W3C home > Mailing lists > Public > www-international@w3.org > January to March 2011

Re: [css3-text] text-transform:capitalize

From: Brady Duga <duga@ljug.com>
Date: Tue, 22 Feb 2011 14:43:11 -0800
Cc: Ambrose LI <ambrose.li@gmail.com>, Christoph Päper <christoph.paeper@crissov.de>, W3C style mailing list <www-style@w3.org>, www-international@w3.org
Message-Id: <DFC37AB0-AE3B-446B-9204-FE2BD17B5A6B@ljug.com>
To: Asmus Freytag <asmusf@ix.netcom.com>
I think I also tend to agree about it being useless. The current recommendation is to uppercase the first letter of every word, using roughly English word breaking rules, though perhaps allowing language specific breaking. Why would anyone want to do that? It seems to have limited utility in the one language where it would behave consistently. And if we expand it to really be title case (that is, correctly uppercasing only those words that should be uppercased for a given language) the likelihood it will work reproducibly across browsers and languages is pretty darn low, especially since those rules don't seem to well stated anywhere. 

--Brady

On Feb 22, 2011, at 12:10 PM, Asmus Freytag wrote:

> On 2/22/2011 11:42 AM, Ambrose LI wrote:
>> 2011/2/22 Christoph Päper<christoph.paeper@crissov.de>:
>>> Lowercasing the rest is helpful for
>>> 
>>>  <abbr>AIDS</abbr>
>>> 
>>> or, better yet,
>>> 
>>>  acronym {text-transform: capitalize;}
>>> 
>>> but harmful for
>>> 
>>>  <h1>Scrooge McDuck richest duck ever</h1>
>> I don't think even lowercasing the rest is helpful for the "AIDS"
>> case. If I write AIDS, it means that according to my stylistic
>> preferences this particular acronym should be in all uppercase, and my
>> preferences don't suddenly change if I put it into a title. We are not
>> even counting acronyms that *must* have correct mixed case, such as
>> "kB", "dB", "DDoS", "GmbH" and the like.
>> 
>> Personally, I think the "title case" transformation is useless. It is
>> almost useless for French, probably useless for other non-English
>> languages, and even if we consider just English it is too
>> content-specific and there's really no sure way to get a result that's
>> right all the time.
> 
> I agree. There are many languages that simply don't use titlecase the way this is understood for English, so in those languages, titlecase should somehow become a no-op for these languages.
> 
> Even in English you have all those rules about particular (small) words that are not to be titlecased. These words would still be sentence cased at the beginning of sentences (and beginning of the title) but not in the middle.
> 
> That makes the problem almost as complex as hyphenation - so if some UA had a great titlecasing engine with near perfect heuristics and good dictionary of exceptions, such a UA might be able to do a half-decent job, but you wouldn't be able to predict the outcome.
> 
> A./
> 
Received on Tuesday, 22 February 2011 22:43:48 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 22 February 2011 22:43:51 GMT