Ok, but I'd like to understand first if the in-line creation is considered so relevant to deserve a specific treatment, or a sort of shortcut that can be eliminated if it is discovered to be a source of complication. Personally, I'm inclined to the second option (KIS)

There is another open issue, that is the flat inclusion of mixins in the xml.

Next week I terminate my course, and come back to play with xml!


2013/11/29 Jean Parpaillon <jean.parpaillon@free.fr>
Hi Augusto !

Le mardi 12 novembre 2013 à 14:58 +0100, Augusto Ciuffoletti a écrit :
> From a very intuitive (i.e., unformal) point of view, it is a very
> compact way to describe an resource together with outgoing links:
> think of a graph representation. I agree with you that this offers an
> embarrassing alias for the operation of creating link, clearly more
> convenient than the "official" way. I decided to use this alternative
> exactly for that reason: it is very attractive, so probably the most
> frequently used. But I found the same point you did: once created, I
> have no reference for the link.
>

One solution could be to require the creation of a "real" link when user
includes it inline. The server must then return a description of created
resources, with id for them.


Jean

>
> So here is the question: is the "in line" creation of a link (as in
> the http rendering paper, sect. 3.4.5) conceptually acceptable (and
> why, we may have overlooked some relevant detail), or not?
>
>
>
> 2013/11/12 Jean Parpaillon <jean.parpaillon@free.fr>
>         Hi Augusto,
>
>
>         Le 12/11/2013 11:40, Augusto Ciuffoletti a écrit :
>         > ...
>         >
>         >     >
>         >     > -) inline links are introduced in the "infrastructure"
>         document. I
>         >     hope
>         >     > that my interpretation is correct. It is true that in
>         this way a
>         >     link is
>         >     > hidden inside the resource. So this is a critical
>         issue in the
>         >     > infrastructure document that should be resolved, or
>         simply clarified
>         >     >
>         >
>         >     Could you point the section where it appears ? I can not
>         find it.
>         >
>         >
>         > Sorry, it was in the "http rendering" document. This is the
>         link in
>         > March 2013 revision:
>         > GFD-P-R.185
>         > March 22, 2013
>         > 3.4.5 Handling Link instances
>         >
>
>
>         Ok, thanks. From what I can understand, an inline link can not
>         be
>         retrieved in the general links collection. This looks weird
>         for me but I
>         have not concrete case in mind.
>
>         > ...
>
>         Regards,
>         --
>         Jean Parpaillon
>         Open Source Consultant
>         Phone: +33 6 30 10 92 86
>         im: jean.parpaillon@gmail.com
>         skype: jean.parpaillon
>         linkedin: http://www.linkedin.com/in/jeanparpaillon/en
>
>         --
>         Augusto Ciuffoletti
>         Dipartimento di Informatica
>         Università di Pisa
>         56100 - Pisa (Italy)

--
Jean Parpaillon
Open Source Consultant
Phone: +33 6 30 10 92 86
im: jean.parpaillon@gmail.com
skype: jean.parpaillon
linkedin: http://www.linkedin.com/in/jeanparpaillon/en




--
Augusto Ciuffoletti
Dipartimento di Informatica
Università di Pisa
56100 - Pisa (Italy)