Re: [uf-discuss] OpenID delegation and XFN

2007-10-13 Thread André Luís
According to [1] it's fine sticking rel attribute in link elements so
I don't see why not, generally. But read on.

I would even extend the question why not adding rev=me since that
url should be mine as well but,this [2] proved me wrong..

[1] http://www.w3.org/TR/html401/struct/links.html#edef-LINK
[2] http://microformats.org/wiki/rel-faq#Should_rev_even_be_used

I would like to add though, that the only situation where I see this
making sense would be when the user would like to have a certain page
related with all others social networks ( or more general, his/her
presences on the web) and not show that information. (just hiding it
with css smells funny to me).

Summarizing:

Case #1: User wants to connect his page with a bunch of others but
doesn't want that explicitly stated on the page = user includes
links with  rel=me attribute.
Case #2; User wants to create a landing page for himself and
explicitly state his other presences out in the open, for everyone to
see = user includes a with rel=me attribute.

Final Note:
This could bring problems to certain implementations that would search
the DOM document for all A elements and not A and LINK. I'm not
sure whether this situation has been discussed earlier in the list or
is in the wiki. If so, I'd be thankful if someone could point me
there.


Cheers,
André Luís

On 10/13/07, Chris Messina [EMAIL PROTECTED] wrote:
 Will Norris made a great suggesting for making OpenID delegation links
 compatible with XFN rel-me links. I actually think that it's okay to
 use link /'s in the header of a document to point to other social
 networks and so on... I understand that this is not kosher with
 marking up visible data, but if it's going to be there anyway (and
 it's part of the OpenID spec) we might as well markup the links in the
 microformats format:

 link rel=openid.server href=http://www.myopenid.com/server; /
 link rel=openid.delegate me href=http://will.norris.name/; /

 Thoughts on this? I know that I'll get push back, but logically why not this?

 link rel=me href=http://flickr.com/photos/factoryjoe; /
 link rel=me href=http://ma.gnolia.com/people/factoryjoe; /

 Chris

 --
 Chris Messina
 Citizen Provocateur 
   Open Source Advocate-at-Large
 Work: http://citizenagency.com
 Blog: http://factoryjoe.com/blog
 Cell: 412 225-1051
 IM: factoryjoe
 This email is:   [ ] bloggable[X] ask first   [ ] private
 ___
 microformats-discuss mailing list
 microformats-discuss@microformats.org
 http://microformats.org/mailman/listinfo/microformats-discuss


___
microformats-discuss mailing list
microformats-discuss@microformats.org
http://microformats.org/mailman/listinfo/microformats-discuss


Re: [uf-discuss] OpenID delegation and XFN

2007-10-12 Thread Tom Morris
On 10/13/07, Chris Messina [EMAIL PROTECTED] wrote:
 Will Norris made a great suggesting for making OpenID delegation links
 compatible with XFN rel-me links. I actually think that it's okay to
 use link /'s in the header of a document to point to other social
 networks and so on... I understand that this is not kosher with
 marking up visible data, but if it's going to be there anyway (and
 it's part of the OpenID spec) we might as well markup the links in the
 microformats format:

 link rel=openid.server href=http://www.myopenid.com/server; /
 link rel=openid.delegate me href=http://will.norris.name/; /

 Thoughts on this? I know that I'll get push back, but logically why not this?

 link rel=me href=http://flickr.com/photos/factoryjoe; /
 link rel=me href=http://ma.gnolia.com/people/factoryjoe; /


Logically, sure, fine. Practically, one might need to check that
OpenID libraries are
parsing the HTML properly (the XFN tools will be fine, since they are coded with
multiple values in mind) - the programmer may have neglected to look for
space-separated values inside the rel value. I know I'd have to edit
the source of
an OpenID function I wrote a while back.

GrokXFN.xsl (the GRDDL parser for XFN) only looks in 'a' elements for XFN links.
I'm not sure about other parsers (and it's 2am, so I'm not lookin'...)

I think it's a good idea. Visible meta-data is a good idea, but not an absolute
requirement. (Of course, all data is visible - 'View Source'). Putting
microformats
on head/link elements is fine, in my book. The more data, the merrier.

-- 
Tom Morris
http://tommorris.org/
___
microformats-discuss mailing list
microformats-discuss@microformats.org
http://microformats.org/mailman/listinfo/microformats-discuss