W3C home > Mailing lists > Public > www-html@w3.org > April 2003

Re: 'acronym' semantics

From: John Lewis <lewi0371@mrs.umn.edu>
Date: Fri, 11 Apr 2003 01:16:35 -0500
Message-ID: <8934129430.20030411011635@cda.mrs.umn.edu>
To: www-html@w3.org

Simon wrote on Thursday, April 10, 2003 at 7:37:22 AM:

> From: "Etan Wexler" <ewexler@stickdog.com>

>> All of your acronyms are abbreviations, by definition. The
>> 'acronym' element type has exactly the same fundamental semantics
>> as the 'abbr' element type has. The 'acronym' element type merely
>> adds the presentational hint, "Pronounce like a normal word".

> Isn't that a bit like saying all of your <strong> elements have the
> same fundamental semantics as your <em> elements, but with a
> presentational hint that "emphasis should be stronger"? I've seen
> suggestions that <strong>...</strong> should be replaced with
> <em><em>...</em></em>.

I think what he's saying is that abbreviations and acronyms, the
concepts, are almost identical. The only difference is a minor one of
presentation. That doesn't make acronym a presentational element (like
b or i). I consider abbr and acronym to both be semantic; however,
abbr is by definition adequate for marking up abbreviations of all
types, including similar but slightly different concepts like acronyms
and the like.

I recall an early math class where the teacher explained that "every
square is a rectangle, but not every rectangle is a square." The same
thing applies to this situation, only the square is acronym and the
rectangle is abbr, and the only difference is a minor one (of what is
essentially presentation). An abbreviation element is adequate for
marking up all types of abbreviations, including acronyms (no matter
what type of abbreviation you think an acronym is). It is less precise
than having a dozen elements for different types of abbreviations, but
not terribly so. It's better for XHTML2 to have one general element
for abbreviations than one for abbreviations and one for each subtype
of abbreviation (acronym included), and similarly I don't think
acronyms are special enough to warrant their own element in a language
like XHTML2.

Oh, and strong is so useless. Unlike the abbr/acronym situation,
nested em elements can outright replace and supersede the strong
element entirely. I still believe there is almost nothing to be gained
by keeping strong and absolutely nothing to be lost by dropping
strong. There is, however, a lot to be said for weeding out useless
elements and streamlining XHTML2 as much as possible. Each useless
element we get rid of increases the likelihood of XHTML2, or future
versions of XHTML, deeming allowable the addition of new useful
elements without bloating the language.

> Back to acronyms and abbreviations. What about specifying the type
> of abbreviation like this? -

[snip examples using the type attribute to delineate the type of

I honestly like the idea; although I'm not sure that the type
attribute is free for this use in XHTML.

> I know it looks a little long-winded, but it gives the purists an
> option to give deeper meaning to an abbreviation, does it not?

Yes. On the other hand, except for UA default styles, this could all
be done via the class attribute. I'm not sure if the lack of a
possibility of a default style is all that bad, since purists will
likely be writing detailed style sheets as well.

I'm wondering: Would it be possible for XHTML2 to "reserve" a number
of class attribute values for this purpose? That means we wouldn't
need a new attribute, and we could still have browser default styles.
Initially, I thought it would conflict with existing documents with
class attributes; but then I realized that it really only affects
certain documents that are converted to XHTML2, since XHTML2 isn't a
recommendation yet. Such conversions are already complex, so an
additional burden is not horrible; more importantly, as far as I know
such conversions are uncommon. As well, care could be taken to use
values that are exceedingly unlikely to be common, such as "-acronym"
and the like (although shorter would be nice). I can't recall if
hyphens are allowed at the beginning of class attribute values, but
you get the idea.

John Lewis
Received on Friday, 11 April 2003 02:21:59 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:06:03 UTC