- From: Holger Knublauch <holger@topquadrant.com>
- Date: Sun, 06 Sep 2015 12:52:03 +1000
- To: public-data-shapes-wg@w3.org
On 9/6/15 11:23 AM, Karen Coyle wrote: > Do we need an issue (or a time slot in Lille) to at least discuss how > we will determine the core? As yet, it doesn't seem that the group has > addressed this. And my efforts in this area haven't been successful. Yes I would appreciate that too. So far the process that lead to the current set of core built-ins was driven by the original requirements/use cases catalog. The WG filtered out several tickets that didn't have enough support (including several that I suggested myself) early in the process. Of course this was only ever meant to be a starting point, yet if someone wants to add other built-ins then they should raise an ISSUE for them similar to the several tickets that I raised recently. > > kc > > On 9/5/15 8:01 AM, Arnaud Le Hors wrote: >> The WG decides what goes into core. If anything is considered missing we >> can add it. >> >> Holger's preference was to start with a very small core set of built-ins >> and let the community build libraries of templates as they see fit. The >> editor's draft reflects this approach although we've already added >> several features to the core. >> >> So, we can and should extend the core to address our use cases. I hope >> the development of the test suite and the user-friendly syntax will help >> identify gaps we ought to fill in. Now that we have a macro mechanism, it is actually quite trivial to add any new built-in. Arnaud, I believe the way that you present things above is not reflecting what happened. I am fine with adding new features, yet I thought that being conservative would be in the best interest of those who plan to not use SPARQL for their engine implementations. Likewise, we can basically forget about the Compact Syntax if we continue to call all these extra features "Core". Finally, no matter how many built-ins we add, we will never reach a sufficient coverage of our (TQ customers') use cases. To me the only remaining question is whether the WG should spawn off a second library/namespace for this extended Core, because once we open the floodgates (and do the FPWD) there will be a constant stream of new feature requests that may require a more flexible workflow - a living standard. This top-down design-by-committee with a hard-coded library of features is so last century. Holger >> -- >> Arnaud Le Hors - Senior Technical Staff Member, Open Web Technologies - >> IBM Software Group >> >> >> Karen Coyle <kcoyle@kcoyle.net> wrote on 09/04/2015 08:14:56 PM: >> >> > From: Karen Coyle <kcoyle@kcoyle.net> >> > To: "public-data-shapes-wg@w3.org" <public-data-shapes-wg@w3.org> >> > Date: 09/04/2015 08:14 PM >> > Subject: Re: Shall we add sh:InversePropertyScope and >> sh:AllObjectsScope? >> > >> > So would this be in the core vocabulary? Because I've been given the >> > impression that the core vocabulary is full - since the answer to my >> > requirements is: >> > >> > On 8/31/15 10:47 PM, Holger Knublauch wrote: >> > > >> > > SHACL can certainly express all this, but maybe not with its Core >> > > Vocabulary. It's still SHACL though. >> > >> > How do some things get into core and others do not? There are at >> least >> > two requirements that have come from the DCMI community that are >> > essential but that we've been told cannot be in the core. How does >> this >> > get decided? >> > >> > kc >> > >> > On 7/17/15 8:43 PM, Holger Knublauch wrote: >> > > Peter, >> > > >> > > in the minutes about ISSUE-62 I see you stated "I would like to >> be able >> > > select objects of a property, not just subjects". We currently >> have [1] >> > > sh:PropertyScope, which selects all subjects for a property. Is my >> > > understanding correct that you'd like to see something like >> > > sh:InversePropertyScope too? >> > > >> > > Also, we have sh:AllSubjectsScope - should I also add >> sh:AllObjectsScope >> > > (which would exclude literals)? >> > > >> > > Both are trivial to add and they seem to make sense, if only for >> symmetry. >> > > >> > > Thanks >> > > Holger >> > > >> > > [1] http://w3c.github.io/data-shapes/shacl/#PropertyScope >> > > >> > > >> > >> > -- >> > Karen Coyle >> > kcoyle@kcoyle.net http://kcoyle.net <http://kcoyle.net/> >> > m: 1-510-435-8234 >> > skype: kcoylenet/+1-510-984-3600 >> > >> >
Received on Sunday, 6 September 2015 02:52:37 UTC