W3C home > Mailing lists > Public > www-validator@w3.org > April 2005

Re: Question re:"allowtransparency" attribute in iframes

From: Lachlan Hunt <lachlan.hunt@lachy.id.au>
Date: Thu, 07 Apr 2005 16:55:52 +1000
Message-ID: <4254D978.10909@lachy.id.au>
To: ryan@getnicestudios.com
CC: www-validator@w3.org

Christopher Ryan McVinney wrote:
> Lachlan Hunt wrote:
>> The validator is correct, it is an invalid attribute.
> 
> That is incorrect, it IS a valid attribute these days,

According to...?

Well, it is certainly *not* in the HTML 4.01 recommendation [1]. 
Therefore, it is *not valid* and *non-conformant*.  It is, and always 
will be, a proprietary non-standard presentational attribute that 
*should not* be used.

> and that is why I was writing to the w3c regarding its use.
> It should be defined in the 4.01 transitional html dtd,
> and I do not know why it isn't.

Because the HTML 4 recommendation should not be updated to include every 
non-standard proprietary, presentational attribute.  Presentational 
aspects like this allowtransparency is something that should be 
specified in a stylesheet, if ever.

Although, there is currently no alternative available in CSS for this 
attribute, I suggest you remove the attribute anyway.

> Microsoft created it, I 'll give you that, but check it out in any other 
> browser's latest version and it works across the board.

Wide spread use and implementation among the popular browsers to cope 
with large number of people like yourself that make use of and depend 
upon non-standard proprietary attributes to make their pages render 
correctly is not a reason for it to be standardised.

There are lots of proprietary extensions copied by other browsers simply 
to cope with all the pages designed and tested with only one browser in 
mind.  If they did not, unfortunately, many pages simply wouldn't work 
and everyone would be stuck with the worst browser availalbe: IE.  So, 
just because browsers are forced to implement these things in order to 
gain any market share, is not a reason for any proprietary extensions to 
be standardised.

In fact, all the elements and attributes available in the Transitional 
DTD have been deprecated for quite some time, and their use is also not 
recommended, and new versions of (X)HTML have dropped transitional 
features completely.

> It should be defined in the dtd. This is the question I am asking -
> why is it not validating?

It will never become a standardised attribute, and will never validate 
in an official HTML DTD.  But if you simply want to get past the 
validator without caring about conformance, you can use a custom DTD [2].

> In other words, why has it not been added, still,
> when many developers are using it successully in every browser...
> at what point does it become a valid attribute, if not then?

Every browser?  Please define what you mean by "every browser".  If 
you've included more than just IE, Mozilla, Opera and maybe Safari, I'll 
be very surprised.  If allow transparency works in all of the following 
plus many more, you may be able to claim "every browser", but I 
guarentee not all of these do:

IE, Mozilla, Opera, Safari, OmniWeb, Konqueror, kHTML, Lynx, Amaya, 
Mosaic, WebTV, and many more...

Lastly, "HTML validation is just a tool" [3] may be a good article for 
you to read, as it will help explain the difference between the formal 
process of validation, and true conformance with HTML.  Although you can 
use a custom DTD to get past a validator, such documents are still 
non-conformant HTML documents.

[1] http://www.w3.org/TR/html401/
[2] http://www.cs.tut.fi/~jkorpela/html/own-dtd.html
[3] http://www.cs.tut.fi/~jkorpela/html/validation.html

-- 
Lachlan Hunt
http://lachy.id.au/
http://GetFirefox.com/     Rediscover the Web
http://GetThunderbird.com/ Reclaim your Inbox
Received on Thursday, 7 April 2005 06:55:58 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 25 April 2012 12:14:18 GMT