Re: naming resources - Slug-Header - ISSUE-43

On 10 Jan 2013, at 10:39, Yves Lafon <ylafon@w3.org> wrote:

> On Tue, 8 Jan 2013, Henry Story wrote:
> 
>> 
>> On 8 Jan 2013, at 11:54, Roger Menday <roger.menday@uk.fujitsu.com> wrote:
>> 
>>> 
>>> hi Henry,
>>> 
>>> An alternative proposal might be to just take the rdfs:label (or maybe from list of established vocab for 'labels') from the POSTed body as a 'hint' for a name (?)
>> 
>> That seems to be a compatible proposal.
> 
> Well, as soon as you have two ways of doing the same thing, you end up dealing with conflicts (setting precedence, be sure that a client that uses only RDF is not ignoring the SLUG header etc...). It is far better to avoid that kind of duplicates.


Yes, but you can't avoid it for non-RDF content. It seems compatible,
modulo the issues I brought up below, that would need to be considered
more carefully, but is less general.

The issues of precedence I think were discussed on the Atom-protocol mailing
list ( I think that was a HUGE discussion ) They had exactly the same 
precedence issues: Atom entries contain atom:title elements and so these 
could also be used for the headers.  I think here we should ask Erik Wilde 
who has been following the Atom world much more closely than I have. 
Is it widely adopted? Where there problems etc... 

> 
> >
>> 1. The advantage of the SLUG Header is that it would work with
>> a non-rdf resource too.
>> 
>> 2. Your proposal would be to POST a document containing
>> Either
>> 
>> (a) a new vocab item
>> <> xxx:proposedTitle "card" .
>>   ...
>> 
>> (b) a well known one
>> <> dc:title "Go Seigen's Friends" .
>> 
>> using (a) requires the LDP server to have a
>> MUST support for this, because it should really then
>> remove that triple if it did not manage to create the file,
>> since that triple was intended only to guide the act
>> of file creation, not for it to be maintained during the
>> whole life of the file. I have a feeling this gets more
>> complicated that the Slug header.
>> 
>> using (b) does not seem harmful having the server use
>> some dc:title like relation to guide its name creation in case
>> there is no Slug Header. But one can see issues with people
>> using the same title across a number of files, and it does
>> not seem fine grained enough.
>> 
>> Perhaps a deciding factor would be what would happen
>> when creating a new Collection. I don't think that there
>> we'd be sending some RDF content along.
>> 
>> But perhaps before we look at all the possible proposals
>> we should just open an ISSUE so that these answers don't
>> get lost?
>> 
>> Henry
>> 
>> 
>>> 
>>> Roger
>>> 
>>>> There are many cases when creating resources the user
>>>> would like to be able if possible to control the name
>>>> of the resource created  at least the last part of it
>>>> (excluding the extension). It makes  sense for example
>>>> that a foaf profile be named "card" when created in a
>>>> collection.
>>>> 
>>>> Do I open an issue for this?
>>>> 
>>>> Since we should give proposals for when we open issues, I
>>>> will do so immediately: I propose that for this we just
>>>> adopt the SLUG-Header as defined by the Atom Protocol.
>>>> 
>>>> https://tools.ietf.org/html/rfc5023#section-9.7
>>>> 
>>>> Henry
>>>> 
>>>> Social Web Architect
>>>> http://bblfish.net/
>>>> 
>>> 
>> 
>> A short message from my sponsors: Vive la France!
>> Social Web Architect
>> http://bblfish.net/
>> 
>> 
> 
> -- 
> Baroula que barouleras, au tiéu toujou t'entourneras.
> 
>        ~~Yves
> 

A short message from my sponsors: Vive la France!
Social Web Architect
http://bblfish.net/