Re: Real examples for Conneg?

On 11/4/19 2:19 AM, Nicholas Car wrote:
> Hi Karen,
> 
> As per the agenda notices for this week's Conneg meeting that I just sent out, we are reviewing 3 x real implementations of Conneg by P:
> 
> * my pyLDAPI toolkit
>   * https://github.com/rdflib/pyLDAPI
>   * exemplified in the Media Types service: https://conneg.info/mediatypes-service/
>   * writeup: https://conneg.info/mediatypes/connegp 
>   * implements all 4 Functional Profiles
> 
> * Rob's OCG specifications implementation
>   * implements the QSA FPs
> 
> * my PHP ConnegP tookit
>   * https://github.com/nicholascar/php-connegp
>   * exemplified in the Aust Gov Linked Data WG's Registers, e.g. http://linked.data.gov.au/org/
>   * writeup http://www.linked.data.gov.au/org/connegp.html 
>   * implements HTTP FP only

To keep information together, could links like these be placed in a
readme file in the Conneg github folder? I had looked in there hoping to
find this kind of thing, and it does seem to be a logical place to look.
I assume that eventually this will all be part of the implementation
report, but in the meanwhile let's make it available to all members of
the working group as they may need this information to make decisions.
Thanks.

> 
> So there are now many working examples (see the writeup links above).
> 
> I'm not sure it's sensible to include these implementations in place of examples in the spec though. The implementations may not be alive as long as the spec and I don't see other W3C specs using implementations as examples.

What I had in mind was perhaps best for an appendix or a primer; not in
place of the examples in the spec, but in addition to (however that
makes sense). To be clearer, I can point to the examples in the original
RDF primer [1]. If you look at Figure 1 in that document you can see
possibly made-up but realistic data, where the name of the person is
"Eric Miller" (it could be any normal-sounding name) and not
"personNameX", and the mailbox is "em@w3.org" and not "mailboxX". The
"realness" of the examples helps make their meaning clear to the reader,
which things that look like variable names do not.

I don't think this has anything to do with PROF, or at least that
doesn't jump out at me when reading the spec. Presumably the examples do
not require PROF.

Thanks,
kc

[1] https://www.w3.org/TR/rdf-primer/

> 
> Did you perhaps just mean profile URIs though? We have started to make a catalogue of profiles described using PROF with the specific entries motivated by necessity for the Conneg implementations:
> 
> * https://w3id.org/profile/
> 
> Currently only Alternates View profile (https://w3id.org/profile/alternates) is written up using PROF and is available as Linked Data. It is used in QSA implementations of Conneg. The other two profiles are incomplete.
> 
> It's really a job for work on PROF, which is taking a back seat until Conneg's at the next stage (3PWD), to demonstrate/deliver more fancy profiles and this URIs for them, not Conneg.
> 
> I expect a primer on Profile description + Conneg for Profiles, perhaps after Conneg & PROF are complete, might showcase better examples. If Profile Guidance gets completed, it too might include better examples.
> 
> Nick
> 
> 
> 
> 
> 
> 
> On 3/11/19, 9:54 pm, "Karen Coyle" <kcoyle@kcoyle.net> wrote:
> 
>     I'm wondering if there are any real examples that can either be used in
>     Conneg or that could accompany the document. I find the use of dummy
>     profile names like "profileX", which are not incorrect, to also not be
>     terribly information rich. I know there is a risk that a few real
>     profiles won't fully express the range of profiles possible, but maybe
>     they can be shown as examples of examples? Or in an informal primer/user
>     guide?
>     
>     Thanks,
>     kc
>     -- 
>     Karen Coyle
>     kcoyle@kcoyle.net http://kcoyle.net
>     skype: kcoylenet
>     
>     
> 

-- 
Karen Coyle
kcoyle@kcoyle.net http://kcoyle.net
skype: kcoylenet

Received on Monday, 4 November 2019 15:12:29 UTC