Changing from Unix to Linux and throwing a few old bones to the OSS crowd isn't "betting the farm". IBM is still very committed to its proprietary software products. For example a few years ago IBM acquired Rational. Immediately afterword they discontinued the popular Visual Test product because it competed with more expensive products IBM owned. They won't sell you a license for it and they won't convert it into an open source project.
IBM's commitment to OSS is very shallow and if OSS disappeared tomorrow IBM would keep right on rolling' like a Hummer running over a dead mouse.
The widespread adoption of BSD's TCP/IP implementation probably has more do to with how poorly the protocol is documented rather than compainies' inablity to do it on their own.
Sure, ASP.NET 2.0 doesn't run on Unix platforms, just as most other web servers only run on Windows and Unix compatible OS's. I was looking for a more significant factor. In my view you pick the web server that works best for your requirements and then choose the OS that it runs best on. If I were using Apache and it ran best under Linux, I couldn't care less whether it also runs less efficiently under Windows.
Of course the primary advantage of a web app is the ability to be used from a variety of OS's. If you have an in-house application and you can control the configuration of the computers, a "fat client" application is probably going to be easier to develop and give a better user experience than an AJAX-based one.
"What's truly significant is the ability to make web requests and processing in the background."
This is really just a hack to get around the fact that web standards weren't designed for anything beyond static pages. A better solution would be to rework these standards so that any user interaction with an HTML element could optionally trigger a request. For most web apps this would make Javascript unnecessary.
I can say that your comments don't logically follow from the evidence, but I don't have enough evidence of my own to conlude that your reasoning is always this poor or if this post isn't representive of your general reasoning abilities.
I said "I know of no..", so naturally unless you know my mind better than I do, my statement is a fact.
As far as the main point is concerned, I don't doubt that there are scientists who think that IQ tests are valid, but repeatable results of an experiment across a large population doesn't prove anything about what is actually being measured.
I used the adjective "comprehensive" quite deliberately because the more narrowly you define intelligence, the easier it is to "measure" and the less meaningful the measurement becomes.
I know of no comprehensive definition of intelligence that is agreed upon by a majority of scientists, but if you have evidence to the contrary feel free to provide it. Obviously, there isn't going to be any scientific definition of "success in the real world".
I think everybody here understands that the GPL doesn't put any restrictions on running the software and that the GPL isn't a EULA.
Having said that a "click-through" agreement doesn't automatically imply a EULA even if in practice it usually is. Since GPL'd software is either distributed with source or available to the user on request and the GPL puts specific restrictions on the use of that source code, it's entirely appropriate that the receiver of this bundled (binary and source) product be made conspicuously aware of and agree to all the terms that they may be bound by.
In short the value of a click-through agreement is exactly the same whether it is a closed source EULA or the GPL: it informs the individual receiving the software of possible legal limitations and makes it much more difficult to for that individual to claim ignorance of the license.
Rochelle is the project manager. Of course he's going to say "everybody is using it". It's just marketing-speak. He probably wouldn't recognize some of Google's employees if he ran into them on the street, let alone know which tools they use.
"Transmeta's processors had their own unique (VLIW) instruction set with a x86 compatibility "code morphing" layer over the top (mainly so you could run existing programs), which is nothing like how CPUs are traditionally designed."
The reason that traditional CPUs aren't designed that way is because it makes the processor slower when it runs code that requires morphing. Transmeta's whole business model was based on running x86 code, not on their lower-level instruction set.
There's a big difference between "Children that are denied human contact" and "not being hugged enough".
In any case, I would have to see some real scientific evidence before I'd believe that these problems were primarily caused by parents behavior toward their children. There are many loving parents who have children with these problems and it's an insult to suggest they are at fault for those conditions.
I think your attitude illustrates exactly what I was talking about.
The only reason OS X is based on NeXTstep is because part of Jobs' price for returning to Apple was to "erase" the Next failure. The core Apple faithful will buy just about anything Jobs comes up with and couldn't care less about "Unixy" features in their OS.
It takes a lot of balls to sue the company whose products form the basis of your business model. If these guys were so brilliant why didn't they create their own unique processor rather than create a super-slow version of Intel's.
You mean MS isn't that interested in developers that don't have the potential to make more money for them? You'd think there were a for-profit company or something.
Slashdot Mirror
"IBM is betting the farm on open source."
Changing from Unix to Linux and throwing a few old bones to the OSS crowd isn't "betting the farm". IBM is still very committed to its proprietary software products. For example a few years ago IBM acquired Rational. Immediately afterword they discontinued the popular Visual Test product because it competed with more expensive products IBM owned. They won't sell you a license for it and they won't convert it into an open source project.
IBM's commitment to OSS is very shallow and if OSS disappeared tomorrow IBM would keep right on rolling' like a Hummer running over a dead mouse.
The widespread adoption of BSD's TCP/IP implementation probably has more do to with how poorly the protocol is documented rather than compainies' inablity to do it on their own.
Sure, ASP.NET 2.0 doesn't run on Unix platforms, just as most other web servers only run on Windows and Unix compatible OS's. I was looking for a more significant factor. In my view you pick the web server that works best for your requirements and then choose the OS that it runs best on. If I were using Apache and it ran best under Linux, I couldn't care less whether it also runs less efficiently under Windows.
Of course the primary advantage of a web app is the ability to be used from a variety of OS's. If you have an in-house application and you can control the configuration of the computers, a "fat client" application is probably going to be easier to develop and give a better user experience than an AJAX-based one.
"What's truly significant is the ability to make web requests and processing in the background."
This is really just a hack to get around the fact that web standards weren't designed for anything beyond static pages. A better solution would be to rework these standards so that any user interaction with an HTML element could optionally trigger a request. For most web apps this would make Javascript unnecessary.
What are the multiple things that other servers understand that ASP.NET servers don't?
Didn't you get the Slashdot memo that said the MS has never invented anyting accept MS Bob? So naturally, what you say can't possibly be true.
I can say that your comments don't logically follow from the evidence, but I don't have enough evidence of my own to conlude that your reasoning is always this poor or if this post isn't representive of your general reasoning abilities.
I said "I know of no..", so naturally unless you know my mind better than I do, my statement is a fact.
As far as the main point is concerned, I don't doubt that there are scientists who think that IQ tests are valid, but repeatable results of an experiment across a large population doesn't prove anything about what is actually being measured.
I used the adjective "comprehensive" quite deliberately because the more narrowly you define intelligence, the easier it is to "measure" and the less meaningful the measurement becomes.
I always thought that the phrase "real world" excluded academic achievement by definition.
I know of no comprehensive definition of intelligence that is agreed upon by a majority of scientists, but if you have evidence to the contrary feel free to provide it. Obviously, there isn't going to be any scientific definition of "success in the real world".
I think everybody here understands that the GPL doesn't put any restrictions on running the software and that the GPL isn't a EULA.
Having said that a "click-through" agreement doesn't automatically imply a EULA even if in practice it usually is. Since GPL'd software is either distributed with source or available to the user on request and the GPL puts specific restrictions on the use of that source code, it's entirely appropriate that the receiver of this bundled (binary and source) product be made conspicuously aware of and agree to all the terms that they may be bound by.
In short the value of a click-through agreement is exactly the same whether it is a closed source EULA or the GPL: it informs the individual receiving the software of possible legal limitations and makes it much more difficult to for that individual to claim ignorance of the license.
Rochelle is the project manager. Of course he's going to say "everybody is using it". It's just marketing-speak. He probably wouldn't recognize some of Google's employees if he ran into them on the street, let alone know which tools they use.
You're talking about a CIO, not a CFO, COO, or CEO. The CIO job doesn't come with huevos. That's a perk reserved for the big boys in the company.
"Don't take my first post so personally, I was just pointing out the irony of the situation"
But it's only ironic if it's true. Perhaps you were trying to be funny rather than ironic.
"Transmeta's processors had their own unique (VLIW) instruction set with a x86 compatibility "code morphing" layer over the top (mainly so you could run existing programs), which is nothing like how CPUs are traditionally designed."
The reason that traditional CPUs aren't designed that way is because it makes the processor slower when it runs code that requires morphing. Transmeta's whole business model was based on running x86 code, not on their lower-level instruction set.
There's a big difference between "Children that are denied human contact" and "not being hugged enough".
In any case, I would have to see some real scientific evidence before I'd believe that these problems were primarily caused by parents behavior toward their children. There are many loving parents who have children with these problems and it's an insult to suggest they are at fault for those conditions.
I think your attitude illustrates exactly what I was talking about.
The only reason OS X is based on NeXTstep is because part of Jobs' price for returning to Apple was to "erase" the Next failure. The core Apple faithful will buy just about anything Jobs comes up with and couldn't care less about "Unixy" features in their OS.
Thank you Doctor. I didn't realize there was a scientific link between OCD and a lack of hugging.
Seriously, this sounds like the typical US tendency to find someone to blame for every medical condition.
If you say so. I certainly didn't.
It takes a lot of balls to sue the company whose products form the basis of your business model. If these guys were so brilliant why didn't they create their own unique processor rather than create a super-slow version of Intel's.
You mean MS isn't that interested in developers that don't have the potential to make more money for them? You'd think there were a for-profit company or something.
Do you realize that GIGO is an older acronym than 99% of all the acronyms you've read on Slashdot.
So your question should be: "Jesus what was going on this world way back when before I was born!"
Of course this was the same pitch McNealy made a decade ago and the marketplace rejected it. This idea has failed. Sun should try something new.
I never said you said ... oh forget it.
It's very common on Slashdot to read that wine is the solution for running Windows applications on linux without mentioning any limitations.