[uf-discuss] Tentative proposal for "What's New" listings

[Andy Mabbett]

In message <6991f8e00609260653x2037b39rb651815aa66b518e at mail.gmail.com>,
Stephen Paul Weber <singpolyma at gmail.com> writes

>The converter currently blindly uses the permalink value as GUID, if
>you use the same permalink twice in your hAtom you get this error.

Is that wise? Using the same link twice is allowed, isn't it?

>  It
>does not affect most feedreaders, but if you can't change your page
>and this is a big issue, I can try hacking a GUID creator into the
>code for you :)

Thank you. You could always append the date-time to the URL.

>>title should not contain HTML: &quot; (6
>>         occurrences)
>>           Friends&amp;quot; section of our links page.
>
>The title is drawn directly from entry-title.  This is not actually
>invalid RSS, it's just something they don't reccomend, but if you want
>it changed, change your hAtom.

By "change my hAtom", what do you mean? Surely you're not suggesting I
replace my escaped ampersands, with invalid ampersand characters?

Since there will be other people who use escaped entities, wouldn't it
be better for you, to deal with them by "unescaping" them?

>>         line 389, column 3: item should contain a guid element
>
>Again, GUID is blindly drawn from permalink.  No permalink, no GUID.
>This again is not invalid, as the results say, only disreccomended.

Again, is that wise? You could always use the URL of the source page
(+date-time) if no linking URL is present.

>>         This feed is valid, but may cause problems for some users. We
>>         recommend fixing these problems.
>
>Again, none of these are actual invalidities

No, but I thought you'd want to know of them anyway (be generous in what
you receive strict in what you send...)

>>         line 74, column 17: description should not contain relative URL
>>         references: ../biblio/BirdLife/1983-0506-42.htm" rel="bookmark"
>>         title="letter in Bird Life magazine
>
>This is in your code... nor is it invalid, just not perfect.

Hey, who are you calling an imperfect coder! ;-)

>  Not sure
>if making this an absolute URL (since it's escaped HTML) is really the
>converter's job

Whose then?

If you see "../biblio/BirdLife/1983-0506-42.htm" on a page on the web,
it can have only one meaning, in the context in which you're seeing, it.

>>         line 100, column 46: Implausible date: Wed, 31 Dec 1969 23:59:59
>>         +0000 (8 occurrences)
>
>This is probably because you use a different date format in your hAtom
>(Y-m-d) instead of the full hAtom-reccomended datestamp
>(Y-m-D\TH:i:sP).

Tath's not my reading of teh spec:

        <http://microformats.org/wiki/hatom#Entry_Updated>

which says:

        use the datetime-design-pattern to encode

then in turn:

        <http://microformats.org/wiki/datetime-design-pattern>

says:

        add a title attribute to the abbr element with the machine
        readable ISO8601 datetime or date as the value

and, according to Wikipedia:

        <http://en.wikipedia.org/wiki/ISO_8601#Years>

the format "yyyy-mm-dd" is allowable.


>  My converter tries to work with alternate date
>formats, but it seems that on a few of your dates Y-m-d is being
>interpreted differently...

As 1969!?!

>I didn't write strtotime, but this code has
>been tested with the full datestamp.  If in doubt, use that.

I'm not in doubt ;-)

>> >The second page seems to be coming through fine now as well.
>>
>> I'm still getting just the one date (in  FireFox's live bookmarks) and
>> an XML Parse error (in Sage).
>
>Firefox live bookmarks only show items that have a link.

Bummer, I'll make a separate post about that.

Nonetheless, Sage is now showing some very odd results (do you see that,
or shall I "flickr" a screenshot?)


>> >From source code (simplified):
>>
>>         <abbr class="updated entry-title" title="2006-09-20">20th</abbr>
>>
>> You're serving:
>>
>>         <title>20th:  </title>
>>
>>         <pubDate>Tue, 26 Sep 2006 20:00:00 +0000</pubDate>

>correct
>according to my understanding of the hAtom spec.

I see that's being discussed separately.

>> >Also, last time I checked RSS 2.0 required a full datestamp in that
>> >format for pubDate... nothing else should be legal
>>
>> That's annoying. If true, we should recognise that in the hAtom spec.
>
>Why?  It's 100% irrelevant to microformats.

Not if the hAtom spec says you can do something, which then leads to
unexpected, or even bizarre, results.

>  ATOM itself requires a
>full timestamp.  Most formats do.  hAtom has it's own datestamp
>requirements.  The converter's job is to make sure the RSS
>requirements are met... which has nothing to do with you.  Your job is
>to meet hAtom, the converter can change the date format just fine.  It
>has no bearing on the hAtom spec, and no human being will ever see the
>date as it is in the RSS unless for some reason they read the code or
>have a feedreader that likes that dateformat.

Well, that's a big "unless" (sage, for instance shows the time).

In fact, you're inserting data into the feed, which isn't on my page!

>> You seem to be inserting an odd character

>I've fixed this now, thanks for the heads-up :)

A pleasure (and in my interest!). Thank you again for your time.

-- 
Andy Mabbett
                Say "NO!" to compulsory ID Cards:  <http://www.no2id.net/>

                Free Our Data:  <http://www.freeourdata.org.uk>