I watched the documentary and it was quite interesting. They had interviewed a couple of ex-Microsofties about what they doing now (meaning 2001).
There was a programmer around 40 or so who spend 9 years at Microsoft and was totally burned out. He was kind of fired in the early 90s because he "forgot" (as he called it) how to program. He was apparently rich, but did not know what to do with his life. In the end he admitted, he just missed working at Microsoft so much.
Than there was a guy who started his own firm (www.sunhawk.com) with the millions he made cashing in his stock options. He loves both music and programming and sells sheet music over the internet. He also admitted that doing business on his own was a lot less fun than working for Bill. Well, who wouldn't say that, when he had lost almost $25 millions during the dot.com crash.
All the women and men who worked there talked about a very competitive environment, very, very long overtime and a lot of pressure. But they also liked the family-like structure and the way Microsoft is doing business: go for the top, don't care about how. They all thought that Microsoft hires only the top-level poeple (and of course they belong to them).
It all looked like a filmed version of Microserfs. All stereotypes and cliches were served.
The parents would have to sit in front of the computer all the time to watch where their child is going. And until they _realize_ that their child is tied to a railroad track, it's too late anyway.
I'm not saying that Microsoft does it better or perfect. But compared to it you don't gain anything in terms of better quality by using a Red Hat or SuSE distribution. Sure, you can hire someone to fix Linux to meet your needs, but your SysAdmin will leave the company sometime, and nobody knows the peculiarities of your "own" Linux. And a tech consultant who fixes Linux kernels is surely very expensive.
"Bull! A Red Hat "2.4.x" kernel usually contains a whole bunch of changes and bugfixes compared to the standard "2.4.x" kernel, even if Linus hasn't got round to adding the fix to "2.4.x+1", that's the beauty of open source."
Yes Red Hat fixes a lot in its kernel, but there are a lot of bugs left. And no, they do not fix it mainly by patching it, but rather have the customer wait for the next release. And the next release will leave too many customer with compatibility issues. The update cycle with Linux (distributions) is crazy. You cannot live with it in a mission critical environment.
Yes, I can pay anyone to fix my Linux problem. If I fix it long enough this way, I will end up with my own operating system.
Microsoft may not be "responsible", but they give support for certain features, e.g. that for a certain period of time kernel features are not touched, so that your application will run in that time frame. Also, they patch the system, also in a improvable way, on a regular basis. Red Hat or SuSE patch they distribution only in the range of 10% of known bugs. They tend to fix some bugs in the next distribution round. But even then only about 30% of the bugs are fixed. I know this from speaking with people who work there. Also, you have the "feature" of Linux to break combatibility with new kernel releases.
The article states
"There are different reasons why people advocate open source. One reason for enterprise is, 'You have the source code; if it doesn't work, you can fix it.' But the fact is, if I'm an enterprise, I don't want to fix it. I want somebody else to fix it," Goldman said.
That's a main concern for big customers, they want a support contract, which they can rely on.
The article states further
"The system vendors -- IBM, HP (NYSE: HWP), Compaq (NYSE: CPQ), Dell (Nasdaq: DELL) -- who sell Linux all support it. They resolve Level One and Level Two problems. If problems come up that they can't solve for the customer, they turn to Red Hat (Nasdaq: RHAT)," Claybrook explained.
This is true to some extent. The big vendors support Linux, but since they are neither the owner nor the main developer, they can't do anything in the end. If, say, there is a critical kernel bug, they will turn to the open source community to make it known, but they will never be responsible to fix it. Maybe the bug will never be fixed, so the customer is left on his own. At least this is what the support contracts of IBM and HP say . The support contracts say, that there is support for installation, configuration and _troubleshooting_, NOT for fixing the problem. That is the part of the community.
Why should someone ever do something as boring as installing Linux on the new iMac. I assume, if you are into Linux, it is for the stability of a Unix system, and maybe because you are used to it. But you are already running Unix on an iMac, and you are also running a FREE Unix, too. So what's the point of Linux on an iMac. I doesn't run smoothly and the GUI is horrible compared to Aqua.
We had the same discussions on heise.de in Germany many times. It's always like this. There are some whiners who insist on "fair use", saying it should be common sense, that no ISP can provide all its users being heavy users. Then there are others who say they know the fact, that all the heavy users are subsidized by the light users. The ISPs should get their calculations right and if it comes out to what Rogers Cable is planning, so be it.
BTW, from what I see posted here, Germany seems to be quite cheap for flat rate DSL - it's just USD25-30/month for 768KB/s down and 128 up, "unlimited":-)
I usually use Red Hat or SuSE Distros with their nice graphical installers, but for some old HP PA-RISC machines I have to rely on the Debian distribution, because Debian is the only one to include the PA-RISC architecture. While the installer is far from good-looking, it does the job. It's like YaST1 from SuSE or the FreeBSD installer. Do we really need a penguin looking at us while installing Linux?
I thought about this too. Why should I run my games under Linux, while Windows is running them faster and with better graphics quality. This has of course nothing to do with money, it's all about ideology.
If _everyone_ would use the _same_ version of Windows we wouldn't complain about compatibility issues:-)
From a more philosophical point of view...
On a meta level you always have to rely on binary logic. In the case of a base 3 logic you have to decide which one of the three possibilities is actually happen. If you go with the states 0,1,2 or -1,0,1, you or the device have to "know" whether it is 0 or 1 or 2 (resp. -1 or 0 or 1).
The "decision" which state is the current one is not a decision based on ternary logic, but based on binary logic. You can say, e.g. at the moment we have state 1 (true) and not 0 or 2 (which are both false). The decision isn't made on ternary logic itself, but falls back to binary.
Ternary logic is NOT self-reflexive, because you can't apply ternary logic to itself. If you do you will end in an endless regress, so it is unlikely to have an advantage after all.
Slashdot Mirror
I watched the documentary and it was quite interesting. They had interviewed a couple of ex-Microsofties about what they doing now (meaning 2001).
There was a programmer around 40 or so who spend 9 years at Microsoft and was totally burned out. He was kind of fired in the early 90s because he "forgot" (as he called it) how to program. He was apparently rich, but did not know what to do with his life. In the end he admitted, he just missed working at Microsoft so much.
Than there was a guy who started his own firm (www.sunhawk.com) with the millions he made cashing in his stock options. He loves both music and programming and sells sheet music over the internet. He also admitted that doing business on his own was a lot less fun than working for Bill. Well, who wouldn't say that, when he had lost almost $25 millions during the dot.com crash.
All the women and men who worked there talked about a very competitive environment, very, very long overtime and a lot of pressure. But they also liked the family-like structure and the way Microsoft is doing business: go for the top, don't care about how. They all thought that Microsoft hires only the top-level poeple (and of course they belong to them).
It all looked like a filmed version of Microserfs. All stereotypes and cliches were served.
How could this device ever have prevented this.
The parents would have to sit in front of the computer all the time to watch where their child is going. And until they _realize_ that their child is tied to a railroad track, it's too late anyway.
I'm not saying that Microsoft does it better or perfect. But compared to it you don't gain anything in terms of better quality by using a Red Hat or SuSE distribution. Sure, you can hire someone to fix Linux to meet your needs, but your SysAdmin will leave the company sometime, and nobody knows the peculiarities of your "own" Linux. And a tech consultant who fixes Linux kernels is surely very expensive.
Isn't Sun also the biggest vendor for Unix workstations (which put Solaris on desktops)? Oops, thats Apple now, isn't it?
"Bull! A Red Hat "2.4.x" kernel usually contains a whole bunch of changes and bugfixes compared to the standard "2.4.x" kernel, even if Linus hasn't got round to adding the fix to "2.4.x+1", that's the beauty of open source." Yes Red Hat fixes a lot in its kernel, but there are a lot of bugs left. And no, they do not fix it mainly by patching it, but rather have the customer wait for the next release. And the next release will leave too many customer with compatibility issues. The update cycle with Linux (distributions) is crazy. You cannot live with it in a mission critical environment. Yes, I can pay anyone to fix my Linux problem. If I fix it long enough this way, I will end up with my own operating system.
Microsoft may not be "responsible", but they give support for certain features, e.g. that for a certain period of time kernel features are not touched, so that your application will run in that time frame. Also, they patch the system, also in a improvable way, on a regular basis. Red Hat or SuSE patch they distribution only in the range of 10% of known bugs. They tend to fix some bugs in the next distribution round. But even then only about 30% of the bugs are fixed. I know this from speaking with people who work there. Also, you have the "feature" of Linux to break combatibility with new kernel releases.
The article states
"There are different reasons why people advocate open source. One reason for enterprise is, 'You have the source code; if it doesn't work, you can fix it.' But the fact is, if I'm an enterprise, I don't want to fix it. I want somebody else to fix it," Goldman said.
That's a main concern for big customers, they want a support contract, which they can rely on.
The article states further
"The system vendors -- IBM, HP (NYSE: HWP), Compaq (NYSE: CPQ), Dell (Nasdaq: DELL) -- who sell Linux all support it. They resolve Level One and Level Two problems. If problems come up that they can't solve for the customer, they turn to Red Hat (Nasdaq: RHAT)," Claybrook explained.
This is true to some extent. The big vendors support Linux, but since they are neither the owner nor the main developer, they can't do anything in the end. If, say, there is a critical kernel bug, they will turn to the open source community to make it known, but they will never be responsible to fix it. Maybe the bug will never be fixed, so the customer is left on his own. At least this is what the support contracts of IBM and HP say . The support contracts say, that there is support for installation, configuration and _troubleshooting_, NOT for fixing the problem. That is the part of the community.
Why should someone ever do something as boring as installing Linux on the new iMac. I assume, if you are into Linux, it is for the stability of a Unix system, and maybe because you are used to it. But you are already running Unix on an iMac, and you are also running a FREE Unix, too. So what's the point of Linux on an iMac. I doesn't run smoothly and the GUI is horrible compared to Aqua.
We had the same discussions on heise.de in Germany many times. It's always like this. There are some whiners who insist on "fair use", saying it should be common sense, that no ISP can provide all its users being heavy users. Then there are others who say they know the fact, that all the heavy users are subsidized by the light users. The ISPs should get their calculations right and if it comes out to what Rogers Cable is planning, so be it. BTW, from what I see posted here, Germany seems to be quite cheap for flat rate DSL - it's just USD25-30/month for 768KB/s down and 128 up, "unlimited" :-)
I usually use Red Hat or SuSE Distros with their nice graphical installers, but for some old HP PA-RISC machines I have to rely on the Debian distribution, because Debian is the only one to include the PA-RISC architecture. While the installer is far from good-looking, it does the job. It's like YaST1 from SuSE or the FreeBSD installer. Do we really need a penguin looking at us while installing Linux?
I thought about this too. Why should I run my games under Linux, while Windows is running them faster and with better graphics quality. This has of course nothing to do with money, it's all about ideology.
:-)
If _everyone_ would use the _same_ version of Windows we wouldn't complain about compatibility issues
pay and be anonymous at the same time? just tell me how.
yeah, pay cash and wear a mask.
From a more philosophical point of view... On a meta level you always have to rely on binary logic. In the case of a base 3 logic you have to decide which one of the three possibilities is actually happen. If you go with the states 0,1,2 or -1,0,1, you or the device have to "know" whether it is 0 or 1 or 2 (resp. -1 or 0 or 1). The "decision" which state is the current one is not a decision based on ternary logic, but based on binary logic. You can say, e.g. at the moment we have state 1 (true) and not 0 or 2 (which are both false). The decision isn't made on ternary logic itself, but falls back to binary. Ternary logic is NOT self-reflexive, because you can't apply ternary logic to itself. If you do you will end in an endless regress, so it is unlikely to have an advantage after all.