Re: @import -- allow at any place in stylesheet.

18.01.2012, 02:52, "Tab Atkins Jr." <jackalmage@gmail.com>:
> 2012/1/17 Marat Tanalin | tanalin.com <mtanalin@yandex.ru>:
>
>> š18.01.2012, 02:31, "Tab Atkins Jr." <jackalmage@gmail.com>:
>>> š2012/1/17 Marat Tanalin | tanalin.com <mtanalin@yandex.ru>:
>>>> šš18.01.2012, 01:35, "Tab Atkins Jr." <jackalmage@gmail.com>:
>>>>> ššStepping outside this problem, you shouldn't really be using @import
>>>>> ššin your stylesheets anyway. šIt's bad for performance, since it makes
>>>>> ššyour stylesheets download one-by-one instead of in parallel. šMaking
>>>>> šš@import more powerful might encourage more use of this, which I'd
>>>>> ššpersonally prefer to avoid. šIt's better to either link in multiple
>>>>> ššstylesheets via multiple <link> elements (so the browser can start
>>>>> ššdownloading all of them at once), or use a preprocessor to munge the
>>>>> ššsheets together into a single file so you can pull them all down with
>>>>> šša single request.
>>>> ššAs I've already said, there are different situations: internet, intRAnet, local webpages opened from CD. Delay concern is generally applicable to former one only, but current limitation is applied forcedly and harmfully to all of them.
>>> šWe only really care about the web, since the vast vast majority of
>>> špages using our tech are web pages. šMaking choices that are bad for
>>> šthe web but offer a minor benefit to non-web usage isn't a good
>>> štradeoff.
>> šPlease don't speak/decide for all web-developers. They should have flexibility.
>
> I'm not. šI'm speaking for the community *designing* the web platform.
> š*We* care mostly about the web, and won't hurt the web in order to
> optimize non-web cases unless it's, like, crazy good.

What do hurt web is creating specs with unnatural theoretical limitations. XHTML2 was one of such specs, and that's why it's dead.

Received on Tuesday, 17 January 2012 23:08:23 UTC