Dates in Atom 0.2 (was RE: comments on Atom 0.2)

["Danny Ayers" <danny666@xxxxxxxxxxx>]

I'm sure this has been said a few times already, but it would be useful for
the dates to map to Dublin Core definitions, so systems set up to use DC in
HTML and/or RDF could use the feed data.

(btw, the issue of timezones came up on Sam's blog, but I can't find the
ref.)

here are the related DC elements :

Label:  Created
  Definition:  Date of creation of the resource.

Label:  Valid
  Definition:  Date (often a range) of validity of a resource.

Label:  Available
  Definition:  Date (often a range) that the resource will become or did
become available.

Label:  Issued
  Definition:  Date of formal issuance (e.g., publication) of the resource.

Label:  Modified
  Definition:  Date on which the resource was changed.

from http://dublincore.org/documents/dcmi-terms/#H3

Cheers,
Danny.


> -----Original Message-----
> From: owner-atom-syntax@xxxxxxxxxxxx
> [mailto:owner-atom-syntax@xxxxxxxxxxxx]On Behalf Of Nelson Minar
> Sent: 08 August 2003 17:12
> To: Evan Martin; Gary F; David.Pawson@xxxxxxxxxxx
> Cc: atom-syntax@xxxxxxx
> Subject: Re: comments on Atom 0.2
>
>
>
> Thanks for all the replies - I'm going to combine my answers here.
>
> >Nelson, one of us is getting the datetime's wrong. I'm not
> saying who though
> >:-) I'm not sure.
> >2003-08-07T09:36-07:00
> >2003-07-15T15:59:60-08:03
> >                ^^^^^^^
>
> I think these are both valid times - your's has seconds, mine doesn't.
> I believe the Atom timestamps are relative to the W3C subset of
> ISO8601, http://www.w3.org/TR/1998/NOTE-datetime-19980827
> There are a variety of forms there that are valid.
>
>
> >If I post in PST but don't tag it with <issued>, then the only date the
> >blog tool has is the modified date, which is in UTC.  How would it know
> >the proper timezone to translate that date into when showing the date
> >you posted with?
>
> I hadn't thought about that. So we'd like to know two different
> things: the absolute time that the entry was created, and the local
> time (to convey things like "this is a 3am rambling").
>
> I still think <issued> should be optional; I don't think most blogs
> will usefully distinguish between the two, so it shouldn't be
> required. I also wonder if maybe best practice is not to make
> <modified> be *local time* with a timezone offset rather than UTC.
> That way you get both absolute time and the local time in one little
> timestamp. That may pose a problem for sites like Blogger and
> LiveJournal, though, do they know what timezone their user is in?
>
> Besides timezone, the other question in <issued> vs <modified> is
> whether issued is the creation time of the post and modified is the
> last modified time. Is that the intent? I could see that being useful
> in some cases, but still I think <issued> should be optional.
>
>
> >>I think <summary> and similar elements should have the same content
> >>encoding attribute that <content> does.
> >And if <content> is multipart how should the aggregator resolve the type
> >of  encoding?
>
> Uh, I dunno. Multipart is complicated. I just want something really
> simple; I want to make a word in my summary bold, or include an
> element like &eAcute; without resorting to character encoding.
> (Honestly, the real reason is I'm lazy. In Blosxom the summary (title)
> is also HTML and converting it to text is awkward, particularly when
> you have entities embedded.)
>
>
>
> PS:
>
> >NOTICE: The information contained in this email and any attachments is
> >confidential and may be legally privileged. ...
>
> I do not consent to any legal terms included in email sent to me.