Your argument doesn't just apply to Caldera but also to Mandrake and SUSE. Redhats 55% of the market claim actually includes sales of the Mandrake distribution (which is now the best selling distribution in the US). Globally SUSE makes more revenue than Redhat, that's not just profit but revenue.
Distributions changing their default GUI to GNOME would be more than just an indication that RedHat knew better it would mean a large part of the GUI for these distributions would change considerably annoying customers and increasing tech support costs.
Despite this Redhat does not act as if it would like KDE to die.
Redhat has changed its KDE position from confrontational (with GNOME support) to neutral (see how quickly Bob apologized to the KDE developers after the RMS/Linux debacle, and notice the improved KDE support in RHL).
Redhat did take an anti-KDE stance with RH 6 and it has hurt them, allowing Mandrake to eat into their market share with superior KDE support.
The situation is a complicated and uncertain one.
Will Borland come out with a linux product with GTK or QT support?
Will Redhat come out with an IDE with GTK support via its Cygnus takeover?
Will a linux distributor like SUSE or Caldera start supporting kdevelop creating a professional edition with proprietary add ons?
Re:KLDs are the anwser ... perhaps
on
FreeBSD at COMDEX
·
· Score: 1
Well to be honest no.
What bothers me is that I spend 2-3hrs a day contributing code to a GPLed app that other people can't reuse.
Why shouldn't people using the Artistic or BSD license be able to use my code (which is useless without the app it contributes to)?
Just because someone might come along and use it in a closed source app? That isn't a good enough reason. I spend 7 hours a days using closed source libraries and I know what a terribly costly decision my employer made when choosing this library. I would switch to a more open equivalent if one was made available, and if there's money to be made in doing it someone will release a more open equivalent.
Well that's how I feel today anyway.
Re:KLDs are the anwser ... perhaps
on
FreeBSD at COMDEX
·
· Score: 1
Reasonable post but I have one serious gripe:
You are the third poster today I have seen state that KDE is GPLed. This is not true, the kde libraries are LGPLed so that developers may create closed source KDE apps. Some parts of the of the libs are also published under the BSD license.
Replacing CDE with KDE is like replacing one disgusting bucket of painful user-obsequiousness with another even flashier one. You aren't getting very far, very fast.
You're being too harsh. So KDE isn't for you, that's fine. I think for the kids in the lab KDE is an improvement over CDE, and serves as a better introduction than TWM even if only because of it's "user-obsequiousness".
I'm almost glad you aren't the system admin at that school. I can imagine you taking glee in the confusion of new students as they first try to use those Sparc machines running twm.
this prevents you from making tarballs for people who don't want to use CVS Not at all include a Make file in your source tarball to merge the patch and the original code.
The Make file would be run when the other developer does a./configure ; make install on your unpacked tarball. A version of the original QT source would be required as a dependency and could be distributed with your tarball of modifications.
The real problem with forking QT is in making it better than the original version, few QT developers would help you as the Trolls are nice guys who are are actively improving QT.
No one is doing it because it becuase it is a dumb idea not because anyone is preventing them.
all you would have to do is port the free version of Qt2 to win32. It has been done. About a month ago someone on the QT interest list said they had done it, and it had only taken them a couple of hours.
You are not alone. I also dislike the X10 adverts.
I personally dislike them because I'm trying to use the web for learning about technical/political issues. Seeing pornographic pictures masquerading as advertisements is very distracting, it breaks my concentration. Moreover it makes me feel uncomfortable when I'm reading an article in the office. (I not against porn but it isn't appropriate in my workplace)
I know this is the whole point of the advert, however it so effective that I now automatically move the leftmost hundred pixels of browser windows showing zdnet articles off the left of the screen, or more often do not read zdnet stuff at all.
look at the bugs in Qt 2.x and you'll see Troll has the same problems
I have reported over a dozen bugs in Qt 2.x, for all but one of those bugs Troll Tech, answered my mail, acknowledged the problem, and fixed it (in cvs) within 24 hours.
(Let's keep this quiet but I don't even have commercial support, I just develop free software and give good quality bug reports). The only bug that took longer than 24 hours to fix was the one I reported at the start of a weekend!
This is just meant to be a data point rather than a "you're wrong", type reply.
I guess I'm saying Troll Tech is an open company and releasing their stuff under a freer license (BSD, LGPL not really GPL) wouldn't necessarily make them anymore receptive to outside contribution. Also RH has released software before it was ready too, so I don't believe a buyout would remedy that situation.
't that only if Troll tech doesn't update Qt for more than a year? No. From http://www.kde.org/kdeqtfoundation.html
It's an OR not an AND. If Troll Tech is bought out or if they don't release a version of QT for 12 months then they will have to release a version of QT under the BSD license.
If it is the new BSD license, without the advertising clause then Red Hat could always relicense it under the LGPL/GPL.
Interesting twist though, the Free QT Foundation clause I mean.
So? It's still lightyears better than C/C++ code which makes it trivial to write a program that dumps core. You're problem is with bugs in the runtime. You'll encounter the same such bugs in MFC, QT, GTK, Motif, etc and other 3rd party libraries used with C++
So in reality Java has these problems (resource leaks, abnormal terminations) just the same as C/C++, that's my point. I've found developing (debugging) C++ applications easier because the implementations are more polished and I have more choices about how things are done, and I don't have to 'debug' my code (when it's perfectly valid!) to get it running on multiple VMs.
(oh, and how about C/C++ compiler optimizer related bugs) They suck, how about bugs introduce due to optimizing VMs, I've had simple (but long) arithmetic expressions work on some VMs but not others.
Yeah, it's pretty sad to do this when time-critical performance isn't an issue, isn't it? Agreed.
The number of apps that require real-time performance pail in comparison to the number of apps that run perfectly fine in Visual Basic, TCL, Perl, Java, and Python. (all of which, feature some form of automatic memory reclaimation, whether it is GC or reference counting) C++ supports reference counting too (eg shared string classes). Some parts of my code are absolutely time critical as my products can't make the user wait longer than our competitors products. (We are meant to be saving the clients time, not wasting it)
Leaks caused by accidently holding onto references *DON'T* build up indefinately. Yes they do.
The only way you can leak resources without bound would be to use a data structure that can grow without bound, and keep adding references to it. Correct, so you just contradicted yourself and explained my point to, thanks.
native-code related leaks are the only leaks that can grow without bound. Not according to you previous sentence.
Do you understand the difference between a piece of Java code holding onto a reference for too long, and a piece of native-code calling malloc() and not freeing? Yes. I do it accidentally (well operator new not malloc) all the time, and it shouldn't happen in Java. But as you've explained these aren't the only kind of resource leaks, and you still have to debug resource leaks in Java.
Which are principally in the AWT related code, not the VM core. Don't you get it? You complain about GUI/desktop related code. Don't you understand that most of the VM related bugs are in the native GUI integration code, and not in the core VM?
Thankfully I didn't have to write GUI code in Java. I found plenty of bugs in VMs writing ordinary (searching/sorting/compression) code. Saying writing GUI code is even worse is something I already know, and is a fact that scares rather than impresses me.
But it's usability is far below. Functional Languages are nice theoretical abstractions Aaah, this was my point, I was debunking the Java is a revolutionary language myth.
and there is a beautiful theory behind the interaction of functional languages and I/O using Monads I've studied and given a presentation on Monads and I think it's an ugly theory.
but functional languages are frankly a *bitch* to work with when it comes to handling environmental state like network, disk, Agreed.
or user interface events. Haggis was interesting, it looked promising.
Yes, Quicksort in 3 lines of Haskell is cool, I remember think the 3 line Quicksort implementation wasn't really a Quicksort implemention as its Big O characteristics were entirely different.
but writing Slashdot in Haskell or a Word Processor would be alot more painful, and mucher slower than either Java or C++. Today that is true.
Most of the world wants something that has immediate utility, a pragmatic advancement. Java, Eiffel, ADA, etc aren't theoretical advancements, they are simple evolutionary advancements that make development less buggier and easier to maintain. I disagree, if this were true more people would be using apps written in these languages, at least on platforms MS doesn't control.
Java certainly isn't the end-all-and-be-all of programming languages. No programming language is! Agreed.
Java was invented to run untrusted code on multiplatforms, and at that, it excels, and happens to function very nicely in a variety of contexts. It has been a spectacular failure. The security models for the popular (netscape, ie) models is different and the netscape model sucks to the point of being broken, which is why you see very few signed applets. But maybe you are talking about it being a theoretical success rather than a practical one.
And by extension, so in Eiffel, Smalltalk, ADA, oh, and not to mention, Linux itself right? yes, Yes, Yes, No.
Could it be that the whole idea of shrinkwrap style apps is a thing of the past? I don't know about you, but except for video games, multimedia, browsers, and Microsoft Office, I spent the vast majority of my time using the web. You said except for.. browsers... I spend.. my time using the web which is strange. Other important apps are Email clients (I pity you if you rely on web based email), and IDEs, also an office suite consists of many apps.
I write shrink wrapped software for a living, so it's very important to me, and I think it still has a bright future. The latency problems inherent with web based applications are currently unsolvable (we are limited by the speed of light in a vacuum) and will prevent current shrink wrapped software from being made obsolete by web based apps. (On the other hand many new types of web based apps are appearing, so really everyone wins).
Fuck, what a waste of time this has been. I don't learn anything here anymore.
Please tell me how you can corrupt the stack in Java. Meaning, expose it to a buffer overflow attack ala Java.
Hmm meant to say heap, makes no difference. All JVMs I've worked with (Sun,Netscape and MS implementations) are buggy and I haven't found it hard to write code that (unintentionally) crashes them, probably due 'exploiting' stack/heap corruption or dangling pointer references in JVM code. I'm not talking about the language here but the VM implementations.
Hmm this sounds like I'm concerned with security. I was/am talking about reliability, I mean I have enough trouble getting code to run properly on different VMs let alone making sure it's secure!
Please tell me how you can corrupt the stack in Java. Meaning, expose it to a buffer overflow attack ala Java. Hmm meant to say heap, makes no difference. All JVMs I've worked with (Sun,Netscape and MS implementations) are buggy and I haven't found it hard to write code that (unintentionally) crashes them, probably due 'exploiting' stack/heap corruption or dangling pointer references in JVM code. I'm not talking about the language here but the VM implementations.
In Java, you can only "leak resources" by holding onto them. You can get garbage collectors for C/C++ too (Great Circle), which leaves you in the same position of only leaking resources when you hold onto them.
Personally I prefer predictable runtime performance (can we say no garbage collection) and still do it manually (using a malloc debugger to find memory leaks). I admit having to do this is pretty sad, but at least in C++ I have the option.
However, these leaks don't build up and usually amount to a waste of memory that is constant. Rubbish.
Until long running C servers which can chew up all the ram in a machine. Huh?
The only "true memory leaks" in Java are related to JNI implementation bugs in classes like java.awt.Image, and java.awt.Dialog, which have been fixed. Never mind the memory leaks in the VMs.
And since you mention Haskell, care to point us to any code you've written, or are you simply blowing smoke out your ass. I haven't written any commercial Haskell code, there aren't any good Haskell compilers, a good optimizing Haskell compiler is quite possibly decades away. My point was that as a language the beauty (simplicity, readability) of Haskell is first rate, far superior to Java and C++.
(Haskell isn't perfect though I suspect many problems will always be best solved with an imperative language).
And if you want to talk about blowing C++ out of the water, let's look at Eiffel, shall we? Haven't really had a serious look at it. I'm not aware of any good compiler implementations, or any significant apps. Post and pre conditions seem like good ideas though.
Java does have its place, servlets, business logic and e-commerce programming come to mind.
However for developing shrinkwrap style apps in a competitive environment (my area of interest) it is a proven failure.
Don't dictionaries order chinese characters by radical and then by number of strokes?
I think he's refering to this kind of problem: In C++ the standard STL string class has no less than or greater than operators defined. Possibly due to the problem of ordering non ansi characters (eg Chinese etc characters).
Other string classes (eg QString in QT or CString in MFC) provide a complete set of inequality operators but I suspect they provide a nonsensical (though consistent) ordering of non ansi characters
I doubt the situation is any better in other (computer) languages.
As far as I can tell part of the problem is unicode itself it reuses character codes for different (but visually similar?) characters, so a Japanese and Chinese character that look (almost) the same maybe (and sometimes have been) given the same character code. The ordering of these characters would depend on what character set the user was using to display the characters in!
Frankly this is a serious, ahh, fuck up, in unicode that apparently has pissed off some Japanese that proposed an alternative coding system that fixed this problem, this alternative has been ignored by the international community.
We (developers) can handle different languages but handling more than one at the same time can be tricky.
I hate to be the one to tell you this, Just because I'm being obnoxious and arrogant doesn't mean I don't want to be corrected if I'm mistaken.
the new version of JDEdwards OneWorld accounting software Never heard of it, will keep a look out for it. "new version" you say, has it been released?
What products does it compete with?
multiple flavors (same source code - cross compilers), including everything running in Java. Cross compiling normally refers to using a compiler on machine A to generate machine code for another type of machine, machine B, but I think I know what you mean.
Well done to them, it can be a fair bit of work getting Java source to compile and run right under different java compilers and VMs. (Different OSs too? Do they have a Mac version? I hear VMs suck especially hard on the Mac).
Frankly, I don't believe most of the Java bashers here ever sat down and wrote any application of substance in Java.
I agree, especially as no one has ever written an application of substance in Java.
Implementations of the language are limited, restricting the language to be used for simple applets, and backoffice and servlet development. Sun still can't get the GUI framework right, by all accounts Swing has serious scalability problems apparently partly due to the limitations of the inflexible (if elegant on paper, and great for small apps) threading model of Java.
The only sizeable shrinkwrapped apps written in Java that I know of are the ones I've help do and Java IDEs, Java isn't used to implement any windows apps I use and isn't relied on in the upcoming *nix desktop environments. Implementations simply aren't competitive, (and I'm not interested in benchmarks I'm interested in substantial apps that are in wide spread use)
Also the language itself can be horribly obtuse at times even for extremely common tasks:
myVector.elementAt( ((Integer)myVector.elementAt( i )).intValue() )
instead of: mVector[mVector[i]]
And don't tell me you can't corrupt the stack and leak resources in Java, because after a few thousand lines of code it's hard not too (and in different ways on different VMs)!
Java does have it uses, I think it's a better language for teaching than Pascal and a better language for developing business logic apps than visual basic.
If you want to develop a large app C is the proven choice, C++ is the Johnny come lately that is proving itself and ADA is used by those with no choice in the matter.
Java is nothing more than a dumbed down, simplified C++, it doesn't advance the state of the art for that I suggest looking at Haskell which is pushing the boundaries of Comp Sci.
Anyway that's a few off the cuff remarks from someone who actually managed to ship a shrink wrapped Java app and who has lived to tell the tale. I really haven't given a proper critique of the language and its implementations but if you search previous posts on Slashdot you should come across some by a clued up Dylan advocate who has given some very nice summaries on the shortcomings of Java.
Actually with an infinite number of monkeys at infinite type writers, they'd write every work ever written, that ever will be written, and ever possible to be written.:) Infinite time just lets them take the time to write out even the LONGEST of long papers. (Of couse it wouldn't matter how long, cause infinite monkeys would still type every single possible variation at the same time)
This is incorrect.
There should be a FAQ about this. Anyone know of a FAQ to show people that make this statement?
I enlightened my quota of would be mathematicians for one life time and I can't be bothered explaining it again.
If the KDE team incorporates some other author's GPL'd code (being able to do this is part of the point of the GPL), and that author hasn't also licensed the work under a Qt-adapted GPL variant, that author could conceivably sue anyone who distributes KDE as well as its authors. That's a different issue from the one I was talking about. It was not my intent to comment on that issue.
If they're going to play these "we promised all derived works would be Free, but, uh, we didn't mean it" games, I wouldn't even consider contributing. That doesn't make any sense to me, I don't know what you are trying to say.
There of course remains a quiestion wheither or not it is legal for third parties to distribute or use KDE
KDE developers suing people for redistributing their code?
That would be like inviting someone over to your house for lunch and then suing them for tresspass.
The KDE authors who GPL their code, upload it to a CVS repository that is open for public access. Anyone may mirror this repository via cvsup and many people do. (The authors are inviting you to redistribute their code).
Consent for redistribution has been implicitly given.
>Is it just me, or has Slashdot gotten really >bland in the last 6 months or so?
Sounds like you're getting bored. Maybe you'll be leaving soon (maybe not I don't know). It'll be sad if you do. I mean you're a prolific writer and I've enjoyed reading your posts, I recognize you as an individual amongst all the humanity here.
Recognizing someone was never an easy thing to do here, and it's become a lot harder as the months have gone by. This place seems more like the crowd outside an express train at peak hour than any kind of community.
Once you've or I've gone I'll miss you along with all the other "reconizables", so many of whom I've forgotten. I'll miss the jerks like meept and Ivan the terrible (you remember the GUIs are evil guy?), I'll miss they cynics Zico Knows, Cassius and like you've become, as well the good guys like "This is linux country on a quiet night.." . Because you've all been part of my neighbourhood, I mean I post here but I don't even talk to my neighbours in meatspace.
I've come to the conclusion that Slashdot is self congratulating and selfish, the HOF (hall of fame) contains no listing of the most recognizable posters, personalities don't count here.
Anyway I just wanted to say good bye to all those who have left and been forgotten, I wish I could remember you all.
Slashdot Mirror
Here, here, write once, debug everywhere.
I've had to work around several bugs in the Solaris, Netscape and IE JVMs.
Your argument doesn't just apply to Caldera but also to Mandrake and SUSE. Redhats 55% of the market claim actually includes sales of the Mandrake distribution (which is now the best selling distribution in the US). Globally SUSE makes more revenue than Redhat, that's not just profit but revenue.
Distributions changing their default GUI to GNOME would be more than just an indication that RedHat knew better it would mean a large part of the GUI for these distributions would change considerably annoying customers and increasing tech support costs.
Despite this Redhat does not act as if it would like KDE to die.
Redhat has changed its KDE position from confrontational (with GNOME support) to neutral (see how quickly Bob apologized to the KDE developers after the RMS/Linux debacle, and notice the improved KDE support in RHL).
Redhat did take an anti-KDE stance with RH 6 and it has hurt them, allowing Mandrake to eat into their market share with superior KDE support.
The situation is a complicated and uncertain one.
Will Borland come out with a linux product with GTK or QT support?
Will Redhat come out with an IDE with GTK support via its Cygnus takeover?
Will a linux distributor like SUSE or Caldera start supporting kdevelop creating a professional edition with proprietary add ons?
Will this be for US residents only again?
Well to be honest no.
What bothers me is that I spend 2-3hrs a day contributing code to a GPLed app that other people can't reuse.
Why shouldn't people using the Artistic or BSD license be able to use my code (which is useless without the app it contributes to)?
Just because someone might come along and use it in a closed source app? That isn't a good enough reason. I spend 7 hours a days using closed source libraries and I know what a terribly costly decision my employer made when choosing this library. I would switch to a more open equivalent if one was made available, and if there's money to be made in doing it someone will release a more open equivalent.
Well that's how I feel today anyway.
Reasonable post but I have one serious gripe:
You are the third poster today I have seen state that KDE is GPLed. This is not true, the kde libraries are LGPLed so that developers may create closed source KDE apps. Some parts of the of the libs are also published under the BSD license.
Sorry I meant 'its "user-obsequiousness"'.
Replacing CDE with KDE is like replacing one disgusting bucket of painful user-obsequiousness with another even flashier one. You aren't getting very far, very fast.
You're being too harsh. So KDE isn't for you, that's fine. I think for the kids in the lab KDE is an improvement over CDE, and serves as a better introduction than TWM even if only because of it's "user-obsequiousness".
I'm almost glad you aren't the system admin at that school. I can imagine you taking glee in the confusion of new students as they first try to use those Sparc machines running twm.
Your babbling.
I notice the license has changed now. Not that it matters.
The minix case is entirely different, the Minix license requires royalties for redistribution of source.
this prevents you from making tarballs for people who don't want to use CVS
./configure ; make install on your unpacked tarball. A version of the original QT source would be required as a dependency and could be distributed with your tarball of modifications.
Not at all include a Make file in your source tarball to merge the patch and the original code.
The Make file would be run when the other developer does a
The real problem with forking QT is in making it better than the original version, few QT developers would help you as the Trolls are nice guys who are are actively improving QT.
No one is doing it because it becuase it is a dumb idea not because anyone is preventing them.
all you would have to do is port the free version of Qt2 to win32.
It has been done. About a month ago someone on the QT interest list said they had done it, and it had only taken them a couple of hours.
If you want I can search for the post.
You are not alone. I also dislike the X10 adverts.
I personally dislike them because I'm trying to use the web for learning about technical/political issues. Seeing pornographic pictures masquerading as advertisements is very distracting, it breaks my concentration. Moreover it makes me feel uncomfortable when I'm reading an article in the office. (I not against porn but it isn't appropriate in my workplace)
I know this is the whole point of the advert, however it so effective that I now automatically move the leftmost hundred pixels of browser windows showing zdnet articles off the left of the screen, or more often do not read zdnet stuff at all.
look at the bugs in Qt 2.x and you'll see Troll has the same problems
I have reported over a dozen bugs in Qt 2.x, for all but one of those bugs Troll Tech, answered my mail, acknowledged the problem, and fixed it (in cvs) within 24 hours.
(Let's keep this quiet but I don't even have commercial support, I just develop free software and give good quality bug reports). The only bug that took longer than 24 hours to fix was the one I reported at the start of a weekend!
This is just meant to be a data point rather than a "you're wrong", type reply.
I guess I'm saying Troll Tech is an open company and releasing their stuff under a freer license (BSD, LGPL not really GPL) wouldn't necessarily make them anymore receptive to outside contribution. Also RH has released software before it was ready too, so I don't believe a buyout would remedy that situation.
't that only if Troll tech doesn't update Qt for more than a year?
No. From http://www.kde.org/kdeqtfoundation.html
It's an OR not an AND. If Troll Tech is bought out or if they don't release a version of QT for 12 months then they will have to release a version of QT under the BSD license.
If it is the new BSD license, without the advertising clause then Red Hat could always relicense it under the LGPL/GPL.
Interesting twist though, the Free QT Foundation clause I mean.
So? It's still lightyears better than C/C++ code which makes it trivial to write a program that dumps core. You're problem is with bugs in the runtime. You'll encounter the same such bugs in MFC, QT, GTK, Motif, etc and other 3rd party libraries used with C++
.. browsers... I spend .. my time using the web which is strange. Other important apps are Email clients (I pity you if you rely on web based email), and IDEs, also an office suite consists of many apps.
So in reality Java has these problems (resource leaks, abnormal terminations) just the same as C/C++, that's my point. I've found developing (debugging) C++ applications easier because the implementations are more polished and I have more choices about how things are done, and I don't have to 'debug' my code (when it's perfectly valid!) to get it running on multiple VMs.
(oh, and how about C/C++ compiler optimizer related bugs)
They suck, how about bugs introduce due to optimizing VMs, I've had simple (but long) arithmetic expressions work on some VMs but not others.
Yeah, it's pretty sad to do this when time-critical performance isn't an issue, isn't it?
Agreed.
The number of apps that require real-time performance pail in comparison to the number of apps that run perfectly fine in Visual Basic, TCL, Perl, Java, and Python. (all of which, feature some form of automatic memory reclaimation, whether it is GC or reference counting)
C++ supports reference counting too (eg shared string classes). Some parts of my code are absolutely time critical as my products can't make the user wait longer than our competitors products. (We are meant to be saving the clients time, not wasting it)
Leaks caused by accidently holding onto references *DON'T* build up indefinately.
Yes they do.
The only way you can leak resources without bound would be to use a data structure that can grow without bound, and keep adding references to it.
Correct, so you just contradicted yourself and explained my point to, thanks.
native-code related leaks are the only leaks that can grow without bound.
Not according to you previous sentence.
Do you understand the difference between a piece of Java code holding onto a reference for too long, and a piece of native-code calling malloc() and not freeing?
Yes. I do it accidentally (well operator new not malloc) all the time, and it shouldn't happen in Java. But as you've explained these aren't the only kind of resource leaks, and you still have to debug resource leaks in Java.
Which are principally in the AWT related code, not the VM core. Don't you get it? You complain about GUI/desktop related code. Don't you understand that most of the VM related bugs are in the native GUI integration code, and not in the core VM?
Thankfully I didn't have to write GUI code in Java. I found plenty of bugs in VMs writing ordinary (searching/sorting/compression) code.
Saying writing GUI code is even worse is something I already know, and is a fact that scares rather than impresses me.
But it's usability is far below. Functional Languages are nice theoretical abstractions
Aaah, this was my point, I was debunking the Java is a revolutionary language myth.
and there is a beautiful theory behind the interaction of functional languages and I/O using Monads
I've studied and given a presentation on Monads and I think it's an ugly theory.
but functional languages are frankly a *bitch* to work with when it comes to handling environmental state like network, disk,
Agreed.
or user interface events.
Haggis was interesting, it looked promising.
Yes, Quicksort in 3 lines of Haskell is cool,
I remember think the 3 line Quicksort implementation wasn't really a Quicksort implemention as its Big O characteristics were entirely different.
but writing Slashdot in Haskell or a Word Processor would be alot more painful, and mucher slower than either Java or C++.
Today that is true.
Most of the world wants something that has immediate utility, a pragmatic advancement. Java, Eiffel, ADA, etc aren't theoretical advancements, they are simple evolutionary advancements that make development less buggier and easier to maintain.
I disagree, if this were true more people would be using apps written in these languages, at least on platforms MS doesn't control.
Java certainly isn't the end-all-and-be-all of programming languages. No programming language is!
Agreed.
Java was invented to run untrusted code on multiplatforms, and at that, it excels, and happens to function very nicely in a variety of contexts.
It has been a spectacular failure. The security models for the popular (netscape, ie) models is different and the netscape model sucks to the point of being broken, which is why you see very few signed applets. But maybe you are talking about it being a theoretical success rather than a practical one.
And by extension, so in Eiffel, Smalltalk, ADA, oh, and not to mention, Linux itself right?
yes, Yes, Yes, No.
Could it be that the whole idea of shrinkwrap style apps is a thing of the past? I don't know about you, but except for video games, multimedia, browsers, and Microsoft Office, I spent the vast majority of my time using the web.
You said except for
I write shrink wrapped software for a living, so it's very important to me, and I think it still has a bright future. The latency problems inherent with web based applications are currently unsolvable (we are limited by the speed of light in a vacuum) and will prevent current shrink wrapped software from being made obsolete by web based apps. (On the other hand many new types of web based apps are appearing, so really everyone wins).
Fuck, what a waste of time this has been. I don't learn anything here anymore.
Please tell me how you can corrupt the stack in Java. Meaning, expose it to a buffer overflow attack ala Java.
Hmm meant to say heap, makes no difference. All JVMs I've worked with (Sun,Netscape and MS implementations) are buggy and I haven't found it hard to write code that (unintentionally) crashes them, probably due 'exploiting' stack/heap corruption or dangling pointer references in JVM code. I'm not talking about the language here but the VM implementations.
Hmm this sounds like I'm concerned with security. I was/am talking about reliability, I mean I have enough trouble getting code to run properly on different VMs let alone making sure it's secure!
"can't corrupt stack and leak resources"
Please tell me how you can corrupt the stack in Java. Meaning, expose it to a buffer overflow attack ala Java.
Hmm meant to say heap, makes no difference. All JVMs I've worked with (Sun,Netscape and MS implementations) are buggy and I haven't found it hard to write code that (unintentionally) crashes them, probably due 'exploiting' stack/heap corruption or dangling pointer references in JVM code. I'm not talking about the language here but the VM implementations.
In Java, you can only "leak resources" by holding onto them.
You can get garbage collectors for C/C++ too (Great Circle), which leaves you in the same position of only leaking resources when you hold onto them.
Personally I prefer predictable runtime performance (can we say no garbage collection) and still do it manually (using a malloc debugger to find memory leaks). I admit having to do this is pretty sad, but at least in C++ I have the option.
However, these leaks don't build up and usually amount to a waste of memory that is constant.
Rubbish.
Until long running C servers which can chew up all the ram in a machine.
Huh?
The only "true memory leaks" in Java are related to JNI implementation bugs in classes like java.awt.Image, and java.awt.Dialog, which have been fixed.
Never mind the memory leaks in the VMs.
And since you mention Haskell, care to point us to any code you've written, or are you simply blowing smoke out your ass.
I haven't written any commercial Haskell code, there aren't any good Haskell compilers, a good optimizing Haskell compiler is quite possibly decades away. My point was that as a language the beauty (simplicity, readability) of Haskell is first rate, far superior to Java and C++.
(Haskell isn't perfect though I suspect many problems will always be best solved with an imperative language).
And if you want to talk about blowing C++ out of the water, let's look at Eiffel, shall we?
Haven't really had a serious look at it. I'm not aware of any good compiler implementations, or any significant apps. Post and pre conditions seem like good ideas though.
Java does have its place, servlets, business logic and e-commerce programming come to mind.
However for developing shrinkwrap style apps in a competitive environment (my area of interest) it is a proven failure.
Don't dictionaries order chinese characters by radical and then by number of strokes?
I think he's refering to this kind of problem:
In C++ the standard STL string class has no less than or greater than operators defined. Possibly due to the problem of ordering non ansi characters (eg Chinese etc characters).
Other string classes (eg QString in QT or CString in MFC) provide a complete set of inequality operators but I suspect they provide a nonsensical (though consistent) ordering of non ansi characters
I doubt the situation is any better in other (computer) languages.
As far as I can tell part of the problem is unicode itself it reuses character codes for different (but visually similar?) characters, so a Japanese and Chinese character that look (almost) the same maybe (and sometimes have been) given the same character code. The ordering of these characters would depend on what character set the user was using to display the characters in!
Frankly this is a serious, ahh, fuck up, in unicode that apparently has pissed off some Japanese that proposed an alternative coding system that fixed this problem, this alternative has been ignored by the international community.
We (developers) can handle different languages but handling more than one at the same time can be tricky.
I hate to be the one to tell you this,
Just because I'm being obnoxious and arrogant doesn't mean I don't want to be corrected if I'm mistaken.
the new version of JDEdwards OneWorld accounting software
Never heard of it, will keep a look out for it. "new version" you say, has it been released?
What products does it compete with?
multiple flavors (same source code - cross compilers), including everything running in Java.
Cross compiling normally refers to using a compiler on machine A to generate machine code for another type of machine, machine B, but I think I know what you mean.
Well done to them, it can be a fair bit of work getting Java source to compile and run right under different java compilers and VMs. (Different OSs too? Do they have a Mac version? I hear VMs suck especially hard on the Mac).
Frankly, I don't believe most of the Java bashers here ever sat down and wrote any application of substance in Java.
I agree, especially as no one has ever written an application of substance in Java.
Implementations of the language are limited, restricting the language to be used for simple applets, and backoffice and servlet development. Sun still can't get the GUI framework right, by all accounts Swing has serious scalability problems apparently partly due to the limitations of the inflexible (if elegant on paper, and great for small apps) threading model of Java.
The only sizeable shrinkwrapped apps written in Java that I know of are the ones I've help do and Java IDEs, Java isn't used to implement any windows apps I use and isn't relied on in the upcoming *nix desktop environments. Implementations simply aren't competitive, (and I'm not interested in benchmarks I'm interested in substantial apps that are in wide spread use)
Also the language itself can be horribly obtuse at times even for extremely common tasks:
myVector.elementAt( ((Integer)myVector.elementAt( i )).intValue() )
instead of:
mVector[mVector[i]]
And don't tell me you can't corrupt the stack and leak resources in Java, because after a few thousand lines of code it's hard not too (and in different ways on different VMs)!
Java does have it uses, I think it's a better language for teaching than Pascal and a better language for developing business logic apps than visual basic.
If you want to develop a large app C is the proven choice, C++ is the Johnny come lately that is proving itself and ADA is used by those with no choice in the matter.
Java is nothing more than a dumbed down, simplified C++, it doesn't advance the state of the art for that I suggest looking at Haskell which is pushing the boundaries of Comp Sci.
Anyway that's a few off the cuff remarks from someone who actually managed to ship a shrink wrapped Java app and who has lived to tell the tale. I really haven't given a proper critique of the language and its implementations but if you search previous posts on Slashdot you should come across some by a clued up Dylan advocate who has given some very nice summaries on the shortcomings of Java.
Actually with an infinite number of monkeys at infinite type writers, they'd write every work ever written, that ever will be written, and ever possible to be written. :) Infinite time just lets them take the time to write out even the LONGEST of long papers. (Of couse it wouldn't matter how long, cause infinite monkeys would still type every single possible variation at the same time)
This is incorrect.
There should be a FAQ about this. Anyone know of a FAQ to show people that make this statement?
I enlightened my quota of would be mathematicians for one life time and I can't be bothered explaining it again.
If the KDE team incorporates some other author's GPL'd code (being able to do this is part of the point of the GPL), and that author hasn't also licensed the work under a Qt-adapted GPL variant, that author could conceivably sue anyone who distributes KDE as well as its authors.
That's a different issue from the one I was talking about. It was not my intent to comment on that issue.
If they're going to play these "we promised all derived works would be Free, but, uh, we didn't mean it" games, I wouldn't even consider contributing.
That doesn't make any sense to me, I don't know what you are trying to say.
There of course remains a quiestion wheither or not it is legal for third parties to distribute or use KDE
KDE developers suing people for redistributing their code?
That would be like inviting someone over to your house for lunch and then suing them for tresspass.
The KDE authors who GPL their code, upload it to a CVS repository that is open for public access. Anyone may mirror this repository via cvsup and many people do. (The authors are inviting you to redistribute their code).
Consent for redistribution has been implicitly given.
>Is it just me, or has Slashdot gotten really
>bland in the last 6 months or so?
Sounds like you're getting bored. Maybe you'll be leaving soon (maybe not I don't know). It'll be sad if you do. I mean you're a prolific writer and I've enjoyed reading your posts, I recognize you as an individual amongst all the humanity here.
Recognizing someone was never an easy thing to do here, and it's become a lot harder as the months have gone by. This place seems more like the crowd outside an express train at peak hour than any kind of community.
Once you've or I've gone I'll miss you along with all the other "reconizables", so many of whom I've forgotten. I'll miss the jerks like meept and Ivan the terrible (you remember the GUIs are evil guy?), I'll miss they cynics Zico Knows, Cassius and like you've become, as well the good guys like "This is linux country on a quiet night.." . Because you've all been part of my neighbourhood, I mean I post here but I don't even talk to my neighbours in meatspace.
I've come to the conclusion that Slashdot is self congratulating and selfish, the HOF (hall of fame) contains no listing of the most recognizable posters, personalities don't count here.
Anyway I just wanted to say good bye to all those who have left and been forgotten, I wish I could remember you all.