[whatwg] size attribute

Ian Hickson ian at hixie.ch
Tue Apr 28 11:33:05 PDT 2009

On Tue, 28 Apr 2009, Kristof Zelechovski wrote:
> The algorithm [1] for converting a character width to pixels is good 
> indeed, except that it should be formulated in terms of glyphs and not 
> characters because stand-alone characters are not displayed or perceived 
> as such for some scripts.

As far as I can tell, "character width" is an adequate term here. I don't 
want to sacrifice readability for pedantism when it's not necessary.

> If the primary font for which the algorithm is being run comprises 
> several scripts, as it typically does, characters of the script of the 
> effective language of the control should take overwhelming weight in the 
> average.

As far as I can tell from testing other browsers, this is not the case.

> This algorithm does not fulfill the requirements for the size attribute 
> formulated in section [2]: when it is applied, it may happen 
> that the user agent will not be able to allow the user to see the whole 
> text it should be able to allow.  The requirement should be modified to 
> match the effect of the algorithm.

The text in doesn't contain a requirement as far as I can tell. 
It's just non-normative text for authors. As such it seems better to be 
simple than precise.

Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'

More information about the whatwg mailing list