From: "Alan J. Flavell" <flavell@ph.gla.ac.uk>
Subject: Re: zen question
Date: 17. lipanj 2004 9:29

On Wed, 16 Jun 2004, Laurence Tureaud wrote, quoting:

> a resource we can all refer to when making the case for <acronym
> title="Cascading Style Sheets">CSS</acronym>-based design.

"CSS" is not an acronym, though.

On Thu, 17 Jun 2004 12:44:22 +0200, Berislav Lopac
<berislav.lopac@dimedia.hr> wrote:

> From: "Alan J. Flavell" <flavell@ph.gla.ac.uk>
> Subject: Re: zen question
> Date: 17. lipanj 2004 9:29
>
> On Wed, 16 Jun 2004, Laurence Tureaud wrote, quoting:
> 
>
> "CSS" is not an acronym, though.

That entirely depends on your definition of "acronym". Several dependable
dictionaries I've consulted define "acronym" without mention of it having
to be pronounceable. The W3 specs don't make it clear either.

Surely we can agree on a few things here, though: an acronym can also be
considered an abbr; IE won't do abbr; no UA's I'm aware of do anything
special with acronym that they don't do with abbr (other than IE observing
it and not the other). So I'm, frankly, inclined to look the other way if
acronym is misused (which I'm not convinced it is here anyway), being that
there no known loss in accessibility or usability.

Neal wrote:

> I'm, frankly, inclined to look the
> other way if acronym is misused (which I'm not convinced it is here
> anyway), being that there no known loss in accessibility or usability.

For my part, I'm confuses as to why the w3c felt the need to have both
<acronym> and <abbr>. Why not <abbr> with optional attribute
type="acronym"? I wonder too about things like <OL> and <UL>. Had it
been my choice, I think I'd have gone with <LIST>, again with optional
attribute type="ordered".

--
Brian (remove ".invalid" to email me)
http://www.tsmchughs.com/

"Neal" <neal413@yahoo.com> wrote in message
news:opr9qvfkp46v6656@news.individual.net...
> On Thu, 17 Jun 2004 12:44:22 +0200, Berislav Lopac
> <berislav.lopac@dimedia.hr> wrote:
> 
>
> That entirely depends on your definition of "acronym". Several dependable
> dictionaries I've consulted define "acronym" without mention of it having
> to be pronounceable.

Can you check again to make sure they don't say it has to be a *word*? All
the resources I trust specify that an acronym is a *word* created from the
initials of other words. I would argue that "CSS" is not a word.

> The W3 specs don't make it clear either.

The specs do make it clear what *they* think: they include "GmbH" and
"F.B.I." as abbreviations that they consider to be acronyms. However, they
didn't invent the word "acronym", nor are they a group of experienced
lexicographers, so nothing they say about the meaning of the word is
dispositive. OTOH, as you observe below, they don't really make it clear why
acronyms and abbreviations need distinct tags anyway, and as the W3
indicates, there are abbreviations (like "SQL") that are spoken both ways,
so they could just as well have used ABBR for both, so practically speaking
it likely doesn't matter.

>
> Surely we can agree on a few things here, though: an acronym can also be
> considered an abbr; IE won't do abbr; no UA's I'm aware of do anything
> special with acronym that they don't do with abbr (other than IE observing
> it and not the other). So I'm, frankly, inclined to look the other way if
> acronym is misused (which I'm not convinced it is here anyway), being that
> there no known loss in accessibility or usability.

"Brian" <usenet3@julietremblay.com.invalid> wrote in message
news:10d3e6c2hic8i09@corp.supernews.com...
> Neal wrote:
> 
>
> For my part, I'm confuses as to why the w3c felt the need to have both
> <acronym> and <abbr>. Why not <abbr> with optional attribute
> type="acronym"? I wonder too about things like <OL> and <UL>. Had it
> been my choice, I think I'd have gone with <LIST>, again with optional
> attribute type="ordered".

I agree with LIST instead of OL and UL, but I don't know about the type
attribute. Indicating that the markers should be bullets or roman numerals
or whatever takes care of that. While bulleted lists don't have their order
indicated by the markers, they are nevertheless ordered in the sense that it
would be unacceptable for a UA to render the items in different order from
the one in which they appear in the HTML.

"Harlan Messinger" <h.messinger@comcast.net> writes:
>  I agree with LIST instead of OL and UL, but I don't know about the type
> attribute. Indicating that the markers should be bullets or roman numerals
> or whatever takes care of that. While bulleted lists don't have their orde
r
> indicated by the markers, they are nevertheless ordered in the sense that 
it
> would be unacceptable for a UA to render the items in different order from
> the one in which they appear in the HTML.

Would it? If it's an unordered list, then the final display order
should be arbitrary. I don't know of any UAs that sort the output into
a different order (alphabetic?) but if it's a real UL, and not an OL
that has "list-style-type: disc" or similar, it shouldn't matter.

--
Chris

"Chris Morris" <c.i.morris@durham.ac.uk> wrote in message
news:87y8mmgoqm.fsf@dinopsis.dur.ac.uk...
> "Harlan Messinger" <h.messinger@comcast.net> writes: 
numerals 
order 
that it 
from 
>
> Would it? If it's an unordered list, then the final display order
> should be arbitrary.

How often, in the real world, would someone type a list of items and then
have a desire to indicate explicitly that the UA should feel free to display
them in another, arbitrary, order?

> I don't know of any UAs that sort the output into
> a different order (alphabetic?) but if it's a real UL, and not an OL
> that has "list-style-type: disc" or similar, it shouldn't matter.

On Thu, 17 Jun 2004, Brian wrote:

> For my part, I'm confuses as to why the w3c felt the need to have both
> <acronym> and <abbr>.

*They* were confused, too!  The discussions showed clearly that
everyone participating was quite certain what the word "acronym" meant
and thus needed no definition in the spec, and that everyone who said
otherwise was wrong.  Unfortunately, the meanings of which they
claimed to be so certain were all different...

OK, so why did they put <acronym> and <abbr> into the spec, and not
any other kind of distinction?  Well, the sceptics would say it was
because Netscape had already implemented <abbr> and MS had already
implemented <acronym> - so they both had to go into the spec or else
the Big Two wouldn't have accepted the spec.

More constructively, some participants reckoned that <acronym> would
give a valuable clue to speaking browsers.  But that fails miserably
when you get folks who reckon that "F.B.I." (*with* the dots,
ferchrissake) is an "acronym".

So the W3folk stuffed both the tags into the spec, woffled vaguely
about what they meant, gave some examples (including the misleading
one just mentioned), and left the users to work it out.  And when it
came to implementing this part of HTML4, MS sat on their hands, it
seems.

Harlan Messinger wrote:

> Brian wrote...
> 
>
> I agree with LIST instead of OL and UL, but I don't know about the
> type attribute. Indicating that the markers should be bullets or
> roman numerals or whatever takes care of that. While bulleted lists
> don't have their order indicated by the markers, they are
> nevertheless ordered in the sense that it would be unacceptable for
> a UA to render the items in different order from the one in which
> they appear in the HTML.

Of course. It would be unacceptable for a ua to change the order of
any characters in the stream, too. Only presntations suggestions from
the author or user should change where an element would otherwise
appear in the document, no? But an UL, even where its LI elements are
specified by the author, does not mean the same thing as an OL.

--
Brian (remove ".invalid" to email me)
http://www.tsmchughs.com/

On Thu, 17 Jun 2004 13:20:01 -0400, Harlan Messinger
<h.messinger@comcast.net> wrote:

> How often, in the real world, would someone type a list of items and then
> have a desire to indicate explicitly that the UA should feel free to
> display
> them in another, arbitrary, order?

For me, the difference between OL and UL is that items in an OL are
intentionally sequential, whether in order of importance, relevance, time,
or other parameter(s). UL, on the other hand, could be put into other
orders if the author so desired, but this is not to say there's any reason
a UA should re-order a UL.

So I see a semantic reason for both UL and OL, in addition to a useful
purpose. (Let us remember that, while we strive to separate content from
presentation, some presentation must be considered for non-CSS rendering.)

Regarding <l marker="disc"> or <l marker="arabic"> - this would be
extendable to allow for all sorts of rendering possibilities. Though we'd
need to have a way to determine default performance if a UA does not
support what we requested, and it becomes something like <l
type="unordered" marker="circle"> or <l type="ordered" marker="romanlc"> -
or perhaps more compactly, <ul marker="circle"> and <ol marker="romanlc">
- then we wonder why the marker style is even necessary, and we're back to
<ol> and <ul>... so I don't see that as an improvement over what we
currently have.

Contrast this to <acronym> and <abbr> which are not really organic to
HTML, and which are marginally useful at best. For this reason I rarely if
ever have used these, and doubt I ever will.