Jack actually points out the fact that "div div div" is pointless. He
also gives a lot of kudos to jQuery and I'm glad for that. He even
mentioned John's nth child script and how well-written it is. I like the
fact that he, in a roundabout way, calls out other libs on the use of
pointless selectors and the tweaking towards XPath.
I would venture to say that when SlickSpeed was first posted by
MooTools, very few developers actually compared the results in IE. IE is
still the dominant browser and if you look at the performance of the
libs in IE, you'll see that jQuery really performed well while others
fell on their face.
While the results don't show us to the be the fastest, I think the post
helps to validate John Resig's position that the selectors in these
tests don't represent real-world examples and certainly shows the lack
of consistent performance by other libs when using the leading browser.
In terms of performance, we're well within effective speed ranges so as
you mentioned, Glen, 10ms is irrelevant. I see this as Jack's way of
quieting the other groups and not specifically targeted at jQuery.
Rey...
Glen Lipka wrote:
http://extjs.com/blog/2007/07/10/css-selectors-speed-myths/
<rant>
"div div div", we are slowest by like 50 ms, but "div div" we are not
slowest, and within 9ms of the leader.
Who uses "div div div"?
<pun>
What "div"ference does it make?
</pun>
Plus, some of the libraries are packed, and some are not. (jQuery is not)
Would that change the results?
Are we really nitpicking about 10's of milliseconds?
No mention of file size again. No mention of capabilities in the core.
</rant>
Glen
--
BrightLight Development, LLC.
954-775-1111 (o)
954-600-2726 (c)
[EMAIL PROTECTED]
http://www.iambright.com
