Yep, looks like he invented the term "hybrid source model" specifically to denigrate the GPL. The GPL is viral in the sense of a limitied life form. A GPL'd program will live on as long as somebody, anybody, has an interest in keeping it alive. If several people have different incompatible ideas of what it should be, it will multiply (divide?) like the Hydra.
Qmail is definitely purebred, and licensed to stay that way.
In general, seems like computer systems are extremely inbred. This from about 30-40 years ago and it can have only gotten more so. Seems like some cross-breeding is desirable, maybe even essential to long-term survival.
"Kenneth Brown is president of the Alexis de Tocqueville Institution... and is accepted at fine restaurants and hotels around the world." See subject.
"However, building a product that starts with the accomplishment of others and announcing it as completely your own work product, is not invention, nor is it innovation." Sounds like Microsoft.
"Another problem with Tanenbaum's logic is that he only presents examples of people that were Unix licensees, had Unix source code, or who were exceptionally familiar with software development. He cannot provide one example reasonably comparable to the Torvalds case." This actually makes it all the more likely that Linus wrote it himself. If he had the programming experience and familiarity with Unix sources, what Linus accomplished in one year would have taken him about three years. The fastest way to comprehend something is to build it from scratch, Ever had a maintenance project where the best and easiest way was to redo it from scratch. How much of the original wound up in the redo? Young, talented, energetic, driven. Reasonable outcome. I doubt that it can be repeated on demand, but all it takes are a few key insights and fortuitous choices early in the game.
"Unix is one of the greatest achievements in the history of computer science." By being onery enough to have outlived its betters. No disrespect to its creators, but Unix (and C) stink to high heaven. Unix is a poor man's Multics and C is a poor man's ALGOL68. C is compilable and Unix is writable and useable. The quote something like "Those who do not learn Unix are doomed to reinvent it. Badly." may well be accurate in that correctness drags along a sufficient overabundance of complexity to make it unatainable. In that case, Brown's rant against "hybrid source" is totally wrong.
"The hybrid source model negatively impacts the intellectual property model for all software, and inevitably the entire IT economy." The same way potable water from the city mains impacts the viability of bottled water. Seems like RedHat isn't doing too badly while the exact same software is available from White-Box? and some others. My understanding is that Linux itself is GPL, hardly fitting any reasonable definition of "hybrid source model". I would imagine that installations use some mixture of BSD, GPL and proprietary software. Solaris with GNU utilities is "hybrid source model"?
Researched? A few leading questions answered quite differently than the direction lead constitutes research substantiating the original direction? Methinks the National Enquirer has higher standards!
And what on earth does he mean by "hybrid source code model"?
What is significant is that Microsoft is buying ad space in a pro-Linux anti-Microsoft medium.
If Microsoft wants to pay for Slashdot publishing the latest Microsoft Worm alerts, it's fine by me. (That is of course my personal opinion, not necessarily related to that of anyone else on Slashdot;)
You do get a bit of a test of journalistic integrity. Seems like newspapers have dealt with similar situations.
The NT security model is that rights are granted to users, not programs.
Which ensures that if I have access to anything with one program, I also have the same access with any other program.
Right-Click IE, no "RunAs" (This is NT4)
If I can communicate with something running with root access and get it to do my bidding, I've effectively got root access regardless of what user rights I supposedly have.
If you run as root, your web browser and email program can do whatever it wants.
I run as root and I assure you that my web browser and email program most certainly do NOT do whatever they want.
Grandparent's point is that even if I choose to run as root (unpatched NT4 no less), there is no reason I should forfeit all my rights to some program just because I choose to run it.
Brown's argument is that no single person could have written a Unix clone in six months.
Brown wouldn't be able to. But that's hardly the same as no one can. It's been done before and will most likely be done again. It's not that big a deal. I've written an OS (not Unix, early OS/360) and I'm nowhere near the league of these big guns.
If I understand it correctly, it is close to impossible to make a language that is not turing complete. Some means of iteration and decision logic is required. Just think, you can compute any computable function in COBOL!
What you're seeing is that it takes extremely advanced programming to catch up with the way beginners are thinking.
The out-of-order evaluation is the primary advatage of spreadsheets. In a way it's cheating because you get the advantages of programming without having to explicity state the execution order, aka programming.
Well that depends on what the exception handler did with the exception doesn't it?
Touche. Maybe going off-topic, but this is starting to become fun.
Life should be better with exception handlers, but what can/should/will they do? I suspect in the majority of systems, the main-line (without exceptions) is tested/debugged pretty well, while the exception-lines with about ten times the complexity get about one tenth the attention.
so spending weeks learning HTML when you want to learn how to program would be completely down the drain
Just like learning the alphabet would be a waste if you wanted to learn to write.
You have to know how to start, how to stop, and what you can put inside what. Each different language has its own peculiar rules, with an emphasis on peculiar.
I doubt that HTML itself is Turing-complete, but it can easily be made so by running it through PHP or JSP.
What it carefully avoids is the discussion of trolls and AC posts.
A lot of what makes Slashdot Slashdot is how Slashdot has handled the problems of trolls and AC posts in real time on a live system. I'm not sure that there even is a "solution", but so far at least, Slashdot has seemed to be able to cope with it. Not bad for a bunch of amateurs.
If you get everything from a single source, it's possible for that single source to collude with itself, hidden well enough that even having all the sources is insufficient to detect the collusion.
The reason it is a classic paper is that no one has been able to pull off the same stunt again. It requires effective control of both parts very early on in the evolution. To repeat it now, you'd need collusion beteen/among Theo, RedHat, Debian and probably IBM. Never gonna happen. Microsoft is a different matter.
Anyone can get ripped off at any level. The good thing is that someone at least *noticed* this one and is now beating restitution out of the victims. [Emphasis added]
Typo? What worries me is that it's probably right as writ.
Then again, it does seem appropriate. The first thing I have to kill when I bother to install MS Office is the stupid wizards and assistants. Now imagine yourself considering those same wizards and assistants as smart. And yourself as smart for using them. Look at what the system wants you to believe.
"I'd rather have a bottle in front of me rather than a frontal lobotomy." Right!
Anthromorphizing things. Water wants to run downhill. Without understanding the effects of gravity, how else would you put it?
How well things work out depends on the intrinsics, but possibly more importantly on the differences between expectations and reality. If something is good, but not quite as good as expected, you get blindsided and things do not work out as expected. Unix is immune to worms and viruses? The Unix Honor virus is an immediate counter-example. All you've got to do is make it credible.
Solaris? Sorry, outa my league. I'm only marginally competent with Linux, learning by doing, on the cheap.
Windows is easier to patch if you just want to go through the motions and do not care about the results. But in that case, why bother patching? What do youdo if a patch ges wrong?
BSD/Linux probably has the higher percentage of computer-literate users. This is due to the OS. Take two equivalent groups, one on Linux, one on Windows. With no difference in external factors, the Linux users will become more competent and the Windows users will become less competent.
Slashdot Mirror
Yep, looks like he invented the term "hybrid source model" specifically to denigrate the GPL. The GPL is viral in the sense of a limitied life form. A GPL'd program will live on as long as somebody, anybody, has an interest in keeping it alive. If several people have different incompatible ideas of what it should be, it will multiply (divide?) like the Hydra.
Qmail is definitely purebred, and licensed to stay that way.
In general, seems like computer systems are extremely inbred. This from about 30-40 years ago and it can have only gotten more so. Seems like some cross-breeding is desirable, maybe even essential to long-term survival.
Google Search Appliance (maybe closer to an orangish yellow)
Ken Brown asserts that Tanenbaum had the Lions notes (illegal Unix copy)
Highly unlikely.
With the Lions book, read, understand, write UNIX. Many years.
Without the Lions book, understand, write LINUX. One year.
at some point, many, many years ago, say 40, a hand-written kernel was booted that was written in assembler.
Correct.
Not once.
Many times.
Most of 'em lost in the dust of history.
You haven't lived until you've keypunched EBCDIC machine language into an IPL deck.
"Kenneth Brown is president of the Alexis de Tocqueville Institution ... and is accepted at fine restaurants and hotels around the world."
See subject.
"However, building a product that starts with the accomplishment of others and announcing it as completely your own work product, is not invention, nor is it innovation."
Sounds like Microsoft.
"Another problem with Tanenbaum's logic is that he only presents examples of people that were Unix licensees, had Unix source code, or who were exceptionally familiar with software development. He cannot provide one example reasonably comparable to the Torvalds case."
This actually makes it all the more likely that Linus wrote it himself. If he had the programming experience and familiarity with Unix sources, what Linus accomplished in one year would have taken him about three years. The fastest way to comprehend something is to build it from scratch, Ever had a maintenance project where the best and easiest way was to redo it from scratch. How much of the original wound up in the redo?
Young, talented, energetic, driven. Reasonable outcome. I doubt that it can be repeated on demand, but all it takes are a few key insights and fortuitous choices early in the game.
"Unix is one of the greatest achievements in the history of computer science."
By being onery enough to have outlived its betters. No disrespect to its creators, but Unix (and C) stink to high heaven. Unix is a poor man's Multics and C is a poor man's ALGOL68. C is compilable and Unix is writable and useable. The quote something like "Those who do not learn Unix are doomed to reinvent it. Badly." may well be accurate in that correctness drags along a sufficient overabundance of complexity to make it unatainable. In that case, Brown's rant against "hybrid source" is totally wrong.
"The hybrid source model negatively impacts the intellectual property model for all software, and inevitably the entire IT economy."
The same way potable water from the city mains impacts the viability of bottled water. Seems like RedHat isn't doing too badly while the exact same software is available from White-Box? and some others. My understanding is that Linux itself is GPL, hardly fitting any reasonable definition of "hybrid source model". I would imagine that installations use some mixture of BSD, GPL and proprietary software. Solaris with GNU utilities is "hybrid source model"?
with a researched presentation
Researched?
A few leading questions answered quite differently than the direction lead constitutes research substantiating the original direction?
Methinks the National Enquirer has higher standards!
And what on earth does he mean by "hybrid source code model"?
What is significant is that Microsoft is buying ad space in a pro-Linux anti-Microsoft medium.
If Microsoft wants to pay for Slashdot publishing the latest Microsoft Worm alerts, it's fine by me. (That is of course my personal opinion, not necessarily related to that of anyone else on Slashdot;)
You do get a bit of a test of journalistic integrity. Seems like newspapers have dealt with similar situations.
The NT security model is that rights are granted to users, not programs.
Which ensures that if I have access to anything with one program, I also have the same access with any other program.
Right-Click IE, no "RunAs" (This is NT4)
If I can communicate with something running with root access and get it to do my bidding, I've effectively got root access regardless of what user rights I supposedly have.
If you run as root, your web browser and email program can do whatever it wants.
I run as root and I assure you that my web browser and email program most certainly do NOT do whatever they want.
Grandparent's point is that even if I choose to run as root (unpatched NT4 no less), there is no reason I should forfeit all my rights to some program just because I choose to run it.
Brown's argument is that no single person could have written a Unix clone in six months.
Brown wouldn't be able to. But that's hardly the same as no one can.
It's been done before and will most likely be done again. It's not that big a deal. I've written an OS (not Unix, early OS/360) and I'm nowhere near the league of these big guns.
To fix it, someone had to go around to all those empty catch blocks (all 10,000 of them) and have the current thread re-interrupt itself in each one.
LOL. There has to be something wrong where one change has to be made in 10,000 places.
If I understand it correctly, it is close to impossible to make a language that is not turing complete. Some means of iteration and decision logic is required. Just think, you can compute any computable function in COBOL!
What you're seeing is that it takes extremely advanced programming to catch up with the way beginners are thinking.
The out-of-order evaluation is the primary advatage of spreadsheets.
In a way it's cheating because you get the advantages of programming without having to explicity state the execution order, aka programming.
Well that depends on what the exception handler did with the exception doesn't it?
Touche.
Maybe going off-topic, but this is starting to become fun.
Life should be better with exception handlers, but what can/should/will they do?
I suspect in the majority of systems, the main-line (without exceptions) is tested/debugged pretty well, while the exception-lines with about ten times the complexity get about one tenth the attention.
Maybe a good example.
The code is functional and will work as intended in both Java and C++, despite fundamental differences in the semantics of new.
Right.
Remember, "The Emperor Has No Clothes" was broken, not by the smart elite, but by a dumb street urchin.
so spending weeks learning HTML when you want to learn how to program would be completely down the drain
Just like learning the alphabet would be a waste if you wanted to learn to write.
You have to know how to start, how to stop, and what you can put inside what. Each different language has its own peculiar rules, with an emphasis on peculiar.
I doubt that HTML itself is Turing-complete, but it can easily be made so by running it through PHP or JSP.
What it carefully avoids is the discussion of trolls and AC posts.
A lot of what makes Slashdot Slashdot is how Slashdot has handled the problems of trolls and AC posts in real time on a live system. I'm not sure that there even is a "solution", but so far at least, Slashdot has seemed to be able to cope with it. Not bad for a bunch of amateurs.
Have you read the Ken Thompson's classic paper ...
Yes. ACM Turing Award paper, IIRC.
If you get everything from a single source, it's possible for that single source to collude with itself, hidden well enough that even having all the sources is insufficient to detect the collusion.
The reason it is a classic paper is that no one has been able to pull off the same stunt again. It requires effective control of both parts very early on in the evolution. To repeat it now, you'd need collusion beteen/among Theo, RedHat, Debian and probably IBM. Never gonna happen. Microsoft is a different matter.
Anyone can get ripped off at any level. The good thing is that someone at least *noticed* this one and is now beating restitution out of the victims. [Emphasis added]
Typo?
What worries me is that it's probably right as writ.
No, that would not be me, because I don't live in the US.
Wrong. You live in the same world.
The effects are small and suble, but multifarious. You pay.
Then again, it does seem appropriate. The first thing I have to kill when I bother to install MS Office is the stupid wizards and assistants.
Now imagine yourself considering those same wizards and assistants as smart. And yourself as smart for using them. Look at what the system wants you to believe.
"I'd rather have a bottle in front of me rather than a frontal lobotomy."
Right!
Anthromorphizing things. Water wants to run downhill. Without understanding the effects of gravity, how else would you put it?
How well things work out depends on the intrinsics, but possibly more importantly on the differences between expectations and reality. If something is good, but not quite as good as expected, you get blindsided and things do not work out as expected. Unix is immune to worms and viruses? The Unix Honor virus is an immediate counter-example. All you've got to do is make it credible.
64-bit Windows virus.
Concept-only. Posted.
Found in wild. Posted.
Same virus.
Different stories.
Different significance.
That would imply that Windows performs automagical frontal lobatomies on users.
How else would you explain the Microsoft Office ad which has its users falling all over each other?
Solaris? Sorry, outa my league.
I'm only marginally competent with Linux, learning by doing, on the cheap.
Windows is easier to patch if you just want to go through the motions and do not care about the results. But in that case, why bother patching? What do youdo if a patch ges wrong?
BSD/Linux probably has the higher percentage of computer-literate users. This is due to the OS. Take two equivalent groups, one on Linux, one on Windows. With no difference in external factors, the Linux users will become more competent and the Windows users will become less competent.