Re: link relations and links in general

[Erik Wilde <dret@xxxxxxxxxxxx>]

Hadrien Gardeur wrote:
> 2009/5/15 Julian Reschke <julian.reschke@xxxxxx 
> <mailto:julian.reschke@xxxxxx>>
>     On the contrary, we want to *harmonize* rel values used in HTML,
>     Atom, and in HTTP Link headers!
> It'll require a very different effort to harmonize rel values between 
> HTML and Atom, as we'll have to redefine the scope of some of these 
> values. For example rel="edit" might be useful in the header of a XHTML 
> page, but not as defined in RFC5023 
> <http://tools.ietf.org/html/rfc5023>. On the other hand, some of these 
> values such as previous/next/first/last can easily be defined within any 
> context.

which is why i though maybe it would be interesting to try to identify 
"universal values", which are registered and well-defined across 
representations, as well as "format-specific values", which could 
explicitly be related to specific representations via mechanisms such as 
QNames or CURIEs (with the usual considerations about how badly both of 
these things are often handled in implementations).

cheers,

dret.