On 18/11/13 03:25, Daniel Cheng wrote:
On Mon, Nov 18, 2013 at 12:19 AM, Tab Atkins Jr. <[email protected]>wrote:
On Sun, Nov 17, 2013 at 5:16 AM, Ryosuke Niwa <[email protected]> wrote:
Without starting a debate on what semantics or aesthetics mean, syntax
is a big deal. A bad syntax can totally kill a feature.
Believe me, I agree; I named my last coding project "Bikeshed", after all.
^_^
This is why I find it puzzling that a syntax accepted by the RICG and
a lot of authors is being shot down by a few implementors. This is
why I've been classifying the objections as "personal aesthetic
concerns" - I don't know how to classify them otherwise. The proposed
syntax doesn't seem to offend average authors, who grasp it well
enough (it's a pretty simple translation from what they already liked
in <picture>). It just offends a few of you from WebKit, some of whom
have been a bit hyperbolic in expressing their dislike.
~TJ
I think it's worth pointing out that there are some Chromium/Blink
developers that don't like the multiple attribute syntax either (for what
it's worth, I am one of them).
Yeah, I think this characterization of the debate as "Apple vs the
World" is inaccurate an unhelpful. I think that the src-N proposal is
very ugly indeed. This ugliness creates real issues e.g. if I have
src-1, src-2 [...] and I decide I want a rule that is consulted between
src-1 and src-2, I need to rewrite all my attribute names. Whilst this
might produce a pleasant rush of nostalgia for children of the 80s
brought up on 8-bit Basic, for everyone else it seems like an
error-prone drag.
So I think the question is not "is this proposal unpleasant"; it is. The
question is "is this less unpleasant than the alternatives". That is
much less clear cut, and there is room for reasonable people to disagree.
