D
David Mark
jdalton said:
As noted in the recent review of QSA tack-ons, your test is rigged
(compares QSA to a script). Thanks for trying though.
[snip bogus results[
As noted in the recent review of QSA tack-ons, your test is rigged
(compares QSA to a script). Thanks for trying though.
[snip bogus results[
J
jdalton
Hi David,
I don't think prending your argument is valid is getting you anywhere
(honestly it's pathetic). Everyone can see that the overwhelming
majority of browsers I have been reporting don't use QSA. Nothing is
rigged, you simply can't accept that you have *failed* to produce a
faster/more complete alternative.
I don't think prending your argument is valid is getting you anywhere
(honestly it's pathetic). Everyone can see that the overwhelming
majority of browsers I have been reporting don't use QSA. Nothing is
rigged, you simply can't accept that you have *failed* to produce a
faster/more complete alternative.
J
jdalton
I don't think pretending...
D
David Mark
jdalton said:
What argument is that? Your quoting is abysmal (as always). Did you
mean the argument that demonstrated you are completely full of shit for
trying to pass off a QSA vs. DOM comparison as a valid test? Your
excuse was that QSA is "buggy in all browsers" (or something like that)
and the valiant "majors" have been feverishly working to combat these
issues, but my add-on is oversimplified because it doesn't include the
slew of workarounds present in theirs. Well:-
http://groups.google.com/group/comp.lang.javascript/browse_thread/thread/7ee2e996c3fe952b#
All I've seen is that your tests are full of holes. And there's not a
single shred of evidence that the non-QSA browsers show mine as "one of
the slowest". Quite the contrary, I run (relevant) tests on non-QSA
browsers all the time and mine is always one of the fastest (usually the
fastest by a large, even exponential margin), particularly in older
and/or limited environments where it counts.
I consider your "results" to be nothing more than random numbers. And
given your track record of one half-truth (or outright lie) after
another, they could very well be made up. I really don't care at this
point (and hard to imagine anyone else does either, excluding your
various alter ego sock puppets).
And the others are all ill-equipped to handle non-QSA browsers as they
can't even read attributes straight, even with browser sniffing in
place. It's spelled f-o-l-l-y. Get it?
Not hardly. You simply can't admit that you don't know what you are
doing. And you need to stop focusing on queries anyway. How many times
do I have to tell you that they are the least important issue at hand.
Trying for weeks to come up with a query-based test that proves my
library is "one of the slowest" just makes you look like a jackass.
And "more complete?" Are you kidding? Mine works on virtually
anything, past, present and (very likely) future. Theirs are software
of the month clubs trying (and failing) to keep with just the latest
versions of three or four browsers (in their default configurations).
One is almost zero cost of ownership, the others are bottomless money
pits. Get that?
D
David Mark
jdalton wrote:
[...]
So what do you have here anyway? Looks like the typical hand-picked
result set. Why not stay consistent from one post to the next? And why
all QSA-less browsers this time around? Why do we need your hacked
version of SlickSpeed for that? Wasn't the whole point to show more
detail than the 0ms QSA results?
Assuming these are less like random numbers than the previous batches
(and aren't made up).
As mentioned, all of the others are disallowed in compatibility mode
(for reasons that should be obvious to anyone who actually reads this
group). Hint: they can't deal with buggy MSHTML attribute methods.
They don't "punt", they throw a Hail Mary into the stands.
So you threw out your previous attempt at "more accurate" tests and came
up with a new batch. And these incomplete results hardly indicate mine
is "one of the slowest". It's unclear if they indicate anything more
than delusion on the part of "jdalton" and a couple of sock puppets.
Odd choices. Do you throw darts at a list of browsers? Or perhaps you
cherry-pick the results that appear to support your ridiculous claims?
Opinion appears divided on the validity of your tests (and the relevance
of the results). Richard says they are meaningless, you say they mean
everything. Who are we to believe?
I say none of these rapid-fire query tests has much in the way of
practical value. The TaskSpeed tests are clearly a better approximation
of a Web application. But at least the other SlickSpeed variations (all
but yours) have some modicum of acceptance in the industry.
Every other test out there shows mine as fast (or much faster). Your
pet test, which doesn't seem to have any practical merits at all, shows
it as middle of the pack at worst. So your ravings about "one of the
slowest" would seem to indicate you are a crackpot.
[...]
So what do you have here anyway? Looks like the typical hand-picked
result set. Why not stay consistent from one post to the next? And why
all QSA-less browsers this time around? Why do we need your hacked
version of SlickSpeed for that? Wasn't the whole point to show more
detail than the 0ms QSA results?
Assuming these are less like random numbers than the previous batches
(and aren't made up).
As mentioned, all of the others are disallowed in compatibility mode
(for reasons that should be obvious to anyone who actually reads this
group). Hint: they can't deal with buggy MSHTML attribute methods.
They don't "punt", they throw a Hail Mary into the stands.
So you threw out your previous attempt at "more accurate" tests and came
up with a new batch. And these incomplete results hardly indicate mine
is "one of the slowest". It's unclear if they indicate anything more
than delusion on the part of "jdalton" and a couple of sock puppets.
Odd choices. Do you throw darts at a list of browsers? Or perhaps you
cherry-pick the results that appear to support your ridiculous claims?
Opinion appears divided on the validity of your tests (and the relevance
of the results). Richard says they are meaningless, you say they mean
everything. Who are we to believe?
I say none of these rapid-fire query tests has much in the way of
practical value. The TaskSpeed tests are clearly a better approximation
of a Web application. But at least the other SlickSpeed variations (all
but yours) have some modicum of acceptance in the industry.
Every other test out there shows mine as fast (or much faster). Your
pet test, which doesn't seem to have any practical merits at all, shows
it as middle of the pack at worst. So your ravings about "one of the
slowest" would seem to indicate you are a crackpot.
J
jdalton
Hi David,
effort.
because you have failed to research them.
created by MooTools/Dojo devs when you bash them and their
frameworks ?
anything.
I didn't say "feverishly working to combat", but they do put forth an
effort.
Yes it is overly simplified. You are completely ignorant to the issues
because you have failed to research them.
Denial. I am curious how do you rationalize using a benchmarking tools
created by MooTools/Dojo devs when you bash them and their
frameworks ?
Delusional. You aren't running my tests or tests that accurately show
anything.
There is no half truths. I have been consistent.
They aren't.
Check the IP addresses from the other posters non map to me.
You are hung up on the few attribute bugs they have. You don't get it.
Then why do you promote your false performance ?
Weeks ? I just posted the Slickspeed tests yesterday.
Mindless ranting. Ahh that creepy wink again.And "more complete?" Are you kidding? Mine works on virtually
anything, past, present and (very likely) future. Delusional.
Theirs are software
of the month clubs trying (and failing) to keep with just the latest
versions of three or four browsers (in their default configurations).
One is almost zero cost of ownership, the others are bottomless money
pits. Get that?
D
David Mark
jdalton said:
Again, this is a newsgroup.
Would that be yesterday's tests or today's? Yesterday's were supposed
to be "more accurate" than something in some way. Today they are
garbage? What about tomorrow's?
Anyone who can read English knows that is false.
Pardon me if I don't take your word for it. You've proven yourself to
be dishonest.
Excuse you? There are two sets of tests on my site. I've said numerous
times which I consider the more relevant. And I didn't invent or
encourage acceptance of either. They are just there.
You are not one to talk about creepy. You come off like an obsessed fan
with your tribute page and seemingly endless fixation with my script.
But I know you are simply trying to stop me from derailing whichever of
the "majors" is your current favorite. Good luck with that!
J
jdalton
David,
I guess people have access to both benchmarks (yours and mine) as well
as the source for both. They can make up their own minds. I am not
cheerleading for any major framework and don't have a fixation on your
script. I just think you are overly critical towards others (endless
nitpicking in code or posts, personal attacks, character
assassination) when your own work is flawed. I wish you treated others
in clj as good as you do in your own g-group (it shows you have the
potential for pleasantness). Good luck.
I guess people have access to both benchmarks (yours and mine) as well
as the source for both. They can make up their own minds. I am not
cheerleading for any major framework and don't have a fixation on your
script. I just think you are overly critical towards others (endless
nitpicking in code or posts, personal attacks, character
assassination) when your own work is flawed. I wish you treated others
in clj as good as you do in your own g-group (it shows you have the
potential for pleasantness). Good luck.
D
David Mark
jdalton said:
I don't have any benchmark. I use what is out there. And you know
_full well_ that the typical library aficionado is neither capable nor
willing to interpret the source. Richard, on the other hand...
Bullshit. You know you lied. Just admit it. You deliberately posted a
test that pits QSA against non-QSA without a disclaimer. That's not
just goofy, but starkly dishonest. Your excuse about "buggy browsers"
and valiant workarounds by the others was shredded to ribbons in five
minutes:-
http://groups.google.com/group/comp.lang.javascript/browse_thread/thread/7ee2e996c3fe952b#
....yet you persist.
That doesn't fit the evidence (certainly not the latter bit).
I think you are a disingenuous, obsessed, attention-starved asshole of
the first order. And a complete non-entity in this industry to boot.
How about that?
You get what you put in. Have a nice life (or whatever you have that
passes for one).
A
Andrea Giammarchi
TaskSpeed is a contest itself, so, is your one a question?
What I am saying is that PureDOM, as baseline, is fine to me.
A TaskSpeed test not focused on tasks as they are but focused only
into "hacks to score" won't be useful, neither it will demonstrate
anything.
// PureHack
dom.innerHTML = "whatever";
// some lib
function $(dom, whatever) {
dom.innerHTML = whatever;
};
$(dom, "whatever");
----------------------
is there any interest into above test? Consider the DOM a library
itself, with methods to do things in a "linear" way to solve tasks.
This is PureDOM, a baseline to compare DOM methods with libraries
methods, and this makes sense to me while the precedent example test
would be pointless.
It's not about being defensive, it's just about understanding PureDOM
column.
Regards
J
jdalton
For those following along I have moved my discussion to
"QSA--buggy in lots of browsers?"
http://groups.google.com/group/comp.../thread/7ee2e996c3fe952b#msg_b16895423bcb4731
"QSA--buggy in lots of browsers?"
http://groups.google.com/group/comp.../thread/7ee2e996c3fe952b#msg_b16895423bcb4731