Re[17]: css with attribues [hardware]

David,

>>>>>> non-realtime UA can parse after downloading of csas-file.
>>>> DD> That still means they have to built an in memory DOM tree and
>>>> modify it based on the stylesheet.
>> DD> This makes them harder to write, require more memory and CPU time

User gets benefit (unnecessary to keep separation in brain),
machine gets loss (necessity to update tree by attributes from css).
Total: benefits of user is more important, than uncomfort of machine.

>> DD> slower,
>> Not actual for non-realtime UA.
DD> (a) If something needs more processing, then it will take
DD> more time given the same hardware.

  Hardware of less company has variable load (burden) during day.
  Hardware of bigger company (which works with all pages of Internet,
or of its region [e.g. RuNet]) is apparently burdened wholly
during whole day. But they govern money, but not we govern their ones.
  What is about cost of spent resource for whole society, then
wasting of resource is justified: as i already said in another letter,
1) CAS will reduce manual job
("multiply to quantity of population, than to quantity of
pages on sites - and you will get real number of duplications,
which are quite not little")
2) CAS allow to write groups, got as result of classification,
and so promote structurezed thinking.

DD> (b) What about "realtime" UAs?

It is UA, which give results to end of time,
which user can (or want) wait.
E.g. if UA is browers, then result is rendering,
time is equal 1 second;
if UA is electronic storage, then result is filling database,
time depends of destination of this storage
(in any case, it is much more, than 1 second).

>>>> upgrade of robots (to append attributes from css-file to tree of
>>>> elements)
>> DD> As I've explained, this is NOT a "relatively small job". It would
>> DD> require major rewrites for many user agents.
>> Following your logic,
>> community must not enter new attributes into tags,
>> because it would require major rewrites for many user agents.
DD> No it wouldn't. Adding new attributes to the markup language wouldn't
DD> require some user agents to completely change the way they process
DD> the page.

"Append attributes from css-file to tree" is also "adding"
(but not "change" - as you are writing).

Also, as i already said, authors don't use standards,
while browsers don't support them.



Dmitry Turin
SQL5      (5.9.1)  http://sql50.euro.ru
HTML6     (6.5.1)  http://html60.euro.ru
Unicode7  (7.2.1)  http://unicode70.euro.ru
Computer2 (2.0.2)  http://computer20.euro.ru

Received on Thursday, 17 January 2008 08:26:32 UTC