Nah, Linux will remain in several niches, all of them relatively virus/trojan/worm/whatever-free.
It gets a bit hilarious when the *BSDs can run Linux binaries and the various distros have trouble running each others.
What's the problem with where RedHat puts Apache? I find it a bit convenient to be able to run two default Apache installations on the same computer at the same time.
Guessing, but if you actually did secure a windows box so that people couldn't write over anything they pleased, what's left for them to write to? Hidden files in \RECYCLER?
I can still run my ancient DOS and Win3.1 programs (for the most part) under Windows 2000; tell me where they've changed the API here? OK. Under DOS, when the CapsLock light is on on the keyboard, pressing an alphabetic key produces an upper-case letter. When the CapsLock light is off on the keyboard, pressing an alphabetic key produces a lower-case letter. Windows 2000 inconsistently reverses that behavior.
Yes, the way people think.
And yes, functional is more productive than procedureal. Composition of functions is associative, so you can sometimes see shortcuts.
Functional concentrates on what is to be done.
Procedural concentrates on the steps required to do it.
Functional is f(g(h(foo(bar))).
Procedural is take bar, then foo it, then h, then g, then f it.
Process file in procedural is CLOSE-FILE( Do-STUFF-in-Middle( Open-File)))
Procedural is prog-style LISP.
Functional is LISP without any variables.
You can do functional in procedural languages (with the possible exception of bare Assembler)
You can do procedural in probably all functional languages.
There are other stunts possible like data-binding.
Functional is nested 30-deep in function calls (ie shallow)
Procedural in the extreme is a Finite State Automaton loaded with gotos.
Program counter is very meaningful in Procedural.
Program counter can be almost all noise in Functional.
It is ludicrous, but the emulator will tend to derandomize the behavior of the apps, making them run more reliably.
Thanks, I have and do. Microsoft Windows 2000 is about what Microsoft Windows 1995 should have been. When it can't do something as straightforward as keeping CapsLock State in sync with keyboard status lights, it's a long way from reliable or stable.
Not just the current generation. FORTRAN has been around and available for more than 40 years that I am personally aware of. Fun thing on GE Timesharing was to use FORTRASH (only the first 3 letters were significant).
The assumption is that everybody knows FORTRAN, at least to the point of being able to read and write Formulas. Programming ability is a completely different subject, although FORTRAN has been used as a sort of universal machine language.
Integers, Floats (called REAL), and Complex numbers are the (very!) basic data types. The results of computations are very predictable. For fun, try to establish bounds on the numeric errors that a C or Pascal compiler is allowed to commit. If you have something like Result = (Large - Large) / (Large - Large), the various algebraicly equivalent forms can yield wildly different results. Experience and instinct dictate spelling it out in painstaking detail if the results are critical.
wanted to integrate Linux into our server pool. What? Why? How? Think about it. You might use Linux servers, or have a pool of Linux servers, but integrating Linux into a pool of (MVS|VMS|MTS|*BSD|NT3.5) servers in nonsense.
kernel-level programming in VB for 8 years COBOL would be a bit more plausible.
I took it upon myself to configure the system from scratch By re-writing the init scripts?
Probably possible to make system thrash by sufficiently bad misconfiguration.
optimised version of gcc 3.1 What on earth is that? Optimised for what?
"server" based operating system How can you base an operating system on a server?
kernel panics caused by Bind and Apache crashing Total misunderstanding. You can possibly get kernel panics from Bind or Apache not crashing, if you work hard enough at it.
Linux kernel itself lacks any support for any type of journaled filesystem, memory protection, SMP support, etc ext3 now standard for RedHat 7.2
I doubt that Linux has ever NOT had separation of address space for different running processes.
RedHat 6.2 had SMP support, so that has been there for a while and getting better.
Enough buzz words so a fast reading looks like he might know something. Closer analysis reveals that it has negative information content.
That's the term.
How many characters can you store in 64k of memory?
Answer. You can store somewhat over 65k characters in 64k (binary kilobytes) of memory.
The convincer (for me;)
How do you express 640k in Megs. 0.640M doesn't really work.
And confusing bits with Bytes is an 800% error (or 87.5%).
It's pretty obvious what 64M means because it's a very round (binary) number. 67.108864M is rather awkward. Use 2 or 3 decimal places and try to add straight.;-)
Patches to fundemental kernal services can have far reaching side-effects... all existing security testing is out-the window and you start from scratch Which raises the question: Are you better off not installing the patch, even with poor details as to the vulnerability? It closes one known hole, maybe and maybe opens up a few others. Too many maybes. If this would be last one for a reasonable length of time (6 months to 2 years), I'd have a different take on it.
I don't think it was criticism. More like Diogenes looking for an honest man. Try to find an honest politician. Try to find one who knows what he's talking about. And they do mean well. Catch a load of CSPAN sometime.
I am not optimistic, however. Those who get it (e.g., you) are pathetically apolitical. You're proud of your apathy. You're disgusted with people who try to persuade politicians. So am I. But while you do nothing, the future of creativity and innovation is sold in DC - typically to the highest, and most disgusting bidder.
But Slashdot is not a political party.
Right. If we were, we would no longer get it.
Great post, BTW.
He just did. You don't think that's the only hole do you?
Unless the right circumstances come up, it's both very stale and very offtopic. Too much like crying "Wolf!". Nobody listens. A basic problem with any access restriction device is that you can't afford to keep out the owner. There's also a problem with janitors (unless you don't want anything cleaned;)
Since there is not an exploit available and this hole is patched (in some of the systems), they go looking. Maybe they find this hole, maybe another one of its neighbors. Whatever the attack is, it will be harder to find and harder to trace. Maybe they did the "right thing", but I do not feel at all safer.
Re:People don't care about security flaws because
on
WinXP Security Flaw
·
· Score: 1
Moved a CD-ROM burner from a Window ME system to a Windows 2000 system.
Made a set of Windows 2000 Professional coasters.
Re:So much for Microsoft's "thorough code review"
on
WinXP Security Flaw
·
· Score: 1
They did the reviewin'. They just didn't get around to the findin' and the fixin'.
What did you expect? That they would actually fix the bugs?
When and what did you last pay for the air you breathe?
There is generally a relationship, but nonlinear and quite often anomalous.
"The best things in life are free."
There is a vast difference between priceless and valueless.
There are those who know the price of everything and the value of nothing.
Then how the hell can you compare a several-years-ago Linux default install with a just-released Windows XP default install? Actually it's pretty easy. Install old version of Linux. Kill off unneeded services and update the few left that matter. Most of the holes are in things you should not be running anyway. System now should be in pretty good shape for next 6-18 months.
Install a just-released Windows XP. Try to find the required updates. All of them. Good for 2 weeks maybe (if you like a false sense of security;)
Slashdot Mirror
The advantage of Open Source.
There are a few paranoids out there than can and will start screaming their heads off if something funny is going on.
Nope, the moderator is right!
It is redundant. Not wrong, redundant.
Microsoft keeps doing stuff to make the world safe for viruses/worms/etc.
>>More tech-savvy than we give him credit for I guess? :)
;)
Understatement?
I wonder.
...
I'm sure you don't, but when enough people start to wonder,
Nah, Linux will remain in several niches, all of them relatively virus/trojan/worm/whatever-free.
It gets a bit hilarious when the *BSDs can run Linux binaries and the various distros have trouble running each others.
What's the problem with where RedHat puts Apache? I find it a bit convenient to be able to run two default Apache installations on the same computer at the same time.
Guessing, but if you actually did secure a windows box so that people couldn't write over anything they pleased, what's left for them to write to? Hidden files in \RECYCLER?
I can still run my ancient DOS and Win3.1 programs (for the most part) under Windows 2000; tell me where they've changed the API here?
OK. Under DOS, when the CapsLock light is on on the keyboard, pressing an alphabetic key produces an upper-case letter. When the CapsLock light is off on the keyboard, pressing an alphabetic key produces a lower-case letter. Windows 2000 inconsistently reverses that behavior.
Yes, the way people think.
And yes, functional is more productive than procedureal. Composition of functions is associative, so you can sometimes see shortcuts.
Functional concentrates on what is to be done.
Procedural concentrates on the steps required to do it.
Functional is f(g(h(foo(bar))).
Procedural is take bar, then foo it, then h, then g, then f it.
Process file in procedural is CLOSE-FILE( Do-STUFF-in-Middle( Open-File)))
Procedural is prog-style LISP.
Functional is LISP without any variables.
You can do functional in procedural languages (with the possible exception of bare Assembler)
You can do procedural in probably all functional languages.
There are other stunts possible like data-binding.
Functional is nested 30-deep in function calls (ie shallow)
Procedural in the extreme is a Finite State Automaton loaded with gotos.
Program counter is very meaningful in Procedural.
Program counter can be almost all noise in Functional.
It is ludicrous, but the emulator will tend to derandomize the behavior of the apps, making them run more reliably.
Thanks, I have and do. Microsoft Windows 2000 is about what Microsoft Windows 1995 should have been. When it can't do something as straightforward as keeping CapsLock State in sync with keyboard status lights, it's a long way from reliable or stable.
Not just the current generation. FORTRAN has been around and available for more than 40 years that I am personally aware of. Fun thing on GE Timesharing was to use FORTRASH (only the first 3 letters were significant).
The assumption is that everybody knows FORTRAN, at least to the point of being able to read and write Formulas. Programming ability is a completely different subject, although FORTRAN has been used as a sort of universal machine language.
Integers, Floats (called REAL), and Complex numbers are the (very!) basic data types. The results of computations are very predictable. For fun, try to establish bounds on the numeric errors that a C or Pascal compiler is allowed to commit. If you have something like Result = (Large - Large) / (Large - Large), the various algebraicly equivalent forms can yield wildly different results. Experience and instinct dictate spelling it out in painstaking detail if the results are critical.
Obviously bogus.
wanted to integrate Linux into our server pool.
What? Why? How? Think about it. You might use Linux servers, or have a pool of Linux servers, but integrating Linux into a pool of (MVS|VMS|MTS|*BSD|NT3.5) servers in nonsense.
kernel-level programming in VB for 8 years
COBOL would be a bit more plausible.
I took it upon myself to configure the system from scratch
By re-writing the init scripts?
Probably possible to make system thrash by sufficiently bad misconfiguration.
optimised version of gcc 3.1
What on earth is that? Optimised for what?
"server" based operating system
How can you base an operating system on a server?
kernel panics caused by Bind and Apache crashing
Total misunderstanding. You can possibly get kernel panics from Bind or Apache not crashing, if you work hard enough at it.
Linux kernel itself lacks any support for any type of journaled filesystem, memory protection, SMP support, etc
ext3 now standard for RedHat 7.2
I doubt that Linux has ever NOT had separation of address space for different running processes.
RedHat 6.2 had SMP support, so that has been there for a while and getting better.
Enough buzz words so a fast reading looks like he might know something. Closer analysis reveals that it has negative information content.
Windows is getting better at an exponential rate.
Hint. The exponent is imaginary.
That's the term.
How many characters can you store in 64k of memory?
Answer. You can store somewhat over 65k characters in 64k (binary kilobytes) of memory.
The convincer (for me;)
How do you express 640k in Megs. 0.640M doesn't really work.
Other unices.
Other than Solaris, (also by Sun).
... shifted from the Win* platform to other (non-Solaris) unices.
He threw you a curve. You missed.
And confusing bits with Bytes is an 800% error (or 87.5%). ;-)
It's pretty obvious what 64M means because it's a very round (binary) number. 67.108864M is rather awkward. Use 2 or 3 decimal places and try to add straight.
Patches to fundemental kernal services can have far reaching side-effects ... all existing security testing is out-the window and you start from scratch
Which raises the question: Are you better off not installing the patch, even with poor details as to the vulnerability? It closes one known hole, maybe and maybe opens up a few others. Too many maybes. If this would be last one for a reasonable length of time (6 months to 2 years), I'd have a different take on it.
Microsoft is a symbol of innovation. It's a front. It's a sham. But it's still a symbol as long as there are enough gullible victims who believe.
I don't think it was criticism. More like Diogenes looking for an honest man. Try to find an honest politician. Try to find one who knows what he's talking about. And they do mean well. Catch a load of CSPAN sometime.
I am not optimistic, however. Those who get it (e.g., you) are pathetically apolitical. You're proud of your apathy. You're disgusted with people who try to persuade politicians. So am I. But while you do nothing, the future of creativity and innovation is sold in DC - typically to the highest, and most disgusting bidder.
But Slashdot is not a political party.
Right. If we were, we would no longer get it.
Great post, BTW.
He just did. You don't think that's the only hole do you?
Unless the right circumstances come up, it's both very stale and very offtopic. Too much like crying "Wolf!". Nobody listens. A basic problem with any access restriction device is that you can't afford to keep out the owner. There's also a problem with janitors (unless you don't want anything cleaned;)
Since there is not an exploit available and this hole is patched (in some of the systems), they go looking. Maybe they find this hole, maybe another one of its neighbors. Whatever the attack is, it will be harder to find and harder to trace. Maybe they did the "right thing", but I do not feel at all safer.
Moved a CD-ROM burner from a Window ME system to a Windows 2000 system.
Made a set of Windows 2000 Professional coasters.
They did the reviewin'. They just didn't get around to the findin' and the fixin'.
What did you expect? That they would actually fix the bugs?
Does this give a new meaning to the term "buffer overflow"?
"....u get what you pay for is the old saying.
When and what did you last pay for the air you breathe?
There is generally a relationship, but nonlinear and quite often anomalous.
"The best things in life are free."
There is a vast difference between priceless and valueless.
There are those who know the price of everything and the value of nothing.
Then how the hell can you compare a several-years-ago Linux default install with a just-released Windows XP default install?
Actually it's pretty easy. Install old version of Linux. Kill off unneeded services and update the few left that matter. Most of the holes are in things you should not be running anyway. System now should be in pretty good shape for next 6-18 months.
Install a just-released Windows XP. Try to find the required updates. All of them. Good for 2 weeks maybe (if you like a false sense of security;)