W3C home > Mailing lists > Public > www-style@w3.org > August 2012

Re: [css-variables] Different syntax for variable definition and use

From: Brian Kardell <bkardell@gmail.com>
Date: Wed, 29 Aug 2012 19:47:38 -0400
Message-ID: <CADC=+jciEET2JrfGQy_=20brvzbCOP8N1qVLRMqSQpHzf7JsGw@mail.gmail.com>
To: "Jens O. Meiert" <jens@meiert.com>
Cc: www-style@w3.org, "Tab Atkins Jr." <jackalmage@gmail.com>
On Aug 29, 2012 1:34 PM, "Jens O. Meiert" <jens@meiert.com> wrote:
>
> > > “var-foo” appears to mean the variable is called “var-foo”; but with
> > > “var(foo)”, the variable seems to be called “foo".
> >
> > I think it's pretty clear: "var-foo" means you're declaring a property
> > for a var named foo.  Change the - to a space, the : to an =, and
> > you're practically writing JS! ^_^
>
> Since nobody here is a linguist:
> “The hyphen (‐) is a punctuation mark used to join words and to
> separate syllables of a single word.” [1]
>
> Wikipedia does not say anything like this about parentheses.
>
> The suggested syntax is confusing.
>
>
> [1] http://en.wikipedia.org/wiki/Hyphen
>
> --
> Jens O. Meiert
> http://meiert.com/en/
>

I would like to state that I don't find this line of reasoning particularly
compelling: The majority of CSS authors are not linguists either... but
they are intimately familliar with css and with the modularization of CSS
we now have an almost ubiquitous pattern where the left hand side is set
explicitly with "module-pattern" style and shorthands are just the module
name and functions are used on the right to access or calculate values.
That fits this model perfectly, think about it.

WRT left right agreement, I think anyone with a sense of that has likely
been biased by an outlier that taught them that, because of all of the
various diverse languages that they may use, the vast majority of them use
a keyword (space) variable name on the left to define and functions are
regularly used to access values.  In other words, a seeming disagreement
isn't exacly so anyways.  Literal laguage comparisons beyond that, no
matter how argued are at least as likely as not to be seen the differently
by different authors of CSS.

When we ran this out in public, without fail, there was a common confusion
in members of each group who independently misread it the same way thinking
that they were both usage - and rightly so because they are coming at it
from a macro/preprocessor pov, which this isn't.

This proposal has had no such confusion and calling them something like
custom properties or author defined properties clears up almost everything
else.

There are lots of macro ideas, and there have been lots in the past.  This
answers questions that those couldn't/can't.  I would really hate to see
something so clearly useful, getting so much progress, acceptance and
attention get held up because of this issue with it.

Anyway, that's my two cents, but I am no one special :)
Received on Wednesday, 29 August 2012 23:48:07 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:59 GMT