Re: Re: mod_enclosures proposal

On Sun, 28 Nov 2004 23:06:19 -0000, [email protected]
<[email protected]> wrote:
> 
> 
> --- In [email protected], Danny Ayers <danny.ayers@g...> wrote:
> > On Sun, 28 Nov 2004 17:55:42 +0100, Reto Bachmann-Gmuer <yahoo@g...>
> 
> 
> wrote:
> >
> > > Well, why should the "simple" xml crowd choose RSS 1.0 anyway?
> >
> > Because it's (out) there - most desktop aggregators do support RSS
> > 1.0, many publishers too, even if there's no general RDF in sight. I
> > suspect this is declining though, following an astonishing amount of
> > publicity for RSS 2.0 and Atom and, well, practically zero for RSS
> > 1.0. As a lot of this data is already being published/consumed, it
> > would be nice to make good use of it (IMHO).
> 
> That's why RSS 1.0 should capitalize on its main advantage - being
> RDF-based. It should not try to be as simple as possible - this is the
> way RSS 2.0 have chosen. It should try to be as powerful and flexible
> as possible. RSS feeds are intended for machine processing.
> Readability should not be main goal. THen, people who need to publish
> complex structured information will look towards RSS 1.0.

Agreed.
 
> This way, I think RSS 1.0 should not be shy to use full power of
> RDF. 

Agreed, noting that this power is largely independent of the syntax.

Trying to handle RSS 1.0 feeds as XML data by imposing RDF
> serialization constraints is not right approach, IMHO.

It might not be the right approach, but it's the approach most
developers have taken. If you look at the tools available for using
RSS (e.g. the CMSs in [1], the aggregators in [2]), only a *tiny*
proportion of them support unconstrained RDF/XML. So if you dropped
the serialization constraints then the vast majority of RSS tools
simply couldn't use the data.

Perhaps things might have been better had RSS 1.0 originally been
specified as general RDF/XML, who knows. But right now RSS is deployed
as constrained. Without RDF, RSS itself is incredibly dull - the same
kind of applications could be built using a small subset of HTML with
the polled delivery system. But RDF doesn't necessarily mean RDF/XML
everywhere (In fact over on www-rdf-interest recent consensus seems to
be that XML developers find the syntax offputting, and that Turtle [3]
is a better pedagogical RDF syntax).

So I believe there is a good case for using constrained syntax as long
as there isn't any significant compromise over what's being expressed.
In the particular instance of <enclosure>, I don't think there is. The
simple form seems to work semantically, and is also likely to be
usable by XML-oriented RSS developers. The more decorated form (with
Alts) can be used as well, as long as you bear in mind that only a
tiny proportion of consumers will currently be able to make sense of
the information.

I do like Bill's suggestion of using seeAlso's with qualified targets
(a lot), but the question is whether it would get deployed by a
significant number of XML-oriented RSS developers.

The thing is the approach to <enclosure> is only an either-or choice
for the Really Simple developers. RDF-oriented developers can use the
simple syntax, the form with Alts as well as seeAlsos.  All of 'em
together or individually per-case. Whatever. But if a simple form
isn't available, the simple tools (i.e. the vast majority) simply
won't support it.

Cheers,
Danny.

[1] http://www.opensourcecms.com/
[2] http://www.lights.com/weblogs/rss.html
[3] http://www.ilrt.bris.ac.uk/discovery/2004/01/turtle/

-- 

http://dannyayers.com


 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/rss-dev/

<*> To unsubscribe from this group, send an email to:
    [email protected]

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/