W3C home > Mailing lists > Public > www-style@w3.org > April 2008

Re: Border model inconsistencies

From: Bert Bos <bert@w3.org>
Date: Tue, 08 Apr 2008 23:52:32 +0200
Message-ID: <47FBE920.5060100@w3.org>
To: www-style@w3.org

Keryx Web wrote:
> 
> Has this issue been discussed?
> 
> http://www.onderhond.com/blog/onderhond/border-model-inconsistencies

No, it has not.

But I'm not sure it's an issue for CSS.

You ask what color the corner pixel is of a 1px thick border that has 
different colors on the different sides. But is that a meaningful question?

As the quoted page itself says, one useful way to join borders in the 
corner is with a diagonal. Devices may differ in how well they can 
render a diagonal. A modern printer has no problem to put a diagonal end 
on a 1px thick line. A screen, which has a much lower resolution, has to 
find some trick. (Although 200dpi screens are already available, see, 
e.g., the XO laptop from OLPC, which has an interesting approach to 
screen resolution.)

Different implementations may differ in how much effort they put into 
delivering the best possible quality. CSS doesn't say what kind of 
anti-aliasing must be used, or even that it must be used at all; it 
doesn't say what algorithm must be used to justify text (and yet that is 
one of the most important factors in the perceived quality of the 
typography); it doesn't say how fast a UA must render a page, nor what 
the maximum size of a page is that a UA must accept. That is left to 
competition and available resources.

And about corners, the Backgrounds and Borders draft currently even says 
explicitly that the way to join different borders is undefined.

B.t.w., the WG did discuss a bit if anything needed to be said about the 
color of rounded corners when the sides are of different color. A smooth 
transition between the colors probably looks best in that case. CSS will 
probably not require that UAs render the corners like that, but we plan 
  to suggest it. (And we won't define what a "smooth transition" is, 
whether that is linear in RGB, S-shaped in HSL, etc...)



Bert
-- 
   Bert Bos                                ( W 3 C ) http://www.w3.org/
   http://www.w3.org/people/bos                               W3C/ERCIM
   bert@w3.org                             2004 Rt des Lucioles / BP 93
   +33 (0)4 92 38 76 92            06902 Sophia Antipolis Cedex, France
Received on Tuesday, 8 April 2008 21:53:17 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:55:04 GMT