did they use there alledged position of power or is it simply the better product (winxp server)? How could you (or a judge) tell the difference?
Ah, but you don't need to tell the difference! The quality of the Windows server system is irrelevant - what matters is that (allegedly) competing server systems can't even attempt to provide equivalent quality services to Windows clients because Microsoft is not revealing the protocols that the clients require. This gives Windows servers unfair leverage in the market. This is the key point - by having dominance in one market (desktops), Microsoft apparently has an automatic advantage in another (servers). By opening up the protocols, other server systems can compete equally to provide services to Windows clients.
Maybe I was assuming to much, but I thought it was plain as day to everyone that Windows isn't the only operating system available.
Doesn't matter. What matters is the market share and how that share is used. This is why there is nothing intrinsically bad about a de-facto monopoly - it can in principle be achieved by quality. (Personally though, I think that much of Windows success is due to bundling with hardware).
Microsoft truly have a de-facto monopoly, and other systems are somewhat excluded, if only because hardware makers work with Microsoft to help their software run better.
But anyway, what matters is how you 'use' this powerful position even if it is achieved through merit.
Despite this fact, stating that MS IS a monopoly without backing it up will never net you a troll moderation here.
RE: Troll moderation. I really wonder how I could get to excellent karma with my opinions on this website. Finally, the day of reckoning is here. At least nobody pretents that this is a site for debate anymore, where presentation of an opionion gets moderated instead of the opinion itself.
You get good karma here by presenting reasoned arguments. I have opinions that are probably radically different from the majority of Slashdotters, but I try and support those opinions with examples and debate. Simply doing what you do and stating 'Microsoft aren't a monopoly' isn't enough. You should try and provide historical examples or legal backing for your view.
Calling a company that hasn't got a 100% marketshare a monopoly is plain wrong. It's also evil because the users of this 85%-truth speculate that some of the negative feelings associated with the word "monopoly" rub onto the company so falsly accused.
There is nothing evil about being a monopoly. There need be no negative feelings about it. However, a monopoly has legal responsibilities. What is wrong is a company that has a monopoly (or "de facto" monopoly, which is when a company has sufficient strength to dominate the market even without 100% of that market) using tht dominance to control purchases in another market. This includes trying to use dominance in personal computer operating systems to encourage purchases of server systems, or to try and dominate multimedia systems. This is unquestionably what Microsoft are trying to do.
However, they were found to be guilty of abusing their marketpower eventhough they only did what's common place in the auto industry, RE: making it impossible to integrate a third party [browser|air conditioner] into their [operating system|car].
As far as the earth goes, the most spectacular environment catastrophe posited is Snowball Earth. Basically, the entire Earth was frozen over with a sheet of ice two miles thick, everything died and there was no oxygen in the atmosphere, for a period of a few hundred million years. It was a rough time, but, ironically, the Earth was saved by an accumulation of 350 times our present level of CO2.
This is almost certainly mistaken, simply because there is no evidence of any extinction - everything didn't die! And unless there was at least some open water to allow gaseous exchange, everything beyond bacteria would probably have died. So, the idea of a totally frozen Earth is not feasible.
We know that radiation can cause cancer. It has been proven time and time again.
Let's discuss quantum theory. This means that radiation with particles of less than a certain energy (frequency) can't cause certain things to happen, such as breaking the bonds in a DNA molecule (which is necessary to cause cancer). There may be some indirect effect - if the radiation causes sufficient heating, but that would have to be significant heating.
Anything in sufficient quantities is bad.
That is true, but does not mean that any radiation of any sort in sufficient quantites causes cancer.
First off, evolution actually fits better with Christian theology than young-earth creation, because it depicts "creation" as an ongoing process rather than a one-off event. It places God in history rather than outside it, which is one of the themes that you'll find running through the Bible.
But this is the problem - it doesn't. The whole point of current neo-Darwinian theory is that evolution is the natural selection of random mutations: there is no sign, or need, for any intervention in the process - no sign or need for an intelligent designer!
Java does not install under one directory. The installer edits(!) MIME-related files in/etc, creates an/etc/.java directory and installs icons and (useless/obsolete) MIME and application registry files to/usr/share.
Big deal! I wonder how you can possibly carry on using your machine with all that on it!
As for the interface issues, the Swing one looks awful.
That statement doesn't make sense. There is no single Swing look - there are literally hundreds. It depends entirely on the platform you are on and the look and feel you choose. If you use Swing on MacOS/X, you are using Apple-designed Swing.
Where can i see the source-code of the java-compiler? how do i know it does not contain any malevolent features? how can i improve it if it has a problem? how can i add a feature or change it if a friend asks for my help? because i want to know what my computer does, i am not free to download the jre. to do so would be to give up a part of the control over my computer. howie.
Sun may not be dodgy. The installer is. Call it snobbery if you will, but as a Debian user I expect a level of quality from software packages that a third party binary installer can not provide.
I used to think that about Debian, but I have had some horrors with a recent update.
PS - tell me how one can run the installer without using the command line!:)
You can open the RPM package for Java using graphical RPM tools, such as provided with KDE.
not sure what world you are living in, you only have to look at how redhat has stagnated over the past year and Sun is still going out backwards. Red Hat is faring better than Sun with its verging on dead proprietry Java.
Not sure what world you are living in, but you only have to look at how successful Red Hat is on the server side compared with Windows, and only someone living in some weird parallel dimension could think that Java is 'dead'. Its use is still growing after a decade - it is now in advance of C++ on sourceforge and check out things like the TIOBE index.
I often find such strange uses of the word 'dead' on Slashdot - they don't conform to uses in our reality!
Hi, I'd just like to say that I also find Java installation a pain. It's not so much that I can't apt-get it, more that it's not installed by default. My OS comes with python, perl, ruby, gcc-everything, and all kinds of other specialised languages like tcl, octave... more than I can count. All this stuff is installed by default, updated by default, and I don't have to give a shit about it.
My sympathy for you having to click on a link to install it has no bounds....
Why does Java think it's so freaking special?
Because it allows you freaking easily write freaking reasonably fast programs with freaking portable GUIs and which are freaking cross platform even with complex things like freaking threading.
Java has a better optimizer right now but I have not found that any issue in practice since the language is almost never a slow point in a program, the problems are almost always algorithmic.
Actually there are other systems that have ported across architectures very easily also. IIRC clisp programs work just fine on 32bit, 64bit, different os etc.
yes, I know - I keep hearing these same arguments, but the point is that languages like python really aren't suitable for some of the true general purpose work that Java is used for. In practice, with a large enough app, performance always end up mattering in one place or another.
Also, clisp simply is not liked that much as a general purpose language.
I know all this stuff seems revolutionary and it is nice that the software industry is paying attention to some of this stuff but it is not new.
It is far from new, but it is revolutionary that it is actually being put into practice.
Heck most c apps for unixes have only needed a recompile to run on just about any platform out there. It is pretty easy to write architecture independent code, it is just that most people that write software should be flipping burgers instead.
No, this is plain wrong. There can be major issues (I have dealt with them, as I have used C for more than 20 years) - threading libraries, different GUI systems, different word lengths, different socket handling for networks... etc, etc...
This also means that a serious application developer has to compile and test the app on each platform. With the JRE you get a certified and tested level of compatibility. The 'write once run anywhere' principle of Java is not a myth - it is used daily by hundreds of thousands of developers.
Why even bother to do such things with C when Java is so much simpler? Portability guaranteed; run-time optimised; memory managed. Sure, C is neat for some quick jobs and other tasks, but for most of us, there is simply no need.
Having said that how many times have top level brass from sun contradicted each other? Sun is clearly an organization where the left hand doesn't know what the right hand is doing.
Clearly? All large organisations have such disagreements. Sun is honest and has them openly.
Java web start requires you to click.
Are we really reduced such a dislike of Java that there is an objection to the fact that the update is only done on demand when the user starts an appliction by clicking a mouse?
To me that is automatic - the JRE gets automatically updated when a program is started.
There is also the Java Control Panel for Windows - that updates Java automatically and you don't even have to click anything.
One set requires an agreement with Sun on redistribution of the runtime and the JDK. IANAL but various distributions have read the license and have chosen not to be bound by it's terms, apparently they found it onorous.
And others haven't.
Maybe you think sun would never sue you but we have seen many examples of large corporations and their legal departments suing people willy nilly and winning. Do you really want to take that risk?
Yes, I do. Because I don't go along with some of the most extreme open source supporters that all software protection is bad and commercial companies are inherently evil.
In a nutshell. Sun does not want you to automatically download and install their software. Not if you are a distributor (a distro) and not if you are an end user.
It is strange, then, that they provide a tool that does just that - it is called Java Web Start. It automatically downloads and installs newer versions of the JRE as required by Java software.
My job is a programmer not a sysadmin. I just have to do sysadmin tasks also. All of our apps are written in python and are for zope. None of the systems need java in any way. Since my main job is writing custom software and I do that with zope why should I make my life harder then it has to be?
In that case, there is obviously no problem. But I suspect that this is rare. Java is increasingly the language of choice for substantial server side development. I think it is going to get hard to get by without it.
It is not like it is a vastly better language then anything else out there.
No; but it is a vastly better development and deployment system than almost anything else. I know of no other mainstream language that allows for production of high-performance apps than can be deployed unchanged on such a wide range of platforms, and even on 32-bit and 64-bit, from the same binary! That has been a revolution for the software industry, no matter how much open source supporters try and dismiss the language.
but if I have to choose between c# and java my choice will be c# because mono is trivial to install and keep updated.
Fine, but you are cutting yourself and your users off from what is probably the most used development language for commercial apps today over what is, to be honest, nothing more than a matter of minor convenience. It is really worth it? How difficult it is to check for Java updates, say, once a month, and script the automatic distribution and installation of these things? Surely that is the kind of thing Linux fans are more than capable of doing? Or are we becoming reduced to the level of Windows users, where if we don't have tools to do things, it is all too much trouble??
Sorry, but this really does seem to me to be trying to set up deliberate hurdles in the way of using Java that would not be put up for other languages. There have been major issues in past years with updates and version tracking for other popular languages - PERL, versions of gcc etc. Did open source advocates shrug off these languages, saying that they were too much bother to support?
When did Linux sysadmins, who have in the past struggled valiantly through things like kernel re-compilations, suddenly become so feeble when faced with a few command-line actions, like those required to install Java?
Not for GNU/Linux users. Java will only work if you use i386 and know how to use the command line.
No. I can do all Java installation using KDE without a command line.
some dodgy third-party binary installer.
Sorry, but calling a Java installer from Sun 'some dodgy third-party binary installer' is simply open-source snobbery, and can't be a comment that I can take seriously! Come on now - Sun is not dodgy!
On GNU/Linux users must open the scary terminal and type a complicated command to get it to install.
No, you don't.
If Sun's license was a little more liberal then proper packages would be made that could go into the non-free section. Then the end user wouldn't need to notice that their program was written in/for/with Java; a JRE would be pulled in automatically when they installed it, just like any other dependancy.
Yes, I agree. However, why can't it work the other way? Why can't the Linux distributions restrictions be a little less restrictive?
Really? You mean I can just apt-get install JDK? You mean if I apt-get install your program the JDK or the JRE will be downloaded and installed? If I apt-get install a python program pyton gets installed, if I apt-get install a C# program mono gets installed automatically. If I don't have the same convenience with java why should I use your program?
For millions of users, the simple click of a link on a web page leads to the download of a JRE. apt-get is a command-line gateway to a selection of software highly constrained (for better or worse) by a set of political ideals.
Personally, I use Debian, and I like it and I like apt, but to consider it the only convenient way to install software is rather extreme.
I mean, if you can use apt-get, are you really trying to tell me you would have trouble downloading a Sun JRE for Linux and running the binary that results?
You can blame me all you want but why should I use your program if I have to go through extra hoops before I can even install it and use it?
Are you seriously trying to say that clicking a link to download a binary, then running the binary is going through extra hoops?
Even novice users on Windows or MacOS can manage that!
Does 'apt-get' limit absolutely everything you do?
Java is just another programming language.
No, it isn't. It is a free language and development system that has helped successfully hold back Microsoft serverside, and has certainly helped Linux become a serious platform for server-side corporate use as an application hosting operating system. No other language or platform has come close to having such a significant influence.
If Sun doesn't want to play nice with open source other people can't make them.
Sure, but that hasn't stopped Java becoming one of the most widely used and successful development languages. Sun should change some of the silly restrictions on their licenses, but at the same time some of the more extreme open sourcers need to drop some of their arrogant attitude towards anything that doesn't quite meet their ideals.
And, anyone who claims that 'Sun doesn't want to play nice with open source' is showing ignorance of Sun's huge contribution to open source. Used GNOME? NFS? Open Office? NetBeans?
"He needs Sun and Java and Torvalds and ESR and Red Hat and everyone else"
No. Emphatically no. It's the other way around. The corps desperately need him. Most of them tried it the proprietary way for years and lost to Microsoft.
Interesting. I would love to hear how Red Hat and Sun (with Java) have lost to Microsoft. On the contrary, Microsoft sees commercial Linux and Java as major threats, and it has singularly failed to compete with them in major areas.
If you include water vapor as a greenhouse gas, humanity contributes 0.28% of the total greenhouse gases in the atmosphere, of which 0.117% is CO2. It's not likely that such a small amount is significant.
That is simply incorrect physics. Different types of gas have different absorption effects. For example, even small amounts of methane can have a dramatic influence.
What matters is not the amount - it is the changes.
How do you know we've had a significant impact on climate change? The most recent climate reconstructions (Moberg and others) have dumped the "hockey stick" graph used in the IPCC for one that includes the Little Ice Age and Medieval Warm Period.
Some people have dumped it; others haven't. The consensus seems to be (for now) that it is real. Also, that many of those fluctuations were somewhat localised. We are now experiencing widespread and definitely global effects.
When you look at those new graphs, it's clear that there've been warming and cooling cycles before Mankind could have caused them.
Of course there have. It is not like the Earth was a simple stable system before mankind. However, to deny that dumping vast amounts of CO2 into the atmophere isn't a highly dangerous thing to do and that it could potentially (and most likely has) cause climatic shifts is extremely poor reasoning.
Recently, researchers have found "that living plants emit 10 to 100 times more methane than dead plants. Scientists had previously thought that plants could only emit methane in the absence of oxygen." None of the climate models currently account for the impact of living plants on GW.
And this is irrelevant. Methane levels are low, and if plants are doing this now, they have always been doing it. As I said, that matters is changes in things.
"Sunshine levels had been decreasing by 2 per cent a decade between 1960 and 1980 - a total decline of about 6 per cent. Now they are going up again. Perhaps this is why our Swiss glaciers are melting," Professor Wild said.
Yes, and this is because of pollution controls and a decrease in suspended particulate matter. This is now why our glaciers are melting - it is why they are melting faster, and why they will melt even faster in future. This adds to the effect of CO2 - it does not explain global warming - it makes it more worrying, as the effect of CO2 on temperature could well have been held back by past dimming of sunlight.
No, the main source of warming is the Sun. CO2 might trap the heat in the Earths atmosphere, but the bulk of that heat comes from the sun.
'Warming' implies increase in temperature. The Sun does not result in an increase in temperature: its output is relatively constant. CO2 increase will result in an increase in trapping, hence 'warming'.
Is that mankinds fault? Or something we can do something about? If the sun gets hotter, so do the planets.
I want him to realise that some who he seems to think are the enemy really aren't.
Sun has very onerous provisions on their java licensing which prevents inclusion of the JDK in a lot of Linux distributions. Why is this good for java? Why is it good for you (the java programmer) that I have to jump trough fifty hoops to install a JDK or a JRE before I can even run your program?
Er - what hoops? One click and it is downloaded. You don't have to jump through the hoops. I am free to redistribute the JRE with any Java program I produce.
Why is it good for you that the java implementation on my linux box is two years out of date and is slow?
That is the fault of the particular Linux distribution, and their political attitudes.
It is also your fault for now downloading the java implementation - which you are free to do at any time,
How would RMS compromise to make all that better for you?
By accepting that not everyone who is not entirely in support of his ideas is against him.
How could Sun compromise to make that better?
Some of their licensing is silly, I agree. That should be changed. But only a wild fanatic would deny the major benefits of Java to platforms such as Linux.
Slashdot Mirror
did they use there alledged position of power or is it simply the better product (winxp server)? How could you (or a judge) tell the difference?
Ah, but you don't need to tell the difference! The quality of the Windows server system is irrelevant - what matters is that (allegedly) competing server systems can't even attempt to provide equivalent quality services to Windows clients because Microsoft is not revealing the protocols that the clients require. This gives Windows servers unfair leverage in the market. This is the key point - by having dominance in one market (desktops), Microsoft apparently has an automatic advantage in another (servers). By opening up the protocols, other server systems can compete equally to provide services to Windows clients.
Do you see?
Maybe I was assuming to much, but I thought it was plain as day to everyone that Windows isn't the only operating system available.
Doesn't matter. What matters is the market share and how that share is used. This is why there is nothing intrinsically bad about a de-facto monopoly - it can in principle be achieved by quality. (Personally though, I think that much of Windows success is due to bundling with hardware).
Microsoft truly have a de-facto monopoly, and other systems are somewhat excluded, if only because hardware makers work with Microsoft to help their software run better.
But anyway, what matters is how you 'use' this powerful position even if it is achieved through merit.
Despite this fact, stating that MS IS a monopoly without backing it up will never net you a troll moderation here.
I suspect this is true!
RE: Troll moderation. I really wonder how I could get to excellent karma with my opinions on this website. Finally, the day of reckoning is here. At least nobody pretents that this is a site for debate anymore, where presentation of an opionion gets moderated instead of the opinion itself.
You get good karma here by presenting reasoned arguments. I have opinions that are probably radically different from the majority of Slashdotters, but I try and support those opinions with examples and debate. Simply doing what you do and stating 'Microsoft aren't a monopoly' isn't enough. You should try and provide historical examples or legal backing for your view.
Calling a company that hasn't got a 100% marketshare a monopoly is plain wrong. It's also evil because the users of this 85%-truth speculate that some of the negative feelings associated with the word "monopoly" rub onto the company so falsly accused.
There is nothing evil about being a monopoly. There need be no negative feelings about it. However, a monopoly has legal responsibilities. What is wrong is a company that has a monopoly (or "de facto" monopoly, which is when a company has sufficient strength to dominate the market even without 100% of that market) using tht dominance to control purchases in another market. This includes trying to use dominance in personal computer operating systems to encourage purchases of server systems, or to try and dominate multimedia systems. This is unquestionably what Microsoft are trying to do.
However, they were found to be guilty of abusing their marketpower eventhough they only did what's common place in the auto industry, RE: making it impossible to integrate a third party [browser|air conditioner] into their [operating system|car].
Yes, because they are a monopoly.
As far as the earth goes, the most spectacular environment catastrophe posited is Snowball Earth. Basically, the entire Earth was frozen over with a sheet of ice two miles thick, everything died and there was no oxygen in the atmosphere, for a period of a few hundred million years. It was a rough time, but, ironically, the Earth was saved by an accumulation of 350 times our present level of CO2.
This is almost certainly mistaken, simply because there is no evidence of any extinction - everything didn't die! And unless there was at least some open water to allow gaseous exchange, everything beyond bacteria would probably have died. So, the idea of a totally frozen Earth is not feasible.
We know that radiation can cause cancer. It has been proven time and time again.
Let's discuss quantum theory. This means that radiation with particles of less than a certain energy (frequency) can't cause certain things to happen, such as breaking the bonds in a DNA molecule (which is necessary to cause cancer). There may be some indirect effect - if the radiation causes sufficient heating, but that would have to be significant heating.
Anything in sufficient quantities is bad.
That is true, but does not mean that any radiation of any sort in sufficient quantites causes cancer.
First off, evolution actually fits better with Christian theology than young-earth creation, because it depicts "creation" as an ongoing process rather than a one-off event. It places God in history rather than outside it, which is one of the themes that you'll find running through the Bible.
But this is the problem - it doesn't. The whole point of current neo-Darwinian theory is that evolution is the natural selection of random mutations: there is no sign, or need, for any intervention in the process - no sign or need for an intelligent designer!
Java does not install under one directory. The installer edits(!) MIME-related files in /etc, creates an /etc/.java directory and installs icons and (useless/obsolete) MIME and application registry files to /usr/share.
Big deal! I wonder how you can possibly carry on using your machine with all that on it!
As for the interface issues, the Swing one looks awful.
That statement doesn't make sense. There is no single Swing look - there are literally hundreds. It depends entirely on the platform you are on and the look and feel you choose. If you use Swing on MacOS/X, you are using Apple-designed Swing.
Where can i see the source-code of the java-compiler? how do i know it does not contain any malevolent features? how can i improve it if it has a problem? how can i add a feature or change it if a friend asks for my help?
because i want to know what my computer does, i am not free to download the jre. to do so would be to give up a part of the control over my computer.
howie.
You can download and change the source for the next version of Java from here:
https://mustang.dev.java.net/
Sun's 'installer' for Java is lousy, it leaves little turds all over your filesystem.
;)
No - it installs under one directory.
[0] except by noticing the crappy user interface
Which one? The openGL-accelerated Swing one? The native SWT one? Or the GTK+ one?
Sun may not be dodgy. The installer is. Call it snobbery if you will, but as a Debian user I expect a level of quality from software packages that a third party binary installer can not provide.
:)
I used to think that about Debian, but I have had some horrors with a recent update.
PS - tell me how one can run the installer without using the command line!
You can open the RPM package for Java using graphical RPM tools, such as provided with KDE.
not sure what world you are living in, you only have to look at how redhat has stagnated over the past year and Sun is still going out backwards. Red Hat is faring better than Sun with its verging on dead proprietry Java.
Not sure what world you are living in, but you only have to look at how successful Red Hat is on the server side compared with Windows, and only someone living in some weird parallel dimension could think that Java is 'dead'. Its use is still growing after a decade - it is now in advance of C++ on sourceforge and check out things like the TIOBE index.
I often find such strange uses of the word 'dead' on Slashdot - they don't conform to uses in our reality!
Hi, I'd just like to say that I also find Java installation a pain. It's not so much that I can't apt-get it, more that it's not installed by default.
My OS comes with python, perl, ruby, gcc-everything, and all kinds of other specialised languages like tcl, octave... more than I can count. All this stuff is installed by default, updated by default, and I don't have to give a shit about it.
My sympathy for you having to click on a link to install it has no bounds....
Why does Java think it's so freaking special?
Because it allows you freaking easily write freaking reasonably fast programs with freaking portable GUIs and which are freaking cross platform even with complex things like freaking threading.
That is freaking nice.
Java has a better optimizer right now but I have not found that any issue in practice since the language is almost never a slow point in a program, the problems are almost always algorithmic.
Actually there are other systems that have ported across architectures very easily also. IIRC clisp programs work just fine on 32bit, 64bit, different os etc.
yes, I know - I keep hearing these same arguments, but the point is that languages like python really aren't suitable for some of the true general purpose work that Java is used for. In practice, with a large enough app, performance always end up mattering in one place or another.
Also, clisp simply is not liked that much as a general purpose language.
I know all this stuff seems revolutionary and it is nice that the software industry is paying attention to some of this stuff but it is not new.
It is far from new, but it is revolutionary that it is actually being put into practice.
Heck most c apps for unixes have only needed a recompile to run on just about any platform out there. It is pretty easy to write architecture independent code, it is just that most people that write software should be flipping burgers instead.
No, this is plain wrong. There can be major issues (I have dealt with them, as I have used C for more than 20 years) - threading libraries, different GUI systems, different word lengths, different socket handling for networks... etc, etc...
This also means that a serious application developer has to compile and test the app on each platform. With the JRE you get a certified and tested level of compatibility. The 'write once run anywhere' principle of Java is not a myth - it is used daily by hundreds of thousands of developers.
Why even bother to do such things with C when Java is so much simpler? Portability guaranteed; run-time optimised; memory managed. Sure, C is neat for some quick jobs and other tasks, but for most of us, there is simply no need.
Having said that how many times have top level brass from sun contradicted each other? Sun is clearly an organization where the left hand doesn't know what the right hand is doing.
Clearly? All large organisations have such disagreements. Sun is honest and has them openly.
Java web start requires you to click.
Are we really reduced such a dislike of Java that there is an objection to the fact that the update is only done on demand when the user starts an appliction by clicking a mouse?
To me that is automatic - the JRE gets automatically updated when a program is started.
There is also the Java Control Panel for Windows - that updates Java automatically and you don't even have to click anything.
One set requires an agreement with Sun on redistribution of the runtime and the JDK. IANAL but various distributions have read the license and have chosen not to be bound by it's terms, apparently they found it onorous.
And others haven't.
Maybe you think sun would never sue you but we have seen many examples of large corporations and their legal departments suing people willy nilly and winning. Do you really want to take that risk?
Yes, I do. Because I don't go along with some of the most extreme open source supporters that all software protection is bad and commercial companies are inherently evil.
In a nutshell. Sun does not want you to automatically download and install their software. Not if you are a distributor (a distro) and not if you are an end user.
It is strange, then, that they provide a tool that does just that - it is called Java Web Start. It automatically downloads and installs newer versions of the JRE as required by Java software.
My job is a programmer not a sysadmin. I just have to do sysadmin tasks also. All of our apps are written in python and are for zope. None of the systems need java in any way. Since my main job is writing custom software and I do that with zope why should I make my life harder then it has to be?
In that case, there is obviously no problem. But I suspect that this is rare. Java is increasingly the language of choice for substantial server side development. I think it is going to get hard to get by without it.
It is not like it is a vastly better language then anything else out there.
No; but it is a vastly better development and deployment system than almost anything else. I know of no other mainstream language that allows for production of high-performance apps than can be deployed unchanged on such a wide range of platforms, and even on 32-bit and 64-bit, from the same binary! That has been a revolution for the software industry, no matter how much open source supporters try and dismiss the language.
but if I have to choose between c# and java my choice will be c# because mono is trivial to install and keep updated.
Fine, but you are cutting yourself and your users off from what is probably the most used development language for commercial apps today over what is, to be honest, nothing more than a matter of minor convenience. It is really worth it? How difficult it is to check for Java updates, say, once a month, and script the automatic distribution and installation of these things? Surely that is the kind of thing Linux fans are more than capable of doing? Or are we becoming reduced to the level of Windows users, where if we don't have tools to do things, it is all too much trouble??
Sorry, but this really does seem to me to be trying to set up deliberate hurdles in the way of using Java that would not be put up for other languages. There have been major issues in past years with updates and version tracking for other popular languages - PERL, versions of gcc etc. Did open source advocates shrug off these languages, saying that they were too much bother to support?
When did Linux sysadmins, who have in the past struggled valiantly through things like kernel re-compilations, suddenly become so feeble when faced with a few command-line actions, like those required to install Java?
Not for GNU/Linux users. Java will only work if you use i386 and know how to use the command line.
No. I can do all Java installation using KDE without a command line.
some dodgy third-party binary installer.
Sorry, but calling a Java installer from Sun 'some dodgy third-party binary installer' is simply open-source snobbery, and can't be a comment that I can take seriously! Come on now - Sun is not dodgy!
On GNU/Linux users must open the scary terminal and type a complicated command to get it to install.
No, you don't.
If Sun's license was a little more liberal then proper packages would be made that could go into the non-free section. Then the end user wouldn't need to notice that their program was written in/for/with Java; a JRE would be pulled in automatically when they installed it, just like any other dependancy.
Yes, I agree. However, why can't it work the other way? Why can't the Linux distributions restrictions be a little less restrictive?
Really? You mean I can just apt-get install JDK? You mean if I apt-get install your program the JDK or the JRE will be downloaded and installed? If I apt-get install a python program pyton gets installed, if I apt-get install a C# program mono gets installed automatically. If I don't have the same convenience with java why should I use your program?
For millions of users, the simple click of a link on a web page leads to the download of a JRE. apt-get is a command-line gateway to a selection of software highly constrained (for better or worse) by a set of political ideals.
Personally, I use Debian, and I like it and I like apt, but to consider it the only convenient way to install software is rather extreme.
I mean, if you can use apt-get, are you really trying to tell me you would have trouble downloading a Sun JRE for Linux and running the binary that results?
You can blame me all you want but why should I use your program if I have to go through extra hoops before I can even install it and use it?
Are you seriously trying to say that clicking a link to download a binary, then running the binary is going through extra hoops?
Even novice users on Windows or MacOS can manage that!
Does 'apt-get' limit absolutely everything you do?
Java is just another programming language.
No, it isn't. It is a free language and development system that has helped successfully hold back Microsoft serverside, and has certainly helped Linux become a serious platform for server-side corporate use as an application hosting operating system. No other language or platform has come close to having such a significant influence.
If Sun doesn't want to play nice with open source other people can't make them.
Sure, but that hasn't stopped Java becoming one of the most widely used and successful development languages. Sun should change some of the silly restrictions on their licenses, but at the same time some of the more extreme open sourcers need to drop some of their arrogant attitude towards anything that doesn't quite meet their ideals.
And, anyone who claims that 'Sun doesn't want to play nice with open source' is showing ignorance of Sun's huge contribution to open source. Used GNOME? NFS? Open Office? NetBeans?
"He needs Sun and Java and Torvalds and ESR and Red Hat and everyone else"
No. Emphatically no. It's the other way around. The corps desperately need him. Most of them tried it the proprietary way for years and lost to Microsoft.
Interesting. I would love to hear how Red Hat and Sun (with Java) have lost to Microsoft. On the contrary, Microsoft sees commercial Linux and Java as major threats, and it has singularly failed to compete with them in major areas.
If you include water vapor as a greenhouse gas, humanity contributes 0.28% of the total greenhouse gases in the atmosphere, of which 0.117% is CO2. It's not likely that such a small amount is significant.
That is simply incorrect physics. Different types of gas have different absorption effects. For example, even small amounts of methane can have a dramatic influence.
What matters is not the amount - it is the changes.
How do you know we've had a significant impact on climate change? The most recent climate reconstructions (Moberg and others) have dumped the "hockey stick" graph used in the IPCC for one that includes the Little Ice Age and Medieval Warm Period.
Some people have dumped it; others haven't. The consensus seems to be (for now) that it is real. Also, that many of those fluctuations were somewhat localised. We are now experiencing widespread and definitely global effects.
When you look at those new graphs, it's clear that there've been warming and cooling cycles before Mankind could have caused them.
Of course there have. It is not like the Earth was a simple stable system before mankind. However, to deny that dumping vast amounts of CO2 into the atmophere isn't a highly dangerous thing to do and that it could potentially (and most likely has) cause climatic shifts is extremely poor reasoning.
Recently, researchers have found "that living plants emit 10 to 100 times more methane than dead plants. Scientists had previously thought that plants could only emit methane in the absence of oxygen." None of the climate models currently account for the impact of living plants on GW.
And this is irrelevant. Methane levels are low, and if plants are doing this now, they have always been doing it. As I said, that matters is changes in things.
"Sunshine levels had been decreasing by 2 per cent a decade between 1960 and 1980 - a total decline of about 6 per cent. Now they are going up again. Perhaps this is why our Swiss glaciers are melting," Professor Wild said.
Yes, and this is because of pollution controls and a decrease in suspended particulate matter. This is now why our glaciers are melting - it is why they are melting faster, and why they will melt even faster in future. This adds to the effect of CO2 - it does not explain global warming - it makes it more worrying, as the effect of CO2 on temperature could well have been held back by past dimming of sunlight.
No, the main source of warming is the Sun. CO2 might trap the heat in the Earths atmosphere, but the bulk of that heat comes from the sun.
'Warming' implies increase in temperature. The Sun does not result in an increase in temperature: its output is relatively constant. CO2 increase will result in an increase in trapping, hence 'warming'.
Is that mankinds fault? Or something we can do something about? If the sun gets hotter, so do the planets.
So what?
You don't want RMS to compromise.
How do you know what I want?
You want him to abandon his ideals and vision.
I certainly don't.
For what exactly?
I want him to realise that some who he seems to think are the enemy really aren't.
Sun has very onerous provisions on their java licensing which prevents inclusion of the JDK in a lot of Linux distributions. Why is this good for java? Why is it good for you (the java programmer) that I have to jump trough fifty hoops to install a JDK or a JRE before I can even run your program?
Er - what hoops? One click and it is downloaded. You don't have to jump through the hoops. I am free to redistribute the JRE with any Java program I produce.
Why is it good for you that the java implementation on my linux box is two years out of date and is slow?
That is the fault of the particular Linux distribution, and their political attitudes.
It is also your fault for now downloading the java implementation - which you are free to do at any time,
How would RMS compromise to make all that better for you?
By accepting that not everyone who is not entirely in support of his ideas is against him.
How could Sun compromise to make that better?
Some of their licensing is silly, I agree. That should be changed. But only a wild fanatic would deny the major benefits of Java to platforms such as Linux.