This is a very very clever plot from GPL friends. It is a preemtive strike against possible future hostile legal fire.
What is happening is that SCO is effectively going to court, trying to attack every single possible legal problem of Linux and the GPL. It does so with enourmous stupidicy and incompetence, to such a degree that the judge will have no choice but to rule against them on every single claim. Then it will escalate all the way up to the supreme court, which will also rule against them on every single claim.
The result? Every possible legal weakness of Linux and the GPL has been strengthen by legal precedence, which is very hard to turn around.
Whoever it is who manages all this deserves every dollar that falls off as a biproduct.
About the tar thingie; I would guess to keep compatibility with POSIX. And the reason POSIX doesn't have it is pretty obvious, when you think of when and why tar was created.
I think one of the greatest things that could happen would be if POSIX was superseded by a new standard that wasn't too rooted in the PDP11. Then we could finally get a generation of MODERN un*xes.
I can't honestly see how including the "Windows 2000 Server" logo is at least not trademark violation. And although I strongly dislike Microsoft, both their products and politics, I would support them in that. Using that logo is just plain stupid and immoral.
This is because the open source community isn't one huge project to develop thingie X, as specified from the customer. It's about an enourmous amount of independent developers viewing the entire open source codebase, and evaluating "Is there anything in here I think sucks, that I could make better?", and then they do an attempt at doing that. Works basically in the same way as evolution.
One of the reasons why this is the best approach is that all developers have different visions of how things should be and what are the real problems with a project. The commercial way of solving that is discussing it, and then let some project leaders pick a compromise that most developers would silently disagree with, but, with slightly lowered motivation, work on anyway. The open source way is that people do what they want, and then afterwards the world can see who was right.
The result of this process is not the maximum code lines possibly produced by millions of developers. It is the most stable and at a certain level "perfect" software possible. This is best illustrated by a Djikstra quote that I don't remember the exact wording of, but it goes something like "The project isn't finnished when it has all the code lines required. It is finnished when it has nothing but the code lines requires". That is quality != quantity.
If open source has one problem, however, it is that the continous enhancement process works like the hillclimbing algorithm. It enhances itself, but revolutions that imply changing lots of projects concurrently to make long term quantum leaps without implying short term enhancements are just not going to happen.
A metre is the length travelled by light in vacuum in 1/299 792 458 of a second. This is approximately the side length of a cube containing one ton (1000 kg) of water. Also, a litre is defined as 0.1^3 metres (same as one kg of water).
The metre,kg and litre are part of the metric system, which is the standard in all civilized and most uncivilized countries.
I've heard tales of a big country across the Atlantic, where they still measure stuff in units like feet, stones, pounds, gallons and grandfathers-length-when-he-died, but I have a hope that we someday will manage to teach them the metric standard, how to read and how to elect sane presidents.
SML is the language on which OCaml is based. OCaml is really just SML with different (read: worse) syntax, some small differences, and then OO support.
SML will probably be easier to start off on, is better supported and better documented.
Once you've got SML down, OCaml will be easy.
I started with SML, but thought it was impractical. I think it was limitations with the library, but when you are generally frustrated, it's pretty hard to identify the exact cause. Anyway, it turned me off ML for a while. Until I discovered O'Caml. Really opened my eyes.
So my advice is to jump directly at O'Caml. SML might have a slightly more pedagogic or clean syntax, but an environment that makes you feel powerfull is more motivating and therefore more important in the start.
C was never a good language. It is a working language who was heavily designed to be easy to compile on a PDP10. As we don't use PDP10's anymore, there's no reason to stick with all the sacrifices made to accomplish easy compilation on that platform.
Actually, the fact that C is so low level actually makes it harder to write good compilers for it. It is much harder to identify the scopes and side effects of optimizations in low level languages.
I'm not sure why it still is popular. Probably a mix between "I don't need no stinking second language", "look at me, I can write in C, I'm l33t" and a lack of coordination between all those attempting to move to better languages.
Haskell:
Nice language, but I've never found any implementation that is complete enough to use for real programs.
O'Caml:
Nice language. Nice implementation. Nice library. Currently what I'm using.
Scheme:
Very small lisp. Often used as an extension language (for example in Gimp). It's a common discussion if this is an academic toy language or if it's fit for real programming. I'm not quite sure. Anyway, it's easy to learn (the standard is only a few pages), and there's a nice book for it (The Little Schemer (don't let the cartoons and food receipies fool you, this is one scary book)).
Common Lisp:
Is paradigm adaptive. That is, if you want to write functional, you can. If you want to write imperative, that's nice too. It's my favourite language, but I've had problems finding an implementation that has ok thread support and costs more than a car (no pun intended:) for a version that is not a strictly non-commercial trial version. That's why I ended up using O'Caml instead.
Already at 'as a professional' I knew you had to be a looser. And when you call NT 'perfectly stable', it's just obvious that you have to be a "web designer" or "VB programmer" or something other in the 'air,food and money wastement' area of work.
Of course you could be paid by Microsoft too. Logs, anyone?
Slashdot Mirror
Why can't everyone just agree that BASIC is dead, should stay dead and preferably staked through the heart?
I personally can't see any use at all for BASIC.
This is a very very clever plot from GPL friends. It is a preemtive strike against possible future hostile legal fire.
What is happening is that SCO is effectively going to court, trying to attack every single possible legal problem of Linux and the GPL. It does so with enourmous stupidicy and incompetence, to such a degree that the judge will have no choice but to rule against them on every single claim. Then it will escalate all the way up to the supreme court, which will also rule against them on every single claim.
The result? Every possible legal weakness of Linux and the GPL has been strengthen by legal precedence, which is very hard to turn around.
Whoever it is who manages all this deserves every dollar that falls off as a biproduct.
About the tar thingie; I would guess to keep compatibility with POSIX. And the reason POSIX doesn't have it is pretty obvious, when you think of when and why tar was created.
I think one of the greatest things that could happen would be if POSIX was superseded by a new standard that wasn't too rooted in the PDP11.
Then we could finally get a generation of MODERN un*xes.
Damn, I'm too quick. I didn't notice before posting that that screenshot was of a WMWare session actually running Windows 2000.
IMHO a pretty bad example.
I can't honestly see how including the "Windows 2000 Server" logo is at least not trademark violation. And although I strongly dislike Microsoft, both their products and politics, I would support them in that. Using that logo is just plain stupid and immoral.
This is because the open source community isn't one huge project to develop thingie X, as specified from the customer. It's about an enourmous amount of independent developers viewing the entire open source codebase, and evaluating "Is there anything in here I think sucks, that I could make better?", and then they do an attempt at doing that. Works basically in the same way as evolution.
One of the reasons why this is the best approach is that all developers have different visions of how things should be and what are the real problems with a project. The commercial way of solving that is discussing it, and then let some project leaders pick a compromise that most developers would silently disagree with, but, with slightly lowered motivation, work on anyway. The open source way is that people do what they want, and then afterwards the world can see who was right.
The result of this process is not the maximum code lines possibly produced by millions of developers. It is the most stable and at a certain level "perfect" software possible. This is best illustrated by a Djikstra quote that I don't remember the exact wording of, but it goes something like "The project isn't finnished when it has all the code lines required. It is finnished when it has nothing but the code lines requires". That is quality != quantity.
If open source has one problem, however, it is that the continous enhancement process works like the hillclimbing algorithm. It enhances itself, but revolutions that imply changing lots of projects concurrently to make long term quantum leaps without implying short term enhancements are just not going to happen.
Does someone have a queue of protocols/formats suited for reverse engineering?
A meter is a tool for measuring something.
A metre is the length travelled by light in vacuum in 1/299 792 458 of a second. This is approximately the side length of a cube containing one ton (1000 kg) of water. Also, a litre is defined as 0.1^3 metres (same as one kg of water).
The metre,kg and litre are part of the metric system, which is the standard in all civilized and most uncivilized countries.
I've heard tales of a big country across the Atlantic, where they still measure stuff in units like feet, stones, pounds, gallons and grandfathers-length-when-he-died, but I have a hope that we someday will manage to teach them the metric standard, how to read and how to elect sane presidents.
I started with SML, but thought it was impractical. I think it was limitations with the library, but when you are generally frustrated, it's pretty hard to identify the exact cause. Anyway, it turned me off ML for a while. Until I discovered O'Caml. Really opened my eyes.
So my advice is to jump directly at O'Caml. SML might have a slightly more pedagogic or clean syntax, but an environment that makes you feel powerfull is more motivating and therefore more important in the start.
C was never a good language. It is a working language who was heavily designed to be easy to compile on a PDP10. As we don't use PDP10's anymore, there's no reason to stick with all the sacrifices made to accomplish easy compilation on that platform.
Actually, the fact that C is so low level actually makes it harder to write good compilers for it. It is much harder to identify the scopes and side effects of optimizations in low level languages.
I'm not sure why it still is popular. Probably a mix between "I don't need no stinking second language",
"look at me, I can write in C, I'm l33t" and a lack of coordination between all those attempting to move to better languages.
There's several possibilities
:) for a version that is not a strictly non-commercial trial version. That's why I ended up using O'Caml instead.
Haskell:
Nice language, but I've never found any implementation that is complete enough to use for real programs.
O'Caml:
Nice language. Nice implementation. Nice library. Currently what I'm using.
Scheme:
Very small lisp. Often used as an extension language (for example in Gimp). It's a common discussion if this is an academic toy language or if it's fit for real programming. I'm not quite sure. Anyway, it's easy to learn (the standard is only a few pages), and there's a nice book for it (The Little Schemer (don't let the cartoons and food receipies fool you, this is one scary book)).
Common Lisp:
Is paradigm adaptive. That is, if you want to write functional, you can. If you want to write imperative, that's nice too. It's my favourite language, but I've had problems finding an implementation that has ok thread support and costs more than a car (no pun intended
Already at 'as a professional' I knew you had to
be a looser. And when you call NT 'perfectly stable', it's just obvious that you have to be
a "web designer" or "VB programmer" or something
other in the 'air,food and money wastement' area
of work.
Of course you could be paid by Microsoft too.
Logs, anyone?
I guess they have environment friendly envelopes :)