You either encountered poorly designed applications, or your negative
experience stems from the pre-1.5 Java GUI. Sun was making steady progress,
improving Swing performance, and as of Java 1.5 (actually, even 1.4.2) the
GUI response is quite comparable with the native apps.

If the server runs Windoze - the weekly (never mind by-weekly) reboot is a
fact of life. Whether the app is written in Java or any other platform is
immaterial. The crap comes from M$ OS.

"Real programmers don't eat quiche". Statements of this kind are a hallmark
of immature wizkids that hold on to this belief until they get into the
'real' world. Then they quickly discover that:

a. lots of 'real' aplications are written in Java (Oracle Management Tools,
for a quick reference)
b. 'real' engineers work with what is most suitable for the task at hand. If
the task calls for a robust, OO, cross-platform, multi-vendor solution, Java
is the way. If the task calls for a distributed, or web-based solution, Java
is the _best_ way.

Alex Molochnikov
Gestalt Corporation

Everybody lives in their own world or company and do what they known. I
have worked at a lot of companies. I would recommend that you do the
same because maturity is a funny thing..

you are 1 person saying so. So you agree with the whole

oh no, the java police.  and they call names too. .  Name-calling helps
to win arguments if your audience are a bunch of fools.  I guess you
consider yourself a leader of lemmings, pointing out trolls so the
helpless ppl(pimples)  don't get lost like lemmings.

so why don't you tell us why java apps aren't mainstream. And do so in
a way that won't make you a troll by your own undocumented definition
of the word.

or is it that anybody that dares claim that java apps aren't mainstream
is a heretic

Oh, you don't like me posting in a way that is semantically anonymous.
If I had called myself   FlintBlah  or PotOfSoup like many people call
themselves maybe that'd be better for you, that is less anonymous, is
it? 'cos that'd really gives away my identity!  Or if I used a fake
name like many people do,  that'd be ok, i wouldn't be anonymous
anymore, and you be any wiser(or less wise) .

you should really see the stupidity in your reasoning.

or are you one of those guys that in order not to be too anonymous,
likes to put all his ideas and thoughts on his t-shirt.  It's mostly
teenagers that do that. Thouigh some immature adults do too.

I don't subscribe to that camp

You should be experienced enough to know that a person that uses their
name on usenet isn't necessary making a useful contribution.  Likewise,
a person that posts as Anonymous is not necessarily a troll. On the
contrary, the troll is often soembody with a regular sounding name like
Paul Ranton  and people might think that is his name.

You are being very very silly, you know that, don't you?

Slow startup is rather serious. Programmers understand that java does a
lot more behind the scenes than other languages. And when jars are
used, there's the decompression issue on startup.  These although
understood, are not tolerated timewise using a C++ alternative exists!!
  It may well be that java is nicer for some to program in. But Using
the app is another matter.

That theoretical test at the link didn't check bulky applications.
Practice shows they kludge along.

That article provides a very weak explanation at the end, that java
isn't really so slow it's all down down to psychology!!

Had they asked people that have tried java applications for specific
reasons why they gave up on them. Then you'll start to see concrete
reasons. Not human psychology.
And these reasons only apply to java apps. Or, apply to java apps
moreso than to C++ apps.

It's possible that now java runs faster - though not as fast as C++.
But because java with swing  was so slow in the past, it has that
stigma, that bad reputation, so people see java apps and download the
other one!  I haven't tried many java apps recently because of past
experience of java programs.  Though actually, a recent java app I
tried was a complete nuisance - GUI wise. Azureus

The app did things that a regular Windows GUI would never do. Because
sometimes the java GUI components aren't quite right.  People tend not
to find the right standard components The find a component with issues
or bugs.
I recall programming a GUI in java where you coulc click and drag the
contents of one box into another. It took me a long time to look for
the right component, but I never found exactly the right one,  and I
remember having to do some ridiculous workarounds.   FWIW it was a team
project - i.e. not just me.  A Standard windows component like in VB
wouldn't have had the problem I ran into in the click and drag.

So, that's just yet another reason why java programs haven't taken
off. The GUI components have issues.

Plus i'm quite sure that java programs use more memory. the bigger the
program, the more memory it takes up.

I remember lecturers unable to do their presentation because running 2
big java apps at the same time was too much for their computer.
Everybody in the hall knew it was java. THat wouldn't have happened had
the prgorams been C++.

very good. Now try adding quotation marks

"java is slow" 16,000
"java is fast"  1,100

Obviously look at the results too, so you know you're not
misinterpreting the results.
Your experiment was intentionally misleading. Without quotation marks
you'll get articles that discuss how to make java faster, and think
they are saying java is fast.

By the way. It is possible that since Version 1.5, java is fast.

By the way. I never said thisi s conclusive. only a guide

Neither of those are the kind of statements one would expect in a
sensible article.

Look at the results and then attempt a quick evaluation of the nature
of the articles. What they are actually saying, and if they are,
whether all the aricles that appear in the first 10 were written by
quacks.
whether they are making a controversial statement - which they
obviously are.

You have to use your intelligence, when looking at results.