[css-syntax][naming] Component values

Le 27/06/2013 00:38, Tab Atkins Jr. a écrit :
> On Wed, Jun 26, 2013 at 4:33 PM, fantasai <fantasai.lists@inkedblade.net> wrote:
>> On 06/26/2013 03:02 AM, Simon Sapin wrote:
>>> Le 26/06/2013 00:12, L. David Baron a écrit :
>>>> I'm a little concerned about the "component value" terminology; I'd
>>>> prefer using a term that doesn't involve "value".  But I don't have
>>>> a better idea right now.
>>>
>>> Previous revisions of this spec used "primitive" for this concept, until
>>> we noticed that the concept is the same as in Values
>>> & Units:
>>>
>>> http://dev.w3.org/csswg/css-values/#value-defs
>>>
>>> But I understand the desire to make a distinction. Does "primitive" sound
>>> better than "component value"?
>>
>> Maybe "component primitive" instead of "component value"?
>> Then at least it ties back to the concept of "component value",
>> but implies it's a slightly lower-level concept.
>
> It's really not a lower-level concept, though.  It's exactly the same
> level as V&U's "component value".  No additional filtering/processing
> happens to the values after this point.

That is correct, but I still dislike the "component value" name. It 
doesn’t make sense outside of declaration values, the only context where 
Values & Units uses it. (The whole {} block for a style rule is a 
value?) It’s also longer than I would like, for something used as often 
as it is in Syntax.

In programming languages, they would often be called "nodes" in the AST 
(abstract syntax tree.) But if we consider the AST to be the output of 
the various css-syntax, algorithms declarations and rules are AST nodes 
too, so we still need a term for other kinds of nodes.

In the Department of Generic English Words, how about "item"?

-- 
Simon Sapin

Received on Thursday, 27 June 2013 09:42:02 UTC