W3C home > Mailing lists > Public > public-semweb-lifesci@w3.org > June 2011

Re: Trying to summarise: Semantic free identifiers

From: Andrea Splendiani <andrea.splendiani@bbsrc.ac.uk>
Date: Wed, 22 Jun 2011 11:47:34 +0100
Cc: Chime Ogbuji <chimezie@gmail.com>, Jim McCusker <mccusj@rpi.edu>, James Malone <malone@ebi.ac.uk>, "andrea splendiani (RRes-Roth)" <andrea.splendiani@rothamsted.ac.uk>, Helena Deus <helenadeus@gmail.com>, Matt Vagnoni <matthew.vagnoni@uth.tmc.edu>, Michel_Dumontier <michel_dumontier@carleton.ca>, "Sivaram Arabandi, MD" <sivaram.arabandi@gmail.com>, "M. Scott Marshall" <mscottmarshall@gmail.com>, "MMVagnoni@mdanderson.org" <mmvagnoni@mdanderson.org>, HCLS <public-semweb-lifesci@w3.org>, Jonathan Rees <jar@creativecommons.org>
Message-Id: <063D7B5F-707E-4F91-B7B5-901FE196F299@bbsrc.ac.uk>
To: Pat Hayes <phayes@ihmc.us>
Hi,

Il giorno 22/giu/2011, alle ore 06.13, Pat Hayes ha scritto:
> 
>> (but primarily to biomedical ontologies - hence the relevance to this interest group mailing list) and as much as I agree with you about the issues associated with requiring opaque identifiers, this discussion is actually important (at least to me) because of the consequences of what the start of the thread was suggesting: that identifiers in the Relations Ontology will be changing.  
>> 
>> The Relations Ontology is very central to many of the recently developed biomedical ontologies and (speaking only for myself - since is the realm of my day job) is sort of a foundation for many of them.
> 
> Well, this is an aside from the theme of this thread, but if this is true then these ontologies have far worse problems than a mere change of identifiers. But leaving that aside, please tell me how having unreadable and meaningless identifiers is going to help with the (admittedly very real) problems which arise when identifiers are changed? 
I think the problem is in the other direction:
It's not about making identifiers opaque. It's about creating identifiers for concepts which don't really map to (current) names. 
In other domains you may http://example/eiffel, and you just need to disambiguate (engineer, tower, language...). After that you can rely on people sharing their knowledge of 'tower'. 
In this domain, you can say #Protein and people associate different meaning to it (sequence, obejct...) in a a way which has not a well established conceptualization. 
Of course this doesn't mean you need opaque identifers, but if you consider your are talking about 10^5 terms x ontology, that you often need to take a 'catalogue' approach, and that each time you see #Protein different people say "it's a different thing"... that's not so insensate. 
That said... I would expect that fundamental relations as part_of, which have a clear definition (at least!) don't pose problems of naming.
And opaqueness has it costs. 

But I was thinking...

The 'web' as a paradigm is (also) about immediacy. All the considerations about opaqueness of identifiers could be served as well by some other infrastructure, not necessarily RDF or the web. It seems like the idea is to use the web as publishing infrastructure. Is it just that ?

ciao,
Andrea


> 
>> The impact of such a change would be quite large.  Certainly, the ontologies I develop and the ones I use would have to undergo what I would consider a substantive revision.
> 
> How would this be any easier if it was a change of meaningless identifiers rather than a change of meaningful ones? And don't tell me that meaningless identifiers never change. 
> 
> Seems to me that the central issue is that important 'foundational' ontologies should simply **not change**: that any change should be a new version, with new identifiers. This is very much what Tim B-L was talking about with his doctrine of 'cool URIs'. Ironically, the use of SW technology - specifically, URIs and XML namespaces - is an elegant way to handle this. To revize something like the RO, leave it alone and change the root URI for the new 'version', while keeping the identifier extensions unchanged. Then a single change in the header of an XML file will be enough to switch all the RO names from their old to their new meanings. (OK, I know this is a simplification, but you get the idea.)  
> 
>> The discussion is also important because the issues that motivated the need for opaque identifiers (as I understand them) are more emphasized in this domain than in others and I'm glad (frankly) to finally have been exposed to the thinking behind them and such open conversation is a necessary step along the way to consensus (or at least to being able to do a decent job of documenting the pros and cons).  
> 
> I agree with you here, of course. 
> 
>> Me personally, I'm a little alarmed that the discussion was only triggered by the realization that the decision had already made and now I feel compelled to find the mailing list of the stewards of the RO to at least attempt to push back before having to decide to either build my own or simply stick with a pre-opaque identifier version!  
>> 
>> And don't get me started on the general topic of absurd / redux / perma threads in the SW community - at least the considerations here are not exclusively in the realm of philosophy.  I've seen worse and I think you know those that I speak of, since you have been involved in many of the ones I listed at the end of a previous email (probably unwittingly).
> 
> Oh yes, I know :-)
> 
> Pat
> 
>> On Tuesday, June 21, 2011 at 11:22 PM, Pat Hayes wrote:
>> 
>>> This entire discussion is simply absurd, if it is supposed to apply to the semantic web generally. OF COURSE people are not going to re-name the RDFS or OWL vocabulary (for example) with 'opaque' names. Programming languages are not going to use opaque identifiers for their reserved vocabularies, and people are not going to start speaking in numerical codes. This would all be riotously funny if some people did not, apparently, take it seriously.
>>> 
>>> How could such a crazy idea ever be enforced? Rest assured that the W3C is not going to re-write its standards to require opaque identifiers. If someone feels that http://whatever.com/opacity/045678723 better suits their methodology than, say, http://www.w3.org/1999/02/22-rdf-syntax-ns#domain, then by all means let them assert the equivalence using owl:sameAs. Of course, this will not actually work in OWL (the RDFS namespaces are reserved vocabulary) but no doubt the extensive Tooling which they will build will have its own special reasoners and hence be able to overcome this minor detail. 
>>> 
>>> This discussion should never have even begun in a Semantic Web forum. Save your energies for more productive discussions, such as how to reconcile the Palestine/Israel conflict. 
>>> 
>>> Pat Hayes
>>> 
>> -- 
>> Chime Ogbuji
>> Sent with Sparrow
>> 
> 
> ------------------------------------------------------------
> IHMC                                     (850)434 8903 or (650)494 3973   
> 40 South Alcaniz St.           (850)202 4416   office
> Pensacola                            (850)202 4440   fax
> FL 32502                              (850)291 0667   mobile
> phayesAT-SIGNihmc.us       http://www.ihmc.us/users/phayes
> 
> 
> 
> 
> 

Andrea Splendiani
Senior Bioinformatics Scientist
Centre for Mathematical and Computational Biology
+44(0)1582 763133 ext 2004
andrea.splendiani@bbsrc.ac.uk
Received on Wednesday, 22 June 2011 10:48:30 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:52:47 UTC