My votes below. Another question: does the new Enhanced Content Model functionality expect to find or rely on this triple?
> (a) when ingesting content model objects, should we enforce a RELS-EXT > assertion to a valid content model for content model objects? or Yes. _Any_ Fedora object should have a content model discoverable through that same old hasModel relationship, and that includes content models themselves. > (b) should we create a Resource Index triple identifying the > fedora-system:ContentModel-3.0 as a default for content model objects when > none is specified in RELS-EXT? No. The RI should reflect as closely as possible the actual state of the objects, if for no other reason than so that folks like Bill can turn up the kinds of things he just did (which started this whole conversation). > (c) should we stop CMA features working (eg the dissemination execution) if > the object identified as the content model does not itself identify through > RELS-EXT that it is a content model object? Yes. If the CM in question doesn't identify itself as a CM, then the only way we know that it is one is when a disseminator gets executed "through" it. (Or until any use of Enhanced Content Model functionality that might turn up the same fact.) That strikes me as unpleasantly "dynamic" for the kinds of purposes Fedora meets. While that kind of "duck-typing" in the repository could be very useful, it could be very dangerous, and I suspect that most Fedora users will more likely find it to be the latter. --- A. Soroka Digital Research and Scholarship R & D the University of Virginia Library On Oct 28, 2010, at 11:11 AM, Steve Bayliss wrote: > Hi Bill > > Thanks for that - particularly confirming that dissemination executions work > even when the content model object doesn't assert its own content model > correctly. > > I wonder if we should add some kind of validation for this - particularly as > there could be an expectation that one should be able to run these sorts of > queries to determine what content models exist. Or whether we we should (in > a similar manner to the "default" object content model) behind-the-scenes > create this triple identifying the default content model for content model > objects. > > So general questions for the list: > > (a) when ingesting content model objects, should we enforce a RELS-EXT > assertion to a valid content model for content model objects? or > (b) should we create a Resource Index triple identifying the > fedora-system:ContentModel-3.0 as a default for content model objects when > none is specified in RELS-EXT? > (c) should we stop CMA features working (eg the dissemination execution) if > the object identified as the content model does not itself identify through > RELS-EXT that it is a content model object? > > Or.... (d) do nothing! > > (I think there are probably issues at least with a and b - how do we tell > that an object is a content model object? Surely the RELS-EXT assertion is > the way to do that, but the scenario is one where this assertion is not > present.) > > Regards > Steve > -----Original Message----- > From: Bill Parod [mailto:bill-pa...@northwestern.edu] > Sent: 28 October 2010 15:56 > To: Support and info exchange list for Fedora users. > Subject: Re: [fcrepo-user] Cmodel discovery? > > Steve, > > On Oct 28, 2010, at 3:09 AM, Steve Bayliss wrote: > >> All Content Model objects *should* contain the triple that Adam has >> identified, the triple isn't generated automatically (however if an object >> declares no content model, then a triple will be generated identifying the >> "default" content model - info:fedora/fedora-system:FedoraObject-3.0). >> > > Thanks very much for confirming the use of <fedora-model:hasModel > rdf:resource="info:fedora/fedora-system:ContentModel-3.0"/> as convention. > > Some of our cmodels do not assert it, so I will update them to do so. > >> I'm not sure if the CMA works correctly if a Content Model object fails to >> declare that triple, perhaps that's what is going on in this case? It would >> be useful to confirm if (a) the RELS-EXT of the CModel objects do indeed >> fail to declare that triple and (b) if any objects belonging to that CModel >> "work" correctly wrt object methods etc. >> > > Yes, they do work, at least as regards CMA binding for dissemination > execution. > >> If you want to query for content models *in use* by objects in the >> repository, you could use the query >> >> select $o >> from <#ri> >> where $s <info:fedora/fedora-system:def/model#hasModel> $o >> >> and use the "Force Distinct" option in the risearch web interface (or the >> distinct=on query parameter if calling the risearch service directly). >> > > That's what I'm doing now, though since your confirmation that cmodel objects > should assert <fedora-model:hasModel > rdf:resource="info:fedora/fedora-system:ContentModel-3.0"/>, I'll update all > our cmodels to do so. I'm then inclined to use Adam's original query > suggestion as it seems more efficient: > > select $cmodel > from <#ri> > where $cmodel <fedora-model:hasModel> > "info:fedora/fedora-system:ContentModel-3.0" > > Thanks very much, > > - Bill > > > > >> Steve >> >>> -----Original Message----- >>> From: aj...@virginia.edu [mailto:aj...@virginia.edu] >>> Sent: 22 October 2010 16:10 >>> To: Support and info exchange list for Fedora users. >>> Subject: Re: [fcrepo-user] Cmodel discovery? >>> >>> >>> Bill-- >>> >>> Perhaps we're seeing CModels created by different means >>> ending up with different RELS-EXT contents. I'm seeing that >>> triple in all of our CModels, but because some of them were >>> taken from external sources (e.g. a JPEG2000 CModel from the >>> LANL Djatoka project) I don't know whether this triple is >>> normatively produced or is just a matter of convention. >>> >>> Perhaps one of the Fedora developers can speak to this? >>> >>> --- >>> A. Soroka >>> Digital Research and Scholarship R & D >>> the University of Virginia Library >>> >>> >>> >>> On Oct 22, 2010, at 10:59 AM, Bill Parod wrote: >>> >>>> Hi Adam, >>>> >>>> I tried it (with slightly different syntax): >>>> select $cmodel >>>> from <#ri> >>>> where $cmodel <fedora-model:hasModel> >>> <info:fedora/fedora-system:ContentModel-3.0> >>>> >>>> and found some of our cmodels but not all, which leads me >>> to think that you've not only provided a query solution but >>> revealed something some of our cmodels RELS-EXT are lacking: >>>> >>>> <fedora-model:hasModel >>> rdf:resource="info:fedora/fedora-system:ContentModel-3.0"/> >>>> >>>> Is that indeed the convention? >>>> >>>> Thanks much! >>>> >>>> Bill >>>> >>>> >>>> >>>> >>>> On Oct 22, 2010, at 9:40 AM, aj...@virginia.edu wrote: >>>> >>>>> Bill-- >>>>> >>>>> If I'm not mistaken, you'll find an RDF statement: >>>>> >>>>> <fedora-model:hasModel >>> rdf:resource="info:fedora/fedora-system:ContentModel-3.0"/> >>>>> >>>>> in every cModel object. Might that be a query that can do >>> the job? Something like: >>>>> >>>>> select $cmodel >>>>> from <#ri> >>>>> where $cmodel <fedora-model:hasModel> >>> "info:fedora/fedora-system:ContentModel-3.0" >>>>> >>>>> >>>>> --- >>>>> A. Soroka >>>>> Digital Research and Scholarship R & D >>>>> the University of Virginia Library >>>>> >>>>> >>>>> >>>>> On Oct 22, 2010, at 10:34 AM, Bill Parod wrote: >>>>> >>>>>> I'm working on a repository introspection service that >>> provides a repository profile from the point of view of >>> cmodels and services. The motivation is to obtain a concise >>> picture of what's in the repository, provide export tools to >>> aid in selective migration, and to generate model/service >>> documentation from the models and services themselves. >>>>>> >>>>>> My question is this: Is there a way, using Fedora REST >>> APIs or triple store queries to obtain a list of the pids of >>> cmodel objects in the repository? >>>>>> >>>>>> I'm currently using the triple store to query for cmodel >>> / service pairings: >>>>>> select $cmodel $service >>>>>> from <#ri> >>>>>> where $cmodel <fedora-model:hasService> $service >>>>>> >>>>>> But that will only find cmodels that have associated >>> services. Ours tend to fall into that category but I'd like >>> to do it in a way that is more general. I'm not very >>> experienced with using the tripe store, so I'm hoping there's >>> a way there to do this, though that seems doubtful from what >>> I'm seeing in cmodel RELS-EXT. >>>>>> >>>>>> I know I can do a sql query on the database ( select >>> distinct cModel from modelDeploymentMap) , but I'd of course >>> like to avoid cutting in at that level, and I'm not sure >>> cmodels without services are represented there anyway. >>>>>> >>>>>> Is there a way to discover cmodels using the APIs or the >>> triple store? Or is it best to put something specific in >>> cmodel RELS-EXT for that purpose? That's what I'm about to >>> do, but thought I'd ask first. >>>>>> >>>>>> Thanks, >>>>>> Bill >>>>>> >>>>>> Bill Parod >>>>>> Library Technology Division - Enterprise Systems >>>>>> Northwestern University Library >>>>>> bill-pa...@northwestern.edu >>>>>> 847 491 5368 >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> >>> -------------------------------------------------------------- >>> ---------------- >>>>>> Nokia and AT&T present the 2010 Calling All >>> Innovators-North America contest >>>>>> Create new apps & games for the Nokia N8 for consumers in >>> U.S. and Canada >>>>>> $10 million total in prizes - $4M cash, 500 devices, >>> nearly $6M in marketing >>>>>> Develop with Nokia Qt SDK, Web Runtime, or Java and >>> Publish to Ovi Store >>>>>> >>> http://p.sf.net/sfu/nokia-dev2dev_____________________________ >>> __________________ >>>>>> Fedora-commons-users mailing list >>>>>> Fedora-commons-users@lists.sourceforge.net >>>>>> https://lists.sourceforge.net/lists/listinfo/fedora-commons-users >>>>> >>>>> >>>>> >>> -------------------------------------------------------------- >>> ---------------- >>>>> Nokia and AT&T present the 2010 Calling All >>> Innovators-North America contest >>>>> Create new apps & games for the Nokia N8 for consumers in >>> U.S. and Canada >>>>> $10 million total in prizes - $4M cash, 500 devices, >>> nearly $6M in marketing >>>>> Develop with Nokia Qt SDK, Web Runtime, or Java and >>> Publish to Ovi Store >>>>> http://p.sf.net/sfu/nokia-dev2dev >>>>> _______________________________________________ >>>>> Fedora-commons-users mailing list >>>>> Fedora-commons-users@lists.sourceforge.net >>>>> https://lists.sourceforge.net/lists/listinfo/fedora-commons-users >>>> >>>> Bill Parod >>>> Library Technology Division - Enterprise Systems >>>> Northwestern University Library >>>> bill-pa...@northwestern.edu >>>> 847 491 5368 >>>> >>>> >>>> >>>> >>>> >>> -------------------------------------------------------------- >>> ---------------- >>>> Nokia and AT&T present the 2010 Calling All >>> Innovators-North America contest >>>> Create new apps & games for the Nokia N8 for consumers in >>> U.S. and Canada >>>> $10 million total in prizes - $4M cash, 500 devices, nearly >>> $6M in marketing >>>> Develop with Nokia Qt SDK, Web Runtime, or Java and Publish >>> to Ovi Store >>>> >>> http://p.sf.net/sfu/nokia-dev2dev_____________________________ >> __________________ >>> Fedora-commons-users mailing list >>> Fedora-commons-users@lists.sourceforge.net >>> https://lists.sourceforge.net/lists/listinfo/fedora-commons-users >> >> >> ---------------------------------------------------------------------------- >> -- >> Nokia and AT&T present the 2010 Calling All Innovators-North America contest >> Create new apps & games for the Nokia N8 for consumers in U.S. and Canada >> $10 million total in prizes - $4M cash, 500 devices, nearly $6M in marketing >> Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store >> http://p.sf.net/sfu/nokia-dev2dev >> _______________________________________________ >> Fedora-commons-users mailing list >> Fedora-commons-users@lists.sourceforge.net >> https://lists.sourceforge.net/lists/listinfo/fedora-commons-users >> >> >> ------------------------------------------------------------------------------ >> Nokia and AT&T present the 2010 Calling All Innovators-North America contest >> Create new apps & games for the Nokia N8 for consumers in U.S. and Canada >> $10 million total in prizes - $4M cash, 500 devices, nearly $6M in marketing >> Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store >> http://p.sf.net/sfu/nokia-dev2dev >> _______________________________________________ >> Fedora-commons-users mailing list >> Fedora-commons-users@lists.sourceforge.net >> https://lists.sourceforge.net/lists/listinfo/fedora-commons-users > > Bill Parod > Library Technology Division - Enterprise Systems > Northwestern University Library > bill-pa...@northwestern.edu > 847 491 5368 > > > > > ------------------------------------------------------------------------------ > Nokia and AT&T present the 2010 Calling All Innovators-North America contest > Create new apps & games for the Nokia N8 for consumers in U.S. and Canada > $10 million total in prizes - $4M cash, 500 devices, nearly $6M in marketing > Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store > http://p.sf.net/sfu/nokia-dev2dev_______________________________________________ > Fedora-commons-users mailing list > Fedora-commons-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/fedora-commons-users ------------------------------------------------------------------------------ Nokia and AT&T present the 2010 Calling All Innovators-North America contest Create new apps & games for the Nokia N8 for consumers in U.S. and Canada $10 million total in prizes - $4M cash, 500 devices, nearly $6M in marketing Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store http://p.sf.net/sfu/nokia-dev2dev _______________________________________________ Fedora-commons-users mailing list Fedora-commons-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/fedora-commons-users
