Re: Re: RSS Profile Comments

On 2 Oct 2007, at 23:19, rcade wrote:

> --- In [email protected], Geoffrey Sneddon <foolistbar@...>
> wrote:
>> 3. |hours| says, "both values to represent midnight", without
>> specifying what values represent midnight (as someone involved in
>> writing various specs, I'd prefer to spell out the blatantly  
>> obvious).
>
> The preceding sentence states that the values are 0 and 24.

I am aware of what the preceding sentence says, but I would still  
rather it be spelt out.

>> 4. |ttl| has two consecutive sentences at the start of the "Most
>> aggregators…" paragraph with the same meaning. There is no need, and
>> as it is worded differently, it takes me a minute to look back and
>> see they mean the same thing and it isn't contradicting itself.
>
> The second sentence makes specific the general statement in the first.

Can we not say something like, "Seven (out of [n] tested) aggregators  
won't check a feed more frequently than its TTL"?

>> 5. |description| says that aggregators should use the |channel| |
>> link| to resolve relative URIs, without specifying what to do if that
>> doesn't exist or is in itself relative. We should probably reference
>> RFC3986 section 5.1, with the afore mentioned element being the base
>> embedded in content.
>
> That's already covered. The channel link can't be relative, because
> it's a URL data type. The definition of URL states that it must begin
> with a URI scheme.
>
> http://www.rssboard.org/rss-profile#data-types-url

I'm am aware that in a conforming feed it can't be relative, but how  
UAs are meant to treat certain non-conforming feeds is undefined.

>> 6. |enclosure| should probably state that @type and @length exist for
>> informative purposes only, and must not be treated as the
>> authoritative MIME type and file length.
>
> I don't see anything in the spec, or in Dave Winer's use-case document
> for enclosure, that says these things. Are you suggesting them as a
> requirement or a recommendation?

As well as what James has already linked to, as a UA author, I want  
to know what I'm meant to do. Am I meant to treat @type as  
authoritative, and more important than what is sent as the Content- 
Type (as in |embed| in HTML)? Am I meant to throw an error when they  
don't match? Am I meant to ignore it?

>> 7. |guid| could recommend something along the lines of using some
>> algorithm that will ensure that the |guid| /is/ globally unique (and
>> not just feed unique).
>
> RSS 2.0 doesn't require global uniqueness, in spite of the name GUID.
> But we do offer the Tag URI scheme as a way to create guids, which has
> the added advantage of being globally unique.

I am aware of what the requirements: I never asked for a new  
requirement, I asked for a recommendation.


- Geoffrey Sneddon




 
Yahoo! Groups Links

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

<*> Your email settings:
    Individual Email | Traditional

<*> To change settings online go to:
    http://groups.yahoo.com/group/rss-public/join
    (Yahoo! ID required)

<*> To change settings via email:
    mailto:[email protected] 
    mailto:[email protected]

<*> 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/