Major change to microformat specs without prior discussion or notification

The change made recently, to the hCard spec:

  <http://microformats.org/wiki?t
itle=hcard-parsing&diff=next&oldid=25563>

affects not only hCard but also other microformats; it has
implications
for both current parsers and published microformats, but
seems to have
been made without prior notification or debate.

Note, for example, that it precludes an event's hCalendar
and the event
organiser's hCard from sharing the URL in a single
"A" element.

I propose that the change be undone; and not reintroduced
until the
implications are more fully understood and widely agreed.

-- 
Andy Mabbett
** via webmail **

_______________________________________________
microformats-discuss mailing list
microformats-discussmicroformats.org
http://microformats.org/mailman/listinfo/microforma
ts-discuss

perhaps there was prior discussion and agreement that was
just a long 
time ago? Have you searched the archives or asked Tantek
directly?

Andy Mabbett wrote:
> The change made recently, to the hCard spec:
>
>   <http://microformats.org/wiki?t
itle=hcard-parsing&diff=next&oldid=25563>
>
> affects not only hCard but also other microformats; it
has implications
> for both current parsers and published microformats,
but seems to have
> been made without prior notification or debate.
>
> Note, for example, that it precludes an event's
hCalendar and the event
> organiser's hCard from sharing the URL in a single
"A" element.
>
> I propose that the change be undone; and not
reintroduced until the
> implications are more fully understood and widely
agreed.
>
>   

_______________________________________________
microformats-discuss mailing list
microformats-discussmicroformats.org
http://microformats.org/mailman/listinfo/microforma
ts-discuss

Thom Shannon wrote:
> perhaps there was prior discussion and agreement that
was just a long
> time ago? Have you searched the archives or asked
Tantek directly?

Here's the IRC log regarding the change to the wiki:

http://rbach.priv.at/Microformats/IRC/2008-02-07#T010220


I agree with the change - I don't agree with not running it
past the
microformats-new list. It seems like a fairly far-reaching
change/update. It invalidates the need for "mfo"
in hcard, doesn't it?
If it were applied to the rest of Microformats, it would
invalidate the
need for "mfo" entirely.

There are logs - so it would be wrong to say the decision
was made in
private, it was done on IRC, without notification to
microformats-new.

-- manu
_______________________________________________
microformats-discuss mailing list
microformats-discussmicroformats.org
http://microformats.org/mailman/listinfo/microforma
ts-discuss

On Feb 7, 2008, at 11:08 AM, Manu Sporny wrote:

>  It invalidates the need for "mfo" in hcard,
doesn't it?
> If it were applied to the rest of Microformats, it
would invalidate  
> the
> need for "mfo" entirely.


Not exactly.  As <hober> said in IRC:

# [02:17:12] <hober> just to quickly clarify, the
opacity parsing rule  
applies to microformat X with regards to nested instances of
X
# [02:17:34] <hober> (as opposed to nested instances
of any, even  
future, microformat)

That "nested instances of any, even future,
microformat" is what MFO  
[1] would cover, but probably never will because piecemeal
solutions  
like this eliminate the various problems MFO would solve all
at once.

On the topic of whether this should have had wider
discussion, I  
thought it was well established long ago that the properties
of an  
AGENT hCard are not inherited by the container hCard, so I
don't see  
anything really changing here.  Was anyone publishing agent
hCards and  
expecting the properties to also apply to the container? 
That seems  
very counter-intuitive to me.

[1] http://microformats.
org/wiki/mfo

Peace,
Scott

_______________________________________________
microformats-discuss mailing list
microformats-discussmicroformats.org
http://microformats.org/mailman/listinfo/microforma
ts-discuss

In message <D992B0BA-4EAB-4AA1-8BF4-6CD9BF81A551randomchaos.com>, Scott 
Reynen <scottrandomchaos.com> writes

>On the topic of whether this should have had wider
discussion, I 
>thought it was well established long ago that the
properties of an 
>AGENT hCard are not inherited by the container hCard, so
I don't see 
>anything really changing here.

The change made today does not refer only to Agent; nor even
only to 
hCard.

-- 
Andy Mabbett
_______________________________________________
microformats-discuss mailing list
microformats-discussmicroformats.org
http://microformats.org/mailman/listinfo/microforma
ts-discuss

In message <47AB4911.5020809digitalbazaar.com>, Manu
Sporny 
<mspornydigitalbazaar.com> writes

>Thom Shannon wrote:
>> perhaps there was prior discussion and agreement
that was just a long
>> time ago? Have you searched the archives or asked
Tantek directly?
>
>Here's the IRC log regarding the change to the wiki:
>
>http://rbach.priv.at/Microformats/IRC/2008-02-07#T010220


I don't consider that to be adequate period of discussion;
and I don't 
consider that forum alone to be an adequate forum for
discussion of such 
a major issue.

What would happen if I implemented one of my outstanding -
and nowhere 
near as harmful - proposals in that way?

Why has it not been reverted yet? If I had any faith that
this was truly 
an open community, I would do so myself.

>I agree with the change

I agree with the general principle behind the change. I
don't agree with 
the method of implementation, or with changing the spec in a
way which 
has such wide- reaching implications, and
backwards-compatibility 
issues, with out doing adequate research, and giving ample
warning, 
first. I don't agree with making the change wholesale rather
than 
examining the implications in each type of use, And I don't
agree with 
changing the way several microformats work, by amending the

parsing-rules page for one of them.

>- I don't agree with not running it past the
>microformats-new list.

It's not just about new microformats, but also existing
microformats; it 
should be discussed here, too.

>It seems like a fairly far-reaching
>change/update. It invalidates the need for
"mfo" in hcard, doesn't it?
>If it were applied to the rest of Microformats, it would
invalidate the
>need for "mfo" entirely.

It also break some previously-valid implementations.

>There are logs - so it would be wrong to say the
decision was made in
>private, it was done on IRC, without notification to
microformats-new.

Who said it was made in private?

-- 
Andy Mabbett
_______________________________________________
microformats-discuss mailing list
microformats-discussmicroformats.org
http://microformats.org/mailman/listinfo/microforma
ts-discuss

On Feb 7, 2008, at 3:03 PM, Andy Mabbett wrote:

>> On the topic of whether this should have had wider
discussion, I  
>> thought it was well established long ago that the
properties of an  
>> AGENT hCard are not inherited by the container
hCard, so I don't  
>> see anything really changing here.
>
> The change made today does not refer only to Agent; nor
even only to  
> hCard.

I don't understand why you think that.  Everything I'm
reading on the  
change page you linked seems to be specific to hCards within
other  
hCards, and the only relationship defined for such nesting
is AGENT.   
I agree major changes should be discussed more than this
was, but I  
still don't see how this is a change at all, much less
major.

Peace,
Scott

_______________________________________________
microformats-discuss mailing list
microformats-discussmicroformats.org
http://microformats.org/mailman/listinfo/microforma
ts-discuss

In message <69FE1D70-C3F3-4920-B196-6A91BA7FF683randomchaos.com>, Scott
Reynen <scottrandomchaos.com> writes

>On Feb 7, 2008, at 3:03 PM, Andy Mabbett wrote:
>
>>> On the topic of whether this should have had
wider discussion, I
>>>thought it was well established long ago that
the properties of an
>>>AGENT hCard are not inherited by the container
hCard, so I don't
>>>see anything really changing here.
>>
>> The change made today does not refer only to Agent;
nor even only to
>>hCard.
>
>I don't understand why you think that.  Everything I'm
reading on the
>change page you linked seems to be specific to hCards
within other
>hCards, and the only relationship defined for such
nesting is AGENT.
>I agree major changes should be discussed more than this
was, but I
>still don't see how this is a change at all, much less
major.

  <http://microformats.org/wiki?title=hcard-parsing&curid=
1121&diff=25580&oldid=25563&rcid=39352>

  (aka <http://tinyurl.com/3e35
8a>)

added:

        Similarly, parsers should treat nested
[[hCalendar]],
        [[hReview]], [[hResume]] [[xFolk]] in the same way,
properties
        inside them {} only apply to the nested
microformat, not
        to the containing microformat.

        All references below to "inside the
hCard", "within the contents
        of the hCard", and similar phrasing {} be
interpreted with
        taking this nesting rule into account.

-- 
Andy Mabbett
_______________________________________________
microformats-discuss mailing list
microformats-discussmicroformats.org
http://microformats.org/mailman/listinfo/microforma
ts-discuss

On Feb 7, 2008, at 5:11 PM, Andy Mabbett wrote:

> added:
>
>        Similarly, parsers should treat nested
[[hCalendar]],
>        [[hReview]], [[hResume]] [[xFolk]] in the same
way, properties
>        inside them {} only apply to the nested
microformat, not
>        to the containing microformat.
>
>        All references below to "inside the
hCard", "within the  
> contents
>        of the hCard", and similar phrasing
{} be interpreted  
> with
>        taking this nesting rule into account.


Aha, I totally missed that part when looking at the change
log twice.   
I'm not clear on the intended scope of this rule, but it's
certainly  
broader than I mistakenly suggested earlier.

Peace,
Scott
_______________________________________________
microformats-discuss mailing list
microformats-discussmicroformats.org
http://microformats.org/mailman/listinfo/microforma
ts-discuss