W3C home > Mailing lists > Public > www-style@w3.org > October 2011

Re: [css3-*] Defining support for the inherit keyword

From: Peter Moulder <peter.moulder@monash.edu>
Date: Fri, 28 Oct 2011 14:56:35 +1100
To: www-style@w3.org
Message-id: <20111028035635.GA16045@bowman.infotech.monash.edu.au>
On Thu, Oct 27, 2011 at 11:53:07PM +0000, Sylvain Galineau wrote:
> 
> [Tab Atkins Jr.:]

> > I really don't want every property to have to include " | inherit |
> > initial" in its value definition.
> > 
> If they are valid values of the property, I really couldn't care less about saving you
> 2 seconds of copy/paste time per property
> ...
> 'Nothing because I'd rather do a little less work' is not acceptable though. 

I didn't see Tab or anyone else complain about the effort involved for
spec editors: Tab's words could just as well be read as an opinion that
it's easier for readers if readers don't have to check for each property
whether 'inherit' or 'initial' are valid values.

> The preference of those spec users who depend on them to implement
> features takes precedence. For the latter, incomplete syntax
> definitions are a problem.
> ...
> e.g. we 
> could add a " | <core values>' link to the section in css3-values that covers all global
> values. That would make the dependency of each property on these two values explicit and
> would be fine. It could also be a separate row in the property definition table, right
> under the property-specific values...Other suggestions welcome.

One approach would be to change the property tables from saying

  Value: <integer> | inherit | initial

to

  Squiggle: <integer>

(after finding a better name than "Squiggle"), and just make sure that we
specify in a relevant place what counts as a valid "property : value"
declaration in a stylesheet by reference to the Squiggle of the property.

The Squiggle is closely related to what the phrase "specified value" has
come to mean; though I hesitate to name the field "Specified value",
partly because the field's main purpose is to specify syntax rather than
the specified value as such (and partly because I find "specified value"
a misleading name given that it differs from what's written in the
stylesheet, which is most directly what's specified; I understand the
history, it's just unfortunate).

pjrm.
Received on Friday, 28 October 2011 03:57:14 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:45 GMT