- From: Holger Knublauch <holger@topquadrant.com>
- Date: Mon, 19 Oct 2015 09:56:48 +1000
- To: public-data-shapes-wg@w3.org
- Message-ID: <562431C0.10305@topquadrant.com>
On 10/18/2015 7:28, Karen Coyle wrote: > I'd be more in favor of this solution if I saw evidence that there > will be a compact syntax that provides this. FWIW attached is a screenshot of TopBraid Composer's dialog to create properties. It uses [0..*] etc while producing the currently discussed RDF triples. Just mentioning this to highlight that there are other means of editing SHACL than via Turtle source code. All kinds of Compact Syntaxes can be produced on-the-fly as "views" without requiring changes to the RDF data model. Holger > > Note that the requirement for min/max reads: > > 4.2.1 R5.2: Property Min/Max Cardinality > > The stated values for a property may be limited by minimum/maximum > cardinality, with typical patterns being [0..1], [1..1], [0..*] and > [1..*]. > > If this wasn't to be included in SHACL as stated, then the group > should have discussed and accepted this change. If we had, it would > have been solved by now. Unfortunately, it is now "carved in code" > which makes changing more difficult. > > I'm working on a table comparing the requirements to the SHACL draft > (which will need input from others).[1] I'd like to make clear what is > in the requirements that is different from the draft, and what is in > the draft that is not in the requirements. The requirements state the > mind of the group at that time, and we can discuss additions or > changes, but we should keep them in the process. > > kc > [1]https://docs.google.com/document/d/1whx2DeJtng-WZXo2DAHc_GZL7ElXNS_B8fBxarGA-0o > - in a Google doc now because I didn't want to put it on the wiki > until i had experimented a bit. has requirements, some SHACL > functions, lots of question marks > > On 10/15/15 2:02 PM, Holger Knublauch wrote: >> I believe having a way to specify unbounded maxCount and open shapes in >> RDF triples is unnecessary and has more costs than benefits. We should >> keep the language consistent - if nothing is stated then nothing is >> constrained. If we do such defaults for maxCount then the same would be >> expected for all other constraint types. However, other syntaxes and >> user interface tools may display this information, i.e. your option 3). >> >> Proposal: Close ISSUE-102 stating that no changes to the RDF data model >> of SHACL are done but other languages such as a Compact Syntax may >> include things like [0..*] >> >> Holger >> >> >> >
Attachments
- image/png attachment: TBC-CreatePropertyDialog.png
Received on Sunday, 18 October 2015 23:57:25 UTC