W3C home > Mailing lists > Public > www-style@w3.org > July 2004

RE: [CSS21] Test Suite

From: Mark Moore <mark.moore@notlimited.com>
Date: Fri, 23 Jul 2004 19:35:06 -0700
To: <www-style@w3.org>
Message-Id: <20040724023842.EDBA4A14E7@frink.w3.org>
I'm hoping this casts some objective light on the decision to use GIF's or
PNG's in the CSS 2.1 validation tests.

As a quick approximation, I used the arbitrary ranking from Hitwise [1], and
grepped the web pages of the top 5 non-Microsoft sites.  (I've arbitrarily
removed Microsoft sites simply to minimize any perceived bias.)

Here are the image formats referenced in the corresponding pages [2]:

1) Yahoo! Mail / Yahoo! / Yahoo! Search / My Yahoo!
2) Google
3) eBay
4) USA Today
5) Pogo

---- yahoo ----
     44 gif
      5 jpg
      4 js
---- google ----
      1 gif
---- ebay ----
     18 gif
     22 js
---- usatoday ----
      1 aspx
     33 gif
      1 gifquote
     11 jpg
      7 js
      1 png
      1 swf
---- pogo ----
     27 gif
      2 jpg

As I mentioned earlier, GIF's may be inferior from a standards and personal
preference perspective, but they seem to be far more widely deployed.  This
is a direct consequence (IMHO) of their broader, more consistent rendering
in existing UA's.

BTW, does anyone know of a resource that characterizes or analyses HTTP
traffic?  That might be a less subjective way of determining which image
format is more widely deployed.

-MM

[1] http://www.hitwise.com/info/whatwedo/index.html?STR_product=rankings
[2] If you're interested in reproducing the results, I've attached the
source files.  See ttt.pl for the HTML scraping, and here's the bash command
that generated the table:

for f in yahoo google ebay usatoday pogo; do echo "---- $f ----"; ttt.pl $f*
| egrep -i '\.[a-z]+$' | sort -u | sed -e 's/^.*\.//' | sort | uniq -c; done



> -----Original Message-----
> From: www-style-request@w3.org [mailto:www-style-request@w3.org] On Behalf
> Of Chris Lilley
> Sent: Friday, July 23, 2004 11:21 AM
> To: Henri Sivonen
> Cc: www-style@w3.org
> Subject: Re: [CSS21] Test Suite
> 
> 
> On Friday, July 23, 2004, 7:04:12 PM, Henri wrote:
> 
> HS> On Jul 23, 2004, at 00:17, Chris Lilley wrote:
> 
> >> They look entirely different on Mac and PC platforms, to take an
> >> example.
> 
> HS> That depends on the display settings.
> 
> Which PNG images can contain and thus display correctly while GIFs
> cannot.
> 
> 
> 
> 
> 
> --
>  Chris Lilley                    mailto:chris@w3.org
>  Chair, W3C SVG Working Group
>  Member, W3C Technical Architecture Group



Received on Friday, 23 July 2004 22:38:43 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:31 GMT