On 2011-11-29 21:24, Tab Atkins Jr. wrote:
On Tue, Nov 29, 2011 at 12:18 PM, Julian Reschke<julian.resc...@gmx.de> wrote:
On 2011-11-29 20:07, Tab Atkins Jr. wrote:
...
Since XPath and Selectors are 95% overlapping in functionality and
100% overlapping in goals and overall structure, I believe it is a bad
idea to try and develop both for the web platform. Instead, we should
...
The proposal wasn't to "develop" XPath as in "doing more work on the
expression language".
What *was* proposed is to fix/standardize the API for XPath, which already
is part of the web platform.
Speccing, implementing, and testing an API is also "developing",
particularly when it appears there's a desire to spec, implement, and
test a new, more convenient API that's currently proprietary to a few
browsers. For the reasons stated in my email, I believe this effort
to be a bad idea.
Well, if people are ready to do the specification, implementation and
testing work, what's the reason to stop them? Yes, I know, "opportunity
cost". But guess what, that's not going to be a convincing argument for
those who really want that feature.
Best regards, Julian
