Re: Does xml:base apply to type="html" content?

[Sjoerd Visscher]

Anne van Kesteren wrote:


Quoting James Holderness <[EMAIL PROTECTED]>:
I think the issue of neutral link bookmarking is unlikely to be a 
problem for Atom aggregators though. Server bugs are another thing, 
but I think most feeds will be broken without an explicit xml:base 
anyway, so maybe that's not worth worrying about. I'm not sure though. 
Should the WG recommend ignoring Content-Location as a base URI, or 
should aggregators follow the RFC exactly as specified?


If `Content-Location` is not usable or can't be used consistent on a 
website

(for example, using it for both Atom and HTML content) I suggest we specify
something that is consistent with what browsers do. And perhaps try to 
obsolete

the relevant header if possible...




The problem is not with Content-Location, but with RFC 3986. It says 
that same-document references must be resolved with respect to the base 
URI. It adds that when those references are resolved, they should not 
result in a new retrieval action, but that does not help with things 
like bookmarking (as James pointed out), and is almost impossible to 
implement.


--
Sjoerd Visscher
http://w3future.com/weblog/

Re: Does xml:base apply to type="html" content?

[A. Pagaltzis]

* Anne van Kesteren <[EMAIL PROTECTED]> [2006-04-04 12:20]:
> Quoting "A. Pagaltzis" <[EMAIL PROTECTED]>:
> >+1, standards aren’t there for people to cherry-pick the parts
> >they find convenient or useful.
> 
> If interoperable implementations for standards are not
> possible, they are useless. The goal of having standards is
> interoperable implementations. Opera has removed support for
> Content-Location (or least partially, not sure of the details)
> for the same reasons as Firefox.
> 
> (This also isn't really about convenient or useful...)

Yes. I considered adding a long diatribe about standards, but
decided against it. In short, I know that standards are not an
end in themselves and that is not what I’m trying to say. Interop
is always the end goal; and that’s exactly why it’s as important
not to cherry-pick standards as it is to not follow them blindly.
It’s a tight-rope walk. Departure from a standard is necessary
sometimes but must be well considered implentor consensus,
otherwise you might as well have no standards at all. And if that
becomes best practice, it should be codified (that is the right
way to write standards, after all).

Etc etc. Let’s not go into all this; it should be clear.

I’m not principally opposed to opposition to `Content-Location`;
it just seems like a precarious proposition to have specs start
to arbitrarily limit the applicability of select other specs.
That way lies spec interaction madness in unforeseen scenarios.
If there is a real problem, the appropriate reaction needs to
become established best practice outside of specs and then should
be codified in the appropriate upstream spec. Let’s not willfully
make things even more complicated than they already are.

Regards,
-- 
Aristotle Pagaltzis // 

Re: Error in example ?

[cedric]

Le Mardi 4 Avril 2006 12:36, cedric a écrit :
> In 3.1.1.3 XHTML of
> http://www.atomenabled.org/developers/syndication/atom-format-spec.php#atom
>.documents
Please do not respond to this wrong mail
>
> There is example using  but this is deprecated in atom 1.0 I
> think.

Re: Error in example ?

[Henri Sivonen]

On Apr 4, 2006, at 13:36, cedric wrote:


In 3.1.1.3 XHTML of
http://www.atomenabled.org/developers/syndication/atom-format- 
spec.php#atom.documents


There is example using  but this is deprecated in atom 1.0  
I think.


It is not deprecated.
See: http://www.atomenabled.org/developers/syndication/atom-format- 
spec.php#element.summary


Actually, there is nothing deprecated in Atom 1.0. There are only  
things that are in the spec and things that are not in the spec.  
There is no Atom 0.3. :-) These are not the droids you are looking  
for. :-)


--
Henri Sivonen
[EMAIL PROTECTED]
http://hsivonen.iki.fi/

Error in example ?

[cedric]

In 3.1.1.3 XHTML of 
http://www.atomenabled.org/developers/syndication/atom-format-spec.php#atom.documents

There is example using  but this is deprecated in atom 1.0 I think.

Re: Does xml:base apply to type="html" content?

[Anne van Kesteren]

Quoting "A. Pagaltzis" <[EMAIL PROTECTED]>:

* James Holderness <[EMAIL PROTECTED]> [2006-04-04 03:25]:

The way I see it, until a standards body tells us otherwise, we
are obliged to support the Content-Location header unless we
can provide a very good argument for ignoring it.


+1, standards aren’t there for people to cherry-pick the parts
they find convenient or useful.


If interoperable implementations for standards are not possible, they are
useless. The goal of having standards is interoperable implementations. Opera
has removed support for Content-Location (or least partially, not sure of the
details) for the same reasons as Firefox.

(This also isn't really about convenient or useful...)


--
Anne van Kesteren

Re: Does xml:base apply to type="html" content?

[Anne van Kesteren]

Quoting James Holderness <[EMAIL PROTECTED]>:

If `Content-Location` is not usable or can't be used consistent on a website
(for example, using it for both Atom and HTML content) I suggest we specify
something that is consistent with what browsers do. And perhaps try 
to obsolete the relevant header if possible...


Isn't this something the HTTP WG should be doing?


I guess so. The HTML WG (W3C, same concept) should be doing a lot of things as
well. That doesn't mean it actually happens...


--
Anne van Kesteren