wfw:commentRss
Alec Muffett
I am looking at using wfw:commentRss from http://wellformedweb.org/CommentAPI/ - have any of you folk encountered it before, please?
-a
--
alec.m...@gmail.com
http://www.crypticide.com/dropsafe/
Joseph Holsten
> I am looking at using wfw:commentRss from http://wellformedweb.org/CommentAPI/ - have any of you folk encountered it before, please?
Can't say I have, but Joe Gregario has a good pedigree. Since that's from 2003, before he wrote atompub, I bet he might suggest a different solution. I've CC'd him just in case I'm spreading blasphemies.
These days, you probably want to use Atom Threading Extensions (http://tools.ietf.org/html/rfc4685) for the simple stuff like what wfw:commentRss provided. Basically add a <link rel="replies" type="application/atom+xml" href="http://www.example.org/mycommentsfeed.xml" thr:count="10" thr:updated="2005-07-28T12:10:00Z" /> to the entry and you're set.
The rest of the CommentAPI is probably best replaced by John Panzer's salmon protocol: http://salmon-protocol.org. And look at that, Ben Laurie's one of the owners of the salmon protocol google code project in here: http://code.google.com/p/salmon-protocol/people/list. Is he still on this list?
Any way, the salmon sample code is in python, and I should be able to help if (nee when) any of salmon's wonky xrd/lrdd/webfinger/hostmeta/push disco glue is confusing.
--
http://josephholsten.com
Alec Muffett
More input would be cool - I only really noticed the functionality since loading Liferea on Ubuntu - NetNewsWire doesn't seem to support it - and I want to see if there's a Minekey-friendly way of implementing comments-visible-in-the-feed; one big challenge is that ATOM is wonderful... but iTunes seems to be really nitpicky and annoying about enclosure-handling in ATOM, so I am currently hoping for a solution which is both ATOM-and-RSS201-friendly.
> These days, you probably want to use Atom Threading Extensions (http://tools.ietf.org/html/rfc4685) for the simple stuff like what wfw:commentRss provided. Basically add a <link rel="replies" type="application/atom+xml" href="http://www.example.org/mycommentsfeed.xml" thr:count="10" thr:updated="2005-07-28T12:10:00Z" /> to the entry and you're set.
Mmmmmmmm yaaaaaah - except of course that I have to generate distinct and separate comments-feeds, per-item, per-subscriber. :-)
> The rest of the CommentAPI is probably best replaced by John Panzer's salmon protocol: http://salmon-protocol.org. And look at that, Ben Laurie's one of the owners of the salmon protocol google code project in here: http://code.google.com/p/salmon-protocol/people/list. Is he still on this list?
Yes-and-no; with Ben it's generally quicker to prod him offline than it is to try and invoke his name in a public forum.
> Any way, the salmon sample code is in python, and I should be able to help if (nee when) any of salmon's wonky xrd/lrdd/webfinger/hostmeta/push disco glue is confusing.
That would be really cool! More to come when my brain finally kicks-in. ++Coffee
Joseph Holsten
> More input would be cool - I only really noticed the functionality since loading Liferea on Ubuntu - NetNewsWire doesn't seem to support it - and I want to see if there's a Minekey-friendly way of implementing comments-visible-in-the-feed; one big challenge is that ATOM is wonderful... but iTunes seems to be really nitpicky and annoying about enclosure-handling in ATOM, so I am currently hoping for a solution which is both ATOM-and-RSS201-friendly.
>
>> These days, you probably want to use Atom Threading Extensions (http://tools.ietf.org/html/rfc4685) for the simple stuff like what wfw:commentRss provided. Basically add a <link rel="replies" type="application/atom+xml" href="http://www.example.org/mycommentsfeed.xml" thr:count="10" thr:updated="2005-07-28T12:10:00Z" /> to the entry and you're set.
>
> Mmmmmmmm yaaaaaah - except of course that I have to generate distinct and separate comments-feeds, per-item, per-subscriber. :-)
Sorry TMTOWTDI, so I was just giving the solution most similar to wfw:commentRss. Here's an example of that:
https://developer.mozilla.org/en/RSS/Article/Why_Well-Formed_Web_RSS_Module_is_Popular_-_Syndicating_Your_Comments
You can also inline the comment entries into the feed right next to the entries they are commenting on. In that case, you'll say something like:
<entry>
<id>tag:example.org,2005:1</id>
<title>Article</title>
</entry>
<entry>
<title>Comment re Article</title>
<thr:in-reply-to
ref="tag:example.org,2005:1"
type="application/xhtml+xml"
href="http://www.example.org/entries/1"/>
</entry>
Is that more your style?
Using entries in this way is what makes salmon work. You post/put to the salmon-reply endpoint with an entry containing a thr:in-reply-to (waves security hands) and it shows up in the salmon-compatible feed.
--
http://josephholsten.com
Alec Muffett
>
> Is that more your style?
Maybe.
I am IMing with Ben and trying to make sense of the protocol; it looks great in a pile of ways, although a lot of what it solves is not immediately relevant to the Mine philosophy of personal platforms - where you are sharing person-to-person you're not likely to require PSHB and/or upstream flow of comments from Aggregators.
Hmmm... I am wondering what's the minimum I need to do, to get away with what I want to achieve; if you consider the Mine to be a "unicast" blog, then the comments represent a bidirectional conversation between mine-owner and mine-subscriber. Having the per-item owner-subscriber-comment-stream available to the subscriber as a feed makes the most sense to me - we're already assuming that feed-reading technologies are ubiquitous, after all.
Maybe salmon is overkill/too cutting edge?
Adriana
turns comments into a more 'legitimate' data in Mine!.
For your information, in the current UI we dealt with this problem
manually, as it were. Basically, if there is a comment on one of the
items/objects, that a user might want to keep e.g. a recommendation or
a response or information that is worth noting, we simply assumed that
a user will 'save comment as an item'. That opens up an item/object
creation page and the content of the comment is dealt with just like
creating another item. This is basically, because we don't know how
many comments people will leave, if they leave any, what kind etc. It
may be that the really interesting information in response to objects
in my Mine! feed, will be in Mine! feeds from other people etc etc.
Until we have a bit more idea of user behaviour (including this
particular user who can't wait to use it!) there is no point in
complicating the matter. This is something that will be on the list of
todo, once we have a usable Mine!, preferrably by yesterday. :)
Also, I attach some Mine! UI wireframes we got so far, they are not
'designed' yet, just simple representation of some clickthroughs and
some of the stuff we envisage on the screen. Work in progress, in
other words. The files are zipped graffle and a zipped visio.
Adriana
Note: Google groups just bounced my message as it was over 4MB -
interesting to know the limit. Not! I am sending just the zipped
graffle file first, then the visio one in a separate email. Oh well.
>> --
>> You received this message because you are subscribed to the Google Groups "themineproject discussion" group.
>> To post to this group, send email to themine...@googlegroups.com.
>> To unsubscribe from this group, send email to themineprojec...@googlegroups.com.
>> For more options, visit this group at http://groups.google.com/group/themineproject?hl=en.
>>
>>
>