Re[27]: css with attribues [hardware]

David,

>>>> should download css before parsing html.
DD> But CSS is supposed to be an optional presentational layer. The
DD> operative word should be "may" not "should".

  You have convinced me, CAS (third layer: with attributes,
but without properties) would more suitable, than CSAS.
  
P.S.
  'optional' theoretically or practically ? I know people,
who browse without downloading pictures (e.g. I),
but i know nobody, who without css.
  As was mentioned, one css-file per hundreds other files of site
is less one persent of traffic, so this is not actual for robots.

---

>> DD> Lynx is a browser. Lynx does not support JavaScript.
>> It cover very little part of population.
>> To my mind, it's anachronism.
DD> To my mind it is a small, fast, useful tool.

  Size of program is not important in epoch of DVD, BluRay, etc.
  It's not important, how much time of rendering is less constants
of human perception (1 sec): 0.05 sec or 0.5 sec
(comparing browsers in identical conditions, i.e. for textual framework).
  And without pictures, it's not useful - observe today's inet.

>>>>>> Nothing disturb author to write in previous manner.
>>>> DD> Initial authors perhaps, but not maintainers who have to find
>>>> DD> out where the previous author decided to put the information.
>> DD> URIs that are being linked to,
>> These are places with unique information -
>> they will remain in htm-file.
DD> Your proposal would allow them to appear in the stylesheet.

Yes, any technology allows author to harm himself.



Dmitry Turin
SQL5      (5.10.0)  http://sql50.euro.ru
HTML6     (6. 5.2)  http://html60.euro.ru
Unicode7  (7. 2.1)  http://unicode70.euro.ru
Computer2 (2. 0.2)  http://computer20.euro.ru

Received on Tuesday, 22 January 2008 14:47:54 UTC