Re: Re: C, MathLink or Java, J/Link
- To: mathgroup at smc.vnet.net
- Subject: [mg28110] Re: [mg28075] Re: C, MathLink or Java, J/Link
- From: murphee <werner.schuster at netway.at>
- Date: Sat, 31 Mar 2001 02:58:45 -0500 (EST)
- References: <99usl9$61j@smc.vnet.net>
- Sender: owner-wri-mathgroup at wolfram.com
At 04:12 30.03.01 -0500, you wrote:
>J/Link is great for the creation of new user interaces. But
>in general Java is too slow, especial if you are looking for
>speed.
That depends.
A hand optimized algorithm that uses all the bells&whistles
of the targeted CPU (like Altivec/MMX/SSE[2]/3DNOW!/...)
will of course always be faster than the most optimized
JavaCode.
But if a programmer wants to write the new Quake3 Successor in
Pure Java (without using eg Java3D ), just because "Java is hip!" and
without some other Justification/Reason, he has to accept to be called
"Naive Idiot"
(or whatever derogatory term you prefer);
>Here at the computer graphics department we had serval
>web-based visualisation applications (2d, 3d) and
>for these applications Java was selected as an "ideal"
>language ...
Selected by whom?
By some manager that has just read an article about
Java in some glossy BusinessMagazine and thought
"Now with gotta do somethin' with that Java thingy,
cause its keeewl!"
The problem with Java was/is the huge hype.
Everyone said: "We gonna do something in Java"
because it was hip to do so.
So now, huge Javaprojects were launched with great Publicity,
although, eg the company didn't have any experienced Java Programmers
("Ahh, the C++ Programmers can do that, its Syntax similar...").
And then, if such a Project failed (or produced suboptimal, slow code),
Java was blamed. And that image damage (for Java) was even bigger,
because everyone was watching those Projects, because they were
done with that new JavaThing.
You also have to keep in mind that the majority of launched Projects
fail anyway (for several reasons, mainly bad Management), but especially
if the Participants were rushed into it & if the Expectations/Goals were
utopic.
And that is the problem:
You mustn't decide to use a technology just because its
shiny & hip & new. You have a problem & look around
for the *best* technology to solve this specific Problem.
If, after a thorough analysis, the best Technology is eg
C++ (eg, because the company has many experience C++
Programmers), then use that, and don't base such vital
decisions on Ramblings from Suns Marketing Dept. or
glossy (but clueless & twofaced) Magazines.
>*All* these applications switch back to C/C++, only
>due incredible slow speed.
Well, you can write slow/bad code in **any** Language.
Java is not the cure for inexperience.
>An applications, that are more complex than to play
>Tick Tack Tao are too slow.
Allright, is that supposed to be a rational statement?
Have you even used Java since 1.02, or for more
than mere Animations Applets.
(I don't mean that as an insult to you Jens, just asking
for your background, to see whether you have the
Qualifications to say something like that).
Also a warning to anyone seriously considering Java:
Please ask experts (real Experts, like the one to be
found on www.javalobby.org, or the comp.lang.java.*
Newsgroups), since nowadays everyone who know
the difference between Java (the computer thing)
and Java (the coffee thing) thinks he/she has the
ultimate wisdom on it (again, Jens, this is really not
meant as an insult or flamebait to you; just general
to warn People from those guys. I don't want to
cause any trouble or upheaval or flamewars in this group);
Werner Schuster
murphee
--
************
my site: www.angelfire.com/co/werners *NEW*
a must: www.linuxtoday.com
so is this: www.javaworld.com
as well as: www.userfriendly.org/static