Re: [css-variables] CSS Variables are a NEW kind of variable

On Wed, Jun 13, 2012 at 3:20 PM, Chris Eppstein <chris@eppsteins.net> wrote:
> On Wed, Jun 13, 2012 at 1:52 PM, Tab Atkins Jr. <jackalmage@gmail.com>
> wrote:
>> On Wed, Jun 13, 2012 at 1:38 PM, Chris Eppstein <chris@eppsteins.net>
>> wrote:
>> > If they are normal properties, why can I not access the value of other
>> > properties?
>>
>> That's a different question - you're now asking about the var use
>> syntax, and why you can't use a similar syntax to access other
>> properties.
>>
>> It's because the rules for whether or not you're allowed to (avoiding
>> cycles from implicit dependencies) are complex, and we'd like to be
>> able to change them over time without worrying about breaking
>> arbitrary web content, or making the rules even more complex.
>> Variable properties have a very simple and explicit set of
>> dependencies - the properties tied to variables used in them - and we
>> won't be changing that in the future (at least, not in a way that's
>> potentially backwards-incompatible).
>
> It's not a different question. They are special and so they are not the
> same. If they are the same, make them the same as properties in all cases
> including read. Otherwise, this is just confusing.

I've given my reasoning for both the similarity and distinction.
Consistency is an excellent and important quality, but we shouldn't
fetishize it.  It's okay for a new class of things to be different in
some ways if the difference has a good reason to exist.  That doesn't
automatically imply that the small difference makes them a completely
new and different thing, and it definitely doesn't automatically imply
that the new class should *look* different, syntax-wise, from the
older things.

~TJ

Received on Wednesday, 13 June 2012 22:50:00 UTC