Re: propdef tables for shorthands

On Thu, May 5, 2016 at 11:17 AM, Alan Stearns <stearns@adobe.com> wrote:
> On 5/5/16, 11:10 AM, "Tab Atkins Jr." <jackalmage@gmail.com> wrote:
>>On Mon, May 2, 2016 at 9:55 AM, fantasai <fantasai.lists@inkedblade.net> wrote:
>>> We've been leaving most fields in shorthand propdef tables
>>> as "see individual properties", but, I think going forward
>>> we should fill these in if all sub-properties have the same
>>> value. This is more useful to people looking things up in
>>> the spec, especially as we are encouraging people to use
>>> shorthands over longhands in many cases.
>>
>>I disagree.  Doing so would suggest that shorthands persist somehow in
>>the CSS data model, when in reality they're expanded into longhands at
>>the very beginning of the cascade process.  They don't apply to
>>anything, they don't inherit, etc.  It doesn't help authors build a
>>good mental model to suggest otherwise.
>>
>>It's also an editing hazard - if we change any subproperty such that
>>not all of them are identical, we have to remember to update the
>>shorthand as well; if we forget, we have confusing incorrect
>>information.
>
> Would it be possible for bikeshed to keep track of these dependencies?

Theoretically, yes. (And I'd like to do so, so I can generate better
indexes.)  But even if I remove the editting hazard by making it
automatable, I still disagree with it for the reasons given in the
first paragraph of my response. ^_^

~TJ

Received on Thursday, 5 May 2016 18:23:43 UTC