I completely agree, but I wouldn't exactly say that Java is dead on the web. If it were then how would you explain sites like jars and javaboutique ?
Personally, I have recently decided to write all future applets for 1.3+. It is just a lot easier to forget about the browser VM's and concentrate on Sun's VM.
At first I thought that nobody would wait for the download. As it turned out, nobody cares about the download. My first 1.3 applet (a swing based ftp client) was a huge success and literally thousands of people have upgraded their VMs to use it.
What people want from an applet is pretty much the same thing they want from an application. They want a useful app that is easy to use and looks good. They could care less who makes the VM. Hands down, you can make a better app with JRE1.4 than with MSVM1.1.3, so the choice for me was clear.
I pretty much agree with you about the 1.0 and 1.1 VMs. MS had a better VM than NS until 4.0.8. Then NS at least had a compatible VM but it was still slower than the MS VM.
However, this is a bit like discussing the best fortran compiler. Its ancient history. I have decided to write all my applets for JDK1.3+ now. Nobody gives a rats about the download. One of my applets is a swing based ftp client, literally thousands of people have downloaded the VM just to use this applet. As long as the process is automatic people don't mind one bit.
And, as far as swing and the ever-popular "speed" issue, well, I admit that swing was pretty slow when it came out, in what, 98-99, but happily very few people are still using the P1-133. It runs plenty fast on my pokey p3-800, and probably even better on a p4-2.8.
The only 1.1 work I do now is on some old code that I am sprucing up to give away.
I don't know how this war/dead/zealot/killer thing got started but I think Java's future is pretty secure. The demise of the MSVM is, if anything, a good thing.
I have noticed that whenever they have a windowing gui in MovieOS it is always some sort of X toolkit. I've seen Motif, Gnome, and even some aqua (only when they are pushing Apple gear) and a wide array of themes.
I guess the important thing is to make sure it looks different than anything anyone uses in real life:-(
Sounds strange coming from IBM. Then Hogan's reply is even more spin, Sun == MS in his estimation. I guess he would love it we all started slagging Sun with the enthusiasm we have for MS.
I sure get sick of corps waving their flags and inciting the community towards one revolution or another.
On my app, I found Resin to be maybe 20% faster, but the ease of configuration is what sold me. Tomcat is just too quriky and unpolished. Even if the speeds were reversed, I would rather buy more hardware than deal with Tomcat.
e.g. When connecting to Apache, resin provides a single connector for both webapps and the webroot, all config is done in the resin.conf. Tomcat has 3 connectors.
e.g.2 A quick look at the config files for resin vs. tomcat, shows how simple the resin is to configure. I have seen people struggle with tomcat for half a day to get it running the first time, resin takes the same person 5-10 minutes.
Tomcat is cool, but I find Resin (Caucho.com) to be easier to configure and use, especially with Apache.
I use Resin, standalone, on 7 load balanced web servers for one of my clients. I have been pretty happy with it. I also use it for my tiny Java web hosting biz <ShamelessPlug>(europasystems.net).</Shamele ssPlug>
Doesn't ATI provide specs and support to the OSS community, that allows them to include the OSS DRI for XFree4?
I use NVidea now, but I was thinking of switching to ATI so I wouldn't have to mess with the drivers as much (since they come with redhat).
Actually mac and win are the proprietary OS's. Unix OS's, by definition, adhere to "open systems" standards. This makes it easy to cross compile the same software for Solaris, HP/UX, OSF-1, IRIS, AIX, BSD, Linux, etc.
That's what unix is, an open standard that anyone can choose to implement. Windows, on the other hand, is a closed standard, which makes life so difficult for the wine project.
Apple and MS choose to not accept open systems standards for their own business reasons. I think MS was even developing an open systems OS before they decided to chuck it and create NT/win32.
Its sometimes called dumping, but only when it happens to be a foreign company. The sad thing is that they consider windows to be better than linux, at the same price.
Actually, I think that they have some sort of network install thing called sms, or something. Not that anyone uses it:-) Probaly costs a bit more than aptget and bash too.
That't the way it works for me, but I am at a loss to explain why the majority of win32 users are still using outlook.
Most people, it seems, are sheep. If you put crap in front of them they will use it. Finding an alternative takes too much thought, so they are willing to put up with one little annoyance after another. After a while they're lives are filled with a big headache that requires them to jump through hoops, buy and install virus checkers that don't always work, and reinstall everything every few months. Why not try to find something better? Too much thought required, and they just can't believe that a bunch of hackers can create something better than a huge gov supported monopoly.
The diamond rio had some support but it was shaky and anything but user-friendly. I was also not impressed with the 64 mb of space either.
The portable hard drive options looked good, lots of space and easy to configure Linux to mount the thing. I was a little worried about how long it would take to move a gb of stuff through USB.
I eventually decided to get a sony diskman with mp3 support. This allowed me to play existing cds, or burn cdrw in mp3 mode. The cd burning utilities were more advanced than the rio tools and I could burn different disks with different stuff and then easily switch out the disks. The thing runs for many hours on a set of batteries, at least 10h. Its not as portable as the rio, but not much bigger than an Archos.
Some history: In 1962 Dr. Benvold Austere introduced the 12 line bar code. His contemporaries pleaded with him to include an extra line, but he didn't. The reason? Computers in the early 60's has very little RAM (random access memory) and so the 13th line was left off to save space.
The Issue: Due to a proliferation of consumer products, mostly due to Lucas Film merchandising, a thirteenth line will be required to be able to identify every possible consumer product.
Predictions: Computer experts agree that 12 line barcode math is pervasive, throughout the industry. This presents a ticking bomb for retailers. Several scanners have been demonstrated to detonate when scanning an extra line. Some scanners merely misread the code.
In a strange twist to this story a mysterious renaissance hermit predicted this event:
"And lo the line will become as the circles and their number will be uncountable by the great IBM and the lawyers will argue whilst molten ash spreads accross the face of the earth. Neither will the sea be saved, much less its canned tuna!" - Frescobaldi Chianti, 1642
What You Should Do Now: 1. Immediately suspend all new IS projects. 2. Pour all available resources into evaluating whether your existing systems are bc13 complient. 3. Hire expert consultants, like Arthur Anderson, to help you evaluate your systems and to ensure that you have properly execised due diligence. 4. Demand a bc13 statement from all your hardware and software suppliers. Even if the software has nothing to do with barcodes, DEMAND this statement anyway! 5. Upgrade all of your commercial software to the latest versions, just to be sure. 6. Hire as many cobol programmers as you can find. 7. Go to bc13 seminars.
Conclusion: bc13 is nothing short of an impending calamity of global scale. It has the propensity to destroy all social interaction sending us scuttling back to the dark ages.
But, do not worry, with care, a great deal of money, and above all, due diligence, we can meet this challenge, just as we met y2k, and, like y2k, we can emerge from this horror, pretty much without a hicup.
The problem has little to do with the Internet and everything to do with the importance of the US market.
Even if you went so far as to split the european internetwork from the north american one, the US would still lock you up for DMCA infractions or posting a picture of Mickey on your web site.
A major problem is the imposition of US laws on other countries. Couple this with the fact that in a lot of instances the US government abdicates their responsibilities to corporations and one gets all manner of mayhem.
Apparently, in the US, international trade is a subject for the courts and US businesses, not the government. We all no how strange the US courts are, and you can imagine how US businesses view international trade. I'll sum up their complex position for you: "My stuff going out is good, Their stuff coming in is bad". So, if you happen to be in the unenviable position of, say, Canada or Mexico, you are stuck. You can't ignore the US economy without a generation of poverty, and yet you can't stop the US.coms from imposing huge dutys on your products; despite any number of deals you sign with.gov or any number of international trade rulings you might secure.
Compared to C and C++ its going to be slower, but against interpreted languages, like python and perl, its going to be much faster in the loops.
It is an easy language from a coding standpoint. Not nearly as dangerous as C and C++, so you won't spend all your time looking for "stacks in my structs, arrays in my queues; I got the segmentation violation, core dump blues" -mod-fortune.
You can compile Java to native code, if speed is of paramount importance, and it has thread support built in. Does FORTRAN have thread support? On a multi-cpu system this might make all the difference.
In short, I would sooner code VB than take up FORTRAN again. I would rather deal with C, hands down. I coded once in FORTRAN 77, back in 1988, even then it was archaic, but it was all the VAX understood.
"Remove Thread Groups" - These things are used for security in applets. A thread must be in the Applet's thread group for you to modify its properties (priority, name, etc.).
"Eliminate primitive data types" - This is just plain stupid. Why create an object to use as a for loop counter? If you want to use objects for all of your primitives, go right ahead, use Float and Long, etc. Personally, I prefer to get something in exchange for giving up efficiency, and Object oriented utopia doesn't cut it.
"Drop all depricated stuff" - Why can't this stuff be phased out slowly. What's with the shock treatment? Why not get rid of all the 1.0.2 stuff depricated in 1.1 now and, when most people have adopted 1.2+ vms, we can start to make the other changes. Either way, what's the big deal? If we take them out of the docs, will that be enough? Don't say "in the interests of efficiency", not after the "eliminate primitives" thing.
"eliminate 1.0.2 event model" - Well, I can't argue with that.
"replace all data formats with xml" - Mighty fashionable, but if you try to pass serialized objects to an earlier VM, say from a 3.0 server to a 1.1 applet, you will also have to pass along several hunded K of parser. I suppose we could use comma sep instead, but since my StringTokenizer is gone too, I don't know what to do.
Most of these suggestions are good, but I see no reason to leap into them in one version. Backward compatibility IS important. A lot of coders "don't get that depricated means don't use this in the code you are writing now", but people who code this way deserve what they get.
The point is a good one, but I think the Linux desktop community is far more experimental than either Apple or MS.
Some times I think that 90% of the effort put into the Linux platform is in the form of GUI work.
Linux has multiple windowing systems, of which XFree is the most popular. Within the X realm, there are two major toolkits, QT and Gnome, maybe a hundred window managers, countless themes.
All of this experimentation has got to produce some good ideas sooner or later, at least different ideas.
Gnome uses a Corba orb for communication between desktop apps, that's pretty different, isn't it? Works pretty well too. I installed a spell checker and my mail client knew about it without me having to restart it.
I don't think that the writer of this article looked very deep before he labeled the Linux GUI projects, enmasse, as "emulating the Windows story". He probably uses a Mac and only played around a bit with Linux to research this story.
And as for the mobile revolution, who is naieve enough to think that meaningful work can be done on the beach with a IPaq? Isn't this is just some sort of fantasy perpetuated by handheld manufacturers?
These guys don't know much about DOS if they think they can just start throwing packets without consequences. They will only succeed in getting their ips banned by the major providers. Unless, of course they operate like the h4x0r crowd and install trojans on a variety of innocent users' machines.
Why do they think that other internet providers will allow them to clog up their net with a bunch of empty packets?
Slashdot Mirror
I completely agree, but I wouldn't exactly say that Java is dead on the web. If it were then how would you explain sites like jars and javaboutique ?
Personally, I have recently decided to write all future applets for 1.3+. It is just a lot easier to forget about the browser VM's and concentrate on Sun's VM.
At first I thought that nobody would wait for the download. As it turned out, nobody cares about the download. My first 1.3 applet (a swing based ftp client) was a huge success and literally thousands of people have upgraded their VMs to use it.
What people want from an applet is pretty much the same thing they want from an application. They want a useful app that is easy to use and looks good. They could care less who makes the VM. Hands down, you can make a better app with JRE1.4 than with MSVM1.1.3, so the choice for me was clear.
I pretty much agree with you about the 1.0 and 1.1 VMs. MS had a better VM than NS until 4.0.8. Then NS at least had a compatible VM but it was still slower than the MS VM.
However, this is a bit like discussing the best fortran compiler. Its ancient history. I have decided to write all my applets for JDK1.3+ now. Nobody gives a rats about the download. One of my applets is a swing based ftp client, literally thousands of people have downloaded the VM just to use this applet. As long as the process is automatic people don't mind one bit.
And, as far as swing and the ever-popular "speed" issue, well, I admit that swing was pretty slow when it came out, in what, 98-99, but happily very few people are still using the P1-133. It runs plenty fast on my pokey p3-800, and probably even better on a p4-2.8.
The only 1.1 work I do now is on some old code that I am sprucing up to give away.
I don't know how this war/dead/zealot/killer thing got started but I think Java's future is pretty secure. The demise of the MSVM is, if anything, a good thing.
I have noticed that whenever they have a windowing gui in MovieOS it is always some sort of X toolkit. I've seen Motif, Gnome, and even some aqua (only when they are pushing Apple gear) and a wide array of themes.
:-(
I guess the important thing is to make sure it looks different than anything anyone uses in real life
Sounds strange coming from IBM. Then Hogan's reply is even more spin, Sun == MS in his estimation. I guess he would love it we all started slagging Sun with the enthusiasm we have for MS.
I sure get sick of corps waving their flags and inciting the community towards one revolution or another.
On my app, I found Resin to be maybe 20% faster, but the ease of configuration is what sold me. Tomcat is just too quriky and unpolished. Even if the speeds were reversed, I would rather buy more hardware than deal with Tomcat.
e.g. When connecting to Apache, resin provides a single connector for both webapps and the webroot, all config is done in the resin.conf. Tomcat has 3 connectors.
e.g.2 A quick look at the config files for resin vs. tomcat, shows how simple the resin is to configure. I have seen people struggle with tomcat for half a day to get it running the first time, resin takes the same person 5-10 minutes.
Tomcat is cool, but I find Resin (Caucho.com) to be easier to configure and use, especially with Apache.
e ssPlug>
I use Resin, standalone, on 7 load balanced web servers for one of my clients. I have been pretty happy with it. I also use it for my tiny Java web hosting biz <ShamelessPlug>(europasystems.net).</Shamel
Doesn't ATI provide specs and support to the OSS community, that allows them to include the OSS DRI for XFree4? I use NVidea now, but I was thinking of switching to ATI so I wouldn't have to mess with the drivers as much (since they come with redhat).
25 is the perfect age to start a business.
1. Failure doesn't matter. He can lose everything, worse case, he gets a job, after learning a bunch about business.
2. At 25 he is probably still living like a student, drinking domestic beer, eating peanut butter, etc. His expenses are nil.
Actually mac and win are the proprietary OS's. Unix OS's, by definition, adhere to "open systems" standards. This makes it easy to cross compile the same software for Solaris, HP/UX, OSF-1, IRIS, AIX, BSD, Linux, etc.
That's what unix is, an open standard that anyone can choose to implement. Windows, on the other hand, is a closed standard, which makes life so difficult for the wine project.
Apple and MS choose to not accept open systems standards for their own business reasons. I think MS was even developing an open systems OS before they decided to chuck it and create NT/win32.
Its sometimes called dumping, but only when it happens to be a foreign company. The sad thing is that they consider windows to be better than linux, at the same price.
Owned by Microsoft vs. 0w3N3d by h4X0r M4sT3r
/. poll:
Sounds like one for a
Would you rather be:
1. Owned by Microsoft
2. 0w3N3d by h4X0r M4sT3r
3. Given away by RMS
4. Used and discarded by nobody in particular
5. CowboyNeal's bitch
Actually, I think that they have some sort of network install thing called sms, or something. Not that anyone uses it :-) Probaly costs a bit more than aptget and bash too.
That't the way it works for me, but I am at a loss to explain why the majority of win32 users are still using outlook.
Most people, it seems, are sheep. If you put crap in front of them they will use it. Finding an alternative takes too much thought, so they are willing to put up with one little annoyance after another. After a while they're lives are filled with a big headache that requires them to jump through hoops, buy and install virus checkers that don't always work, and reinstall everything every few months. Why not try to find something better? Too much thought required, and they just can't believe that a bunch of hackers can create something better than a huge gov supported monopoly.
The diamond rio had some support but it was shaky and anything but user-friendly. I was also not impressed with the 64 mb of space either.
The portable hard drive options looked good, lots of space and easy to configure Linux to mount the thing. I was a little worried about how long it would take to move a gb of stuff through USB.
I eventually decided to get a sony diskman with mp3 support. This allowed me to play existing cds, or burn cdrw in mp3 mode. The cd burning utilities were more advanced than the rio tools and I could burn different disks with different stuff and then easily switch out the disks. The thing runs for many hours on a set of batteries, at least 10h. Its not as portable as the rio, but not much bigger than an Archos.
Is this even a word? They mean "moves" right?
bc13 is rapidly approaching, are you prepared?
Some history:
In 1962 Dr. Benvold Austere introduced the 12 line bar code. His contemporaries pleaded with him to include an extra line, but he didn't. The reason? Computers in the early 60's has very little RAM (random access memory) and so the 13th line was left off to save space.
The Issue:
Due to a proliferation of consumer products, mostly due to Lucas Film merchandising, a thirteenth line will be required to be able to identify every possible consumer product.
Predictions:
Computer experts agree that 12 line barcode math is pervasive, throughout the industry. This presents a ticking bomb for retailers. Several scanners have been demonstrated to detonate when scanning an extra line. Some scanners merely misread the code.
In a strange twist to this story a mysterious renaissance hermit predicted this event:
"And lo the line will become as the circles and their number will be uncountable by the great IBM and the lawyers will argue whilst molten ash spreads accross the face of the earth. Neither will the sea be saved, much less its canned tuna!" - Frescobaldi Chianti, 1642
What You Should Do Now:
1. Immediately suspend all new IS projects.
2. Pour all available resources into evaluating whether your existing systems are bc13 complient.
3. Hire expert consultants, like Arthur Anderson, to help you evaluate your systems and to ensure that you have properly execised due diligence.
4. Demand a bc13 statement from all your hardware and software suppliers. Even if the software has nothing to do with barcodes, DEMAND this statement anyway!
5. Upgrade all of your commercial software to the latest versions, just to be sure.
6. Hire as many cobol programmers as you can find.
7. Go to bc13 seminars.
Conclusion:
bc13 is nothing short of an impending calamity of global scale. It has the propensity to destroy all social interaction sending us scuttling back to the dark ages.
But, do not worry, with care, a great deal of money, and above all, due diligence, we can meet this challenge, just as we met y2k, and, like y2k, we can emerge from this horror, pretty much without a hicup.
The problem has little to do with the Internet and everything to do with the importance of the US market.
.coms from imposing huge dutys on your products; despite any number of deals you sign with .gov or any number of international trade rulings you might secure.
Even if you went so far as to split the european internetwork from the north american one, the US would still lock you up for DMCA infractions or posting a picture of Mickey on your web site.
A major problem is the imposition of US laws on other countries. Couple this with the fact that in a lot of instances the US government abdicates their responsibilities to corporations and one gets all manner of mayhem.
Apparently, in the US, international trade is a subject for the courts and US businesses, not the government. We all no how strange the US courts are, and you can imagine how US businesses view international trade. I'll sum up their complex position for you: "My stuff going out is good, Their stuff coming in is bad". So, if you happen to be in the unenviable position of, say, Canada or Mexico, you are stuck. You can't ignore the US economy without a generation of poverty, and yet you can't stop the US
Compared to C and C++ its going to be slower, but against interpreted languages, like python and perl, its going to be much faster in the loops.
It is an easy language from a coding standpoint. Not nearly as dangerous as C and C++, so you won't spend all your time looking for "stacks in my structs, arrays in my queues; I got the segmentation violation, core dump blues" -mod-fortune.
You can compile Java to native code, if speed is of paramount importance, and it has thread support built in. Does FORTRAN have thread support? On a multi-cpu system this might make all the difference.
In short, I would sooner code VB than take up FORTRAN again. I would rather deal with C, hands down. I coded once in FORTRAN 77, back in 1988, even then it was archaic, but it was all the VAX understood.
"Remove Thread Groups" - These things are used for security in applets. A thread must be in the Applet's thread group for you to modify its properties (priority, name, etc.).
"Eliminate primitive data types" - This is just plain stupid. Why create an object to use as a for loop counter? If you want to use objects for all of your primitives, go right ahead, use Float and Long, etc. Personally, I prefer to get something in exchange for giving up efficiency, and Object oriented utopia doesn't cut it.
"Drop all depricated stuff" - Why can't this stuff be phased out slowly. What's with the shock treatment? Why not get rid of all the 1.0.2 stuff depricated in 1.1 now and, when most people have adopted 1.2+ vms, we can start to make the other changes. Either way, what's the big deal? If we take them out of the docs, will that be enough? Don't say "in the interests of efficiency", not after the "eliminate primitives" thing.
"eliminate 1.0.2 event model" - Well, I can't argue with that.
"replace all data formats with xml" - Mighty fashionable, but if you try to pass serialized objects to an earlier VM, say from a 3.0 server to a 1.1 applet, you will also have to pass along several hunded K of parser. I suppose we could use comma sep instead, but since my StringTokenizer is gone too, I don't know what to do.
Most of these suggestions are good, but I see no reason to leap into them in one version. Backward compatibility IS important. A lot of coders "don't get that depricated means don't use this in the code you are writing now", but people who code this way deserve what they get.
What do their clients in India think about paying for a busted net connection?
"Also used by British Nuclear Fuels Ltd." WTF!
And how would you defend against a ddos attack?
The point is a good one, but I think the Linux desktop community is far more experimental than either Apple or MS.
Some times I think that 90% of the effort put into the Linux platform is in the form of GUI work.
Linux has multiple windowing systems, of which XFree is the most popular. Within the X realm, there are two major toolkits, QT and Gnome, maybe a hundred window managers, countless themes.
All of this experimentation has got to produce some good ideas sooner or later, at least different ideas.
Gnome uses a Corba orb for communication between desktop apps, that's pretty different, isn't it? Works pretty well too. I installed a spell checker and my mail client knew about it without me having to restart it.
I don't think that the writer of this article looked very deep before he labeled the Linux GUI projects, enmasse, as "emulating the Windows story". He probably uses a Mac and only played around a bit with Linux to research this story.
And as for the mobile revolution, who is naieve enough to think that meaningful work can be done on the beach with a IPaq? Isn't this is just some sort of fantasy perpetuated by handheld manufacturers?
These guys don't know much about DOS if they think they can just start throwing packets without consequences. They will only succeed in getting their ips banned by the major providers. Unless, of course they operate like the h4x0r crowd and install trojans on a variety of innocent users' machines.
Why do they think that other internet providers will allow them to clog up their net with a bunch of empty packets?