On 26/1/05 5:02 PM, "Asbjørn Ulsberg" <[EMAIL PROTECTED]> wrote:
>> How is a resource which shows the last 15 entries as of today an
>> "alternative" representation of an entry which was published six months
>> ago and has long slipped out of the sliding window?
>
> It isn't, and that's not what
On Fri, 21 Jan 2005 17:35:05 +1100, Eric Scheid
<[EMAIL PROTECTED]> wrote:
How is a resource which shows the last 15 entries as of today an
"alternative" representation of an entry which was published six months
ago and has long slipped out of the sliding window?
It isn't, and that's not what
On 20/1/05 3:46 AM, "Arve Bersvendsen" <[EMAIL PROTECTED]> wrote:
>>> a) Change rel="alternate" to rel="feed", rel="subscription" or similar.
>>
>> On one (technical) level both make sense. However given that the
>> current version is pretty widely deployed, a) at least could be
>> counter-produ
On 21/1/05 1:07 PM, "Asbjørn Ulsberg" <[EMAIL PROTECTED]> wrote:
>> a) Change rel="alternate" to rel="feed", rel="subscription" or similar.
>
> What I have a problem with is «what is the feed for if it isn't an
> alternative to the resource you're looking at?».
How is a resource which shows the
On Wed, 19 Jan 2005 14:40:17 +0100, Arve Bersvendsen <[EMAIL PROTECTED]>
wrote:
a) rel="alternate" is a misnomer. Not all feeds are alternate content.
They may have some other relationship, such as rel="related".
I agree.
b) Specifying a mime type for a feed is not always appropriate. If we
On 20/1/05 3:49 AM, "Arve Bersvendsen" <[EMAIL PROTECTED]> wrote:
>>> b) Drop 'type' as a required attribute for link. Make it a 'MAY'.
>>
>> @type isn't a required attribute.
>
> In the Atom autodiscovery document I referred to, type is required:
mea culpa
> 4.2 type attribute
> The type
On Wed, 19 Jan 2005 15:52:22 +0100, Danny Ayers <[EMAIL PROTECTED]>
wrote:
II) The MIME type may differ, depending on the client requesting it.
Re. II) I'm not so sure about this, it depends somewhat on how you
interpret the element - I think an interpretation consistent with
webarch could be: th
On Thu, 20 Jan 2005 02:17:35 +1100, Eric Scheid
<[EMAIL PROTECTED]> wrote:
b) Drop 'type' as a required attribute for link. Make it a 'MAY'.
@type isn't a required attribute.
In the Atom autodiscovery document I referred to, type is required:
4.2 type attribute
The type attribute MUST be pr
On 20/1/05 2:29 AM, "Danny Ayers" <[EMAIL PROTECTED]> wrote:
>> +1 for "feed", for the link to the resource which is a "sliding window
>> representation" of the most recently updated content from which the current
>> resource was once published via. Don't use "feed" for links to Atom Feed
>> Docu
On Thu, 20 Jan 2005 02:17:35 +1100, Eric Scheid
<[EMAIL PROTECTED]> wrote:
> +1 for "feed", for the link to the resource which is a "sliding window
> representation" of the most recently updated content from which the current
> resource was once published via. Don't use "feed" for links to Atom F
On 20/1/05 12:40 AM, "Arve Bersvendsen" <[EMAIL PROTECTED]> wrote:
>
> I have just reread the Atom autodiscovery protocol, at
> http://diveintomark.org/rfc/draft-pilgrim-atom-autodiscovery-02.html>.
> I have also read http://www.w3.org/Provider/Style/URI>, and I believe
> the autodiscovery proto
On Wed, 19 Jan 2005 14:40:17 +0100, Arve Bersvendsen
<[EMAIL PROTECTED]> wrote:
>
> I have just reread the Atom autodiscovery protocol, at
> http://diveintomark.org/rfc/draft-pilgrim-atom-autodiscovery-02.html>.
> I have also read http://www.w3.org/Provider/Style/URI>, and I believe
> the autodis
I have just reread the Atom autodiscovery protocol, at
http://diveintomark.org/rfc/draft-pilgrim-atom-autodiscovery-02.html>.
I have also read http://www.w3.org/Provider/Style/URI>, and I believe
the autodiscovery protocol needs revisiting:
a) rel="alternate" is a misnomer. Not all feeds are
13 matches
Mail list logo
