Re: Flexible schema inclusions and overloading in ShEx.

Unfortunately, it happens quite frequently in real life :)
And reshipping a new data definition + banning the older version
is usually not acceptable. (or at least extremely complicated).

If my memory is correct, some of the work about XMLSchema 1.1 was about
handling
this topic, and they basically failed to address it properly (imho :)

BTW, it may be interesting to have the opinion of XSD guys about our
discussions.

PS: btw, is James Clark still in activity? He probably has some insights
about
RelaxNG, its birth and reasons for existence.



On Tue, Jul 22, 2014 at 12:48 PM, Peter F. Patel-Schneider <
pfpschneider@gmail.com> wrote:

> I find overriding to be a very scary feature.  It is hard to define in
> this context and hard to understand.  I strongly suggest that it not be a
> part of whatever comes out of a spec for RDF constraints.
>
> Overriding does not appear to be a part of the ShEx member submission.
>
> peter
>
>
>
>
> On 07/22/2014 03:35 AM, Olivier Rossel wrote:
>
>> Hi all.
>> I would like to point out that ShEx has some ideas about a nice feature
>> coming
>> from RelaxNG:
>> "extending and overriding shapes"
>>
>> cf the discussion here
>> http://www.w3.org/2001/sw/wiki/Discussion_SHEX_format#
>> Discussion_2.2C_redefining_and_including_other_SHEX_definitions
>>
>> This is really elegant and handy feature when you deal with data
>> descriptions
>> on the long term (in standards, for example).
>>
>> I am not sure SPIN or ICV have any support for an equivalent feature.
>> True?
>>
>>
>>
>>