List Info

Thread: dfn design pattern (proposal)




dfn design pattern (proposal)
user name
2007-08-17 06:44:27
This is may be somewhat premature as the results of the
assistive  
technology tests aren't back yet:
http://microformats.org/wiki/assistive-technology-
abbr-results

But if we do need to look at alternatives to the abbr design
pattern,  
one of the ideas that came up in discussion was to use the
dfn element.

I've put together a page on the wiki outlining the thinking
behind  
this proposal as well as highlighting some of the potential
downsides:

http:
//microformats.org/wiki/dfn-design-pattern

On the plus side, I think it's better not to widen the title
design  
pattern to apply to any element (which is essentially what
the span  
proposal is saying) so this would only be a slight
expansion.

On the down side, the semantics of "defining
instance" aren't always  
going to be applicable for datetime, geo, etc. But I think
it could  
well cover 80% of use cases.

What's needed:

* Arguments for or against the use of dfn as a container for
the  
title design pattern: is this semantic abuse or is it simply
 
stretching semantics (like the abbr design pattern).

* Document usage of the dfn element in the wild: I believe
it is  
often used in conjunction with the title attribute.

* Test results from screen readers to find out if dfn is
treated as a  
special case (like abbr and acronym) or whether it is
"safe" to use.

* Feedback from the people building parsers (Mike Kaply,
Brian Suda,  
etc.) on whether this would be tricky or easy to implement.

I'm fairly certain that this proposed design pattern would
*not*  
cover 100% of use cases but it might cover enough situations
to be  
viable as *an alternative* to the abbr design pattern.

Note that I am not suggesting that the abbr deisgn pattern
should be  
deprecated. I believe it has its place but I think it would
be good  
to provide an alternative to address the accessibility
question.

For instance, even if the dfn design pattern is adopted, I
will still  
use the abbr element for cases like this:

<abbr class="dtstart"
title="2007-08-19">August 19th</abbr>

That's because I believe exposing the string
"2007-08-19" either to  
sighted or blind users is an acceptable, readable,
understandable way  
of formating a date. But for a string like
"2007-08-19T12:39:00" I  
would like to have an alternative that wouldn't directly
expose that  
format to the user.

That's my own call, of course. I suspect that others,
offered the  
choice of an alternative to abbr, will always go for the
alternative.  
And others will choose to always stick with abbr. I think
that all of  
those positions should be accommodated.

Look forward to getting your feedback,

Bye,

Jeremy

-- 
Jeremy Keith

a d a c t i o

http://adactio.com/


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

Re: dfn design pattern (proposal)
country flaguser name
United States
2007-08-21 04:28:25
> * Arguments for or against the use of dfn as a
container for the title
>   design pattern: is this semantic abuse or is it
simply  stretching
>   semantics (like the abbr design pattern).

In the interest of forward-compatibility, I think using
dfn/title is a
bad idea.

The HTML 5 draft goes into much more detail about
<dfn> then any
previous HTML version, specifying both how to find the
<dfn>'s term, and
how to find the relevant <dfn> for some use of the
term elsewhere in the
document.

       http://www.whatwg.org/specs/web-apps/current-work/#the
-dfn

The incompatibility comes in here:

> Defining term: If the dfn element has a title
attribute, then the
> exact value of that attribute is the term being
defined. Otherwise, if
> it contains exactly one element child node and no child
text nodes,
> and that child element is an abbr element with a title
attribute, then
> the exact value of that attribute is the term being
defined.
> Otherwise, it is the exact textContent of the dfn
element that gives
> the term being defined.
> 
> If the title attribute of the dfn element is present,
then it must
> only contain the term being defined.

So using dfn/title to contain something other than the
term being
defined is probably not a good idea.

This is similar to how we're trying to avoid the use of rev now,
since
HTML 5's dropped it.


-- 
Ted

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

Re: Re: dfn design pattern (proposal)
user name
2007-08-21 04:47:06
On 8/21/07, Edward O'Connor <hober0gmail.com> wrote:
> > * Arguments for or against the use of dfn as a
container for the title
> >   design pattern: is this semantic abuse or is it
simply  stretching
> >   semantics (like the abbr design pattern).
>
> In the interest of forward-compatibility, I think using
dfn/title is a
> bad idea.
>
> The HTML 5 draft goes into much more detail about
<dfn> then any
> previous HTML version, specifying both how to find the
<dfn>'s term, and
> how to find the relevant <dfn> for some use of
the term elsewhere in the
> document.

--- i wouldn't worry too much about HTML5. Firstly, it is
years away
from completion, and secondly, it also gives us the
<time> element

http://www.
w3.org/html/wg/html5/#the-time

which would be the most semantic element for this pattern.
The trouble
is that we are searching for a solution to the issues NOW.
If/when
HTML5 is finished in 2010 as planned, how many years before
it gets
into browsers, and/or into older text-only browsers like
LYNX  or
aural browsers that don't just hook into IE?

The DFN and other proposals are searching for a solution
that can be
used forever with HTML4. With HTML5 and other future
languages, the
semantics will be defined by new (hopefully better)
elements. We can
cross those bridges as we get there.

Thanks for the heads-up on DFN, i have noted it on the
dfn-design-pattern page:

http://microformats.org/wiki/dfn-design-pattern#HTML5

-brian

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

Re: dfn design pattern (proposal)
country flaguser name
United States
2007-08-21 11:34:13
Brian wrote:

> it[...] gives us the <time> element[,] which
would be the most
> semantic element for this pattern.
[...]
> The trouble is that we are searching for a solution to
the issues NOW.
[...]
> The DFN and other proposals are searching for a
solution that can be
> used forever with HTML4.

If browsers introduced a new parsing mode for HTML5, this
might be a
reasonable way forward, but it's pretty clear that they
don't intend to
do so.[1] The stated goal is to process all versions of HTML
with the
same, HTML5, algorithm.

Basically, HTML5 moves us away from HTML language *versions*
and toward
CSS-like language *levels*. We should try to formulate our
markup with
that in mind.

When the <time> element becomes available, yes,
microformats should
prefer its use. But with my microformat parser implementor
hat on, I
don't intend to reject use of the <abbr> design
pattern once <time> is
available. They'll both work, it's just that one will be
preferred. I
can't say the same about the <dfn> design pattern.



1. http://lists.w3.org/Archives/Public/public-html
/2007Apr/1609.html

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

Re: Re: dfn design pattern (proposal)
country flaguser name
United Kingdom
2007-08-21 13:23:17
In message 
<21e770780708210247g7bd3a101jabb2e8b6cf9d05bamail.gmail.com>, Brian 
Suda <brian.sudagmail.com> writes

>i wouldn't worry too much about HTML5. Firstly, it is
years away from 
>completion, and secondly, it also gives us the
<time> element
>
>http://www.
w3.org/html/wg/html5/#the-time
>
>which would be the most semantic element for this
pattern.

for "dtstart" and other date-times, perhaps, but
not for Geo, or
tel-types in languages of other than English.

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

Re: Re: dfn design pattern (proposal)
country flaguser name
United States
2007-08-21 21:48:07

Brian Suda (21 août 2007 - 18:47) :
> On 8/21/07, Edward O'Connor <hober0gmail.com> wrote:
>> In the interest of forward-compatibility, I think
using dfn/title  
>> is a
>> bad idea.
>>
>> The HTML 5 draft goes into much more detail about
<dfn> then any
>> previous HTML version, specifying both how to find
the <dfn>'s  
>> term, and
>> how to find the relevant <dfn> for some use
of the term elsewhere  
>> in the
>> document.
>
> --- i wouldn't worry too much about HTML5.

The definition of [HTML 5 dfn][1] is not incompatible with
[HTML 4  
dfn][2]
It is just a better definition, less vague.
For the record, XHTML 2 WD gives another interesting pattern
for  
[XHTML 2 dfn][3] with the attribute role.

Though as you said html 5 is only an *editor* draft for
now.

[1]: http://www.w
3.org/html/wg/html5/#the-dfn
[2]: 
http://www.w3.org/TR/html4/struct/text.html#edef-DFN
[3]: http://www.w3.org/TR/xhtml2/mod-text.html#edef_text_dfn


-- 
Karl Dubost - http://www.w3.org/Peop
le/karl/
W3C Conformance Manager, QA Activity Lead
   QA Weblog - http://www.w3.org/QA/
      *** Be Strict To Be Cool ***




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

[1-6]

about | contact  Other archives ( Real Estate discussion Medical topics )