this (what I think to be false)
> premise, it doesn't invalidate their needs or make them less useful
> for driving forward the semantic web.

I don't dispute that applications like these can be very useful for
driving forward the Semantic Web.

Correct me if I'm wrong, but the drive for a new version of OWL does
come from a section of the OWL user community, in particular the
segment currently using DL more or less exclusively of the current OWL
sublanguages. But any new version of OWL will have an impact beyond
this group.

Now it may be, whatever the particular motivations, that the
requirements of this group are entirely in line with those of the
broader Semantic Web community - or at least do not run counter to
those needs. Speaking personally, I wouldn't be unhappy to see a new
version of OWL that was a superset of the current version, whether or
not the extra parts were used outside of the existing DL community.

What does concern me is the possibility that any new version of OWL
might negatively impact the Semantic Web as a whole. Given the current
state of deployment of RDF and OWL, I'd suggest maintenance of
compatibility with existing specifications is very important. Given
this, statements like "OWL 1.1 does not have an RDF-compatible
semantics" may be cause for concern.

The suggestion that the increment proposed is small seems a little
disingenuous - new semantics, new concrete syntax, functional syntax
style of specification. If the difference so small, why not let it
simply be defined as another species : OWL Full, OWL DL...OWL Plus?

Let me be clear here : my personal position is not anti-OWL v.next,
it's pro-Semantic Web. I don't really have a problem with any
additions, as long as the new material works well with what we've
already got in RDF and OWL. I'm not sure of the timing either way, but
clearly something needs to be resolved now. There may be no conflict
in the proposals, and everybody can benefit. But as I've stated
before, I believe the onus should be on the proposers to convincingly
demonstrate that.

Cheers,
Danny.

[1] http://clarkparsia.com/weblog/2006/11/10/owled-2006-is-here/
[2] http://www.w3.org/DesignIssues/Architecture.html

--

http://dannyayers.com

[Bijan Parsia-3]

On Nov 18, 2006, at 1:40 AM, Bijan Parsia wrote:

> Sure. And there are large TBoxes from life sciences (for which the  
> EL++ fragment was developed...but EL++ includes qualified number  
> restrictions...so we sort of have to add them to the whole).
[snip]
> Of course, none of us, at all, have done a formal market survey, so  
> we're all giving impressions. Impressions are better at determining  
> positives rather than negatives. We all hear, "rational subsets!"  
> but perhaps you didn't hear, "Qualified number restrictions" (and,  
> in EL++, they unite).
[snip]

SIgh. Ian points out to me that I got confused. EL++ does not include  
QCRs, but it does include a form of complex role inclusions (which  
was a motivation for adding them to OWL 1.1).

Mes apologies.

Cheers,
Bijan.

[Jim Hendler]

At 9:39 AM +0000 11/19/06, Bijan Parsia wrote:

>On Nov 18, 2006, at 1:40 AM, Bijan Parsia wrote:
>
>>  Sure. And there are large TBoxes from life sciences (for which the
>>EL++ fragment was developed...but EL++ includes qualified number
>>restrictions...so we sort of have to add them to the whole).
>[snip]
>>  Of course, none of us, at all, have done a formal market survey,
>>so we're all giving impressions. Impressions are better at
>>determining positives rather than negatives. We all hear, "rational
>>subsets!" but perhaps you didn't hear, "Qualified number
>>restrictions" (and, in EL++, they unite).
>[snip]
>
>SIgh. Ian points out to me that I got confused. EL++ does not
>include QCRs, but it does include a form of complex role inclusions
>(which was a motivation for adding them to OWL 1.1).
>
>Mes apologies.
>
>Cheers,
>Bijan.

Bijan - for those of us who aren't OWL insiders, how about a pointer
to this EL++ thing, whatever it is.
  -JH
--
Prof James Hendler hendler@...
Dept of Computer Science http://www.cs.umd.edu/~hendler
AV Williams Bldg 301-405-2696 (work)
Univ of Maryland 301-405-6707 (Fax)
College Park, MD 20853 USA

[Alan Ruttenberg-2]

On Nov 19, 2006, at 9:30 AM, Jim Hendler wrote:

>
> Bijan - for those of us who aren't OWL insiders, how about a  
> pointer to this EL++ thing, whatever it is.


http://owl1_1.cs.manchester.ac.uk/tractable.html#2
http://lat.inf.tu-dresden.de/%7Eclu/papers/archive/ijcai05.pdf
http://lat.inf.tu-dresden.de/research/papers/2006/BaaLutSun-IJCAR-06.pdf

-Alan

[Bijan Parsia-3]

On Nov 19, 2006, at 5:05 PM, Alan Ruttenberg wrote:

> On Nov 19, 2006, at 9:30 AM, Jim Hendler wrote:
>
>> Bijan - for those of us who aren't OWL

1.1? I can never think of you as *not* an OWL insider :)

It's not so much "insiderness" here. None of the OWL 1.1 people are  
involved in the development of the EL family (or tools for it, though  
that will change...).

>> insiders, how about a pointer to this EL++ thing, whatever it is.
>
>
> http://owl1_1.cs.manchester.ac.uk/tractable.html#2
> http://lat.inf.tu-dresden.de/%7Eclu/papers/archive/ijcai05.pdf
> http://lat.inf.tu-dresden.de/research/papers/2006/BaaLutSun- 
> IJCAR-06.pdf


Thanks Alan.

I'll also point out the EL+ reasoner Cel:
        http://lat.inf.tu-dresden.de/systems/cel/

(other publications on that page).

There was a company at ISWC that claimed a complete EL++ reasoner as  
a product, but I can't find it on the Web.

Also, to my not *huge* surprise, I read last night that EL at least  
can be reduced to datalog. See chapter 2 of Yevgeny Kazakov's thesis:
        http://www.mpi-inf.mpg.de/~ykazakov
He did a prototype implementation as well which did pretty well (and  
he's not a razzle-dazzle programmer, by any means).

Cheers,
Bijan.

[Alan Ruttenberg-2]

On Nov 19, 2006, at 4:24 PM, Bijan Parsia wrote:

There was a company at ISWC that claimed a complete EL++ reasoner as a product, but I can't find it on the Web.

http://www.arity.com/index3.html

-Alan

[Bijan Parsia-3]

On Nov 19, 2006, at 9:37 PM, Alan Ruttenberg wrote:

> On Nov 19, 2006, at 4:24 PM, Bijan Parsia wrote:
>
>> There was a company at ISWC that claimed a complete EL++ reasoner  
>> as a product, but I can't find it on the Web.
>
> http://www.arity.com/index3.html

Yeah, but I can't see specifically on that page a mention of EL++.

I don't care much for that flash website...spent way too much time  
splunking to no good end!

Cheers,
Bijan.