The offensive part is the platitudes. It's one big general diatribe on nothing at all. Reminds me of the preachy father from Hamlet - all words and no content.
Is his point that people need to be more serious about security? Fine. But he's utterly unqualified to be giving technical advice or making technical judgements on a field he obviously knows nothing about (use Python, Ruby, or Java and never C++? I call buzzword bullshit bingo!) other than the vague general knowledge a "technologist" would have.
It's like me saying plastic surgery has become dangerous because there are so many quacks, and they all need to sharpen their scalpels and pay attention to sterility.
This "technologist" is carrying on about bad programmers and security? Wow - I assume he's a seasoned professional with many large-scale projects under his belt?
With such trenchant insights as "Don't use C/C++"! "Don't use Windows!" "Watch out for user input"!.
Wow. How truly insightful. I'm not even going to bother pointing out the utter absurdity of claiming that using or not using C/C++ has anything to do with it, or the added security problems with using high level languages (do you trust the implementation?).
I'm just going to say I've had bloody poops with more useful information in them than this article.
Wow, someone lives in fairy land, with cute little translucently purple winged pixies flitting from nerd to nerd, granting their every drooling little fruity wish.
"Would you like World Peace? Abundant cash? And end to suffering in all forms?" "N'Glaven, No! I want Microsoft to be crushed!!!"
100 threads - maybe. 10 threads on a single CPU, nonsense. And the number of CPU's has no bearing. Threads in this situation are just a way to increase responsiveness and to ease development effort. Writing complex event loops is silly unless you're doing intense numerical computations or something else where the CPUs are going to be nearing 100% load.
A lot of development models require threads that usually just sit there, but will occasionally wake up and act. Say you have a UI that needs updating, a logfile that needs written to, a few network sockets that need listening to, etc... There's no reason not to use as many threads as you need.
The ease in programming outwieghs the nonexistent issues in performance because most of your threads aren't doing anything most of the time anyway.
Now, if you're talking about hardcore computation in multiple threads, then obviously the fewer threads the better.
Again, real Unix operating systems have had working, stable M:N thread models for a long time. Why would it be any problem for Linux?
Does this whitepaper by a (relatve) newcomer somehow invalidate 10 years of real experience by the rest of the world and by large commercial companies which make real money selling Unix operating systems? Hardly.
There's nothing more "correct" about NPTL. Real Unix operating systems have used M:N thread semantics for a long, long time.
It was hubris for the developers to even release LinuxThreads ("Oh, our context switches are so fast we don't need user threads!") and the result has been a painful last few years. Oh what joy, every thread is a process! How, umm, clever! Thank God for the new choices (especially NGPT).
You're a filthy hypocrit. I'm sure if this was some evel corporation suing a mom-n-pop company over their use of some idiotic patent you'd be crying out against the horrible inhumanity of it all.
a.) The EU is more interested in economics than racism. If you think their actions are motivated because of racism over economics you're clueless.
b.) What thesis are we supposed to invalidate? His idiotic comparison to Star Wars which is just stupid. Or his.. uhhh, what? What _IS_ his thesis, that the book doesn't impart some life-changing moral?? Good lord.
And I think it explains somewhat what the honored maitres were afraid of. At about 1/3 of the book it foreshadows some very powerful force that was sent out into the galaxy to germinate, and that force is probably what scares the shit out of everyone inthe last Dune book by papa Herbert.
Are you kidding? Historically Unix security has been absolutely horrible. Look at the number of bugs it's had. It _still_ has more flaws (when you take 'Unix' as a whole) than Windows, and Unix has had 30 years to improve!
I can agree with the FreeBSD issue, I guess... The rest I don't, but the point's moot.
I'm actually more thankful it isn't Perl. Good lord, I wish people would quit trying to develop "real" applications with that lumbering piece of merde. It's great for scripts, it's crap for real software development.
Re:Java: The Britney of Programming Languages
on
Yahoo Moving to PHP
·
· Score: 0
You've not done a lot of Java development, have you? Maybe..umm, none beyone a Hello World program or some trivial applet? I find that a lot with Java naysayers.
Your portability arguments were, I believe, thrown in for shock value? Of course Java is very portable. If you don't use JNI code you can run the vast, vast majority of code on any OS with a decent JVM.
As for usability, you're completely wrong there as well. Java is very quick to develop in - it's got a comprehensive, self-consistent set of standard libraries. It's got an amazing number of external tools and libraries, and there are several (several) free and commercial high quality servlet containers.
I think you probably equate Java with "web client Java", which is meaningless nowadays. Java is a full-blown language and runtime environment - the web is only tangentially relevent to this.
Also, Java has several scripting solutions available - Python, Tcl, Ruby, and countless others. They're even available through a semi-standard interface with the IBM bean scripting framework.
Java offers maintainability, rapid development, speed (are you on crack? Of course it's faster than PHP or Perl unless you're talking the JVM startup time, which is meaningless in real applications), extensive industry support, several nice IDE's.
The Java world is growing, not shrinking. I'll admit if they have that many FreeBSD machines and Java sucks on FreeBSD that may be a legitimate reason not to use it. Your points, however, show a decided lack of understanding.
As for the Britney Spears analogy, I've crapped more trenchant insights. Not funny, not applicable.
The real problem with Java is that people think they understand it because they briefly tried to write an applet 5 years ago, and decided it sucked. It's GOD in the client server arena, and in the servlet/JSP domain.
Their reasoning for throwing out Java seems like utter nonsense. What on Earth does FreeBSD have to do with anything?
They used a red herring argument abuot thread support on FreeBSD (they should change OS's anyway) to discount what's obviously the best choice - Java technology.
I hope more than this one PHP cheerleader is making the decisions on this.
That's strange. There are billions and billions of dollars who each say you're completely wrong, and your opinions insignificant.
Also - this "CISC" you mention, and the alleged violation of "Amdahl's law" (ooooh, important) seem to come to odds with the fact that they still make chips faster than the vastr, vast majority of CPU's out there. Strange, huh corky?
Way to go. Any binary that used the 'HZ' variable (a constant defined in a header file) will need to be recompiled for these new kernels. Way to go, Linux. Keep it up.
Are you daft? How do you know he didn't steal the actual physical film or digital media with the film on it? Are you some kind of "super-insider" on the case - maybe you should start a little newsletter. Got any inside dirt on the Maryland sniper??
Slashdot Mirror
I'm so tempted to go through one of them and find a security hole just for the fun of it. Soo tempted, but sooo lazy. Anyone got some time to spare?
The offensive part is the platitudes. It's one big general diatribe on nothing at all. Reminds me of the preachy father from Hamlet - all words and no content.
Is his point that people need to be more serious about security? Fine. But he's utterly unqualified to be giving technical advice or making technical judgements on a field he obviously knows nothing about (use Python, Ruby, or Java and never C++? I call buzzword bullshit bingo!) other than the vague general knowledge a "technologist" would have.
It's like me saying plastic surgery has become dangerous because there are so many quacks, and they all need to sharpen their scalpels and pay attention to sterility.
I detect a subtle amount of nerd sarcasm. Did Matt Groening develop the Simpsons' 'Comic Book Guy' character with you as a reference?
Was my post "Worst Post Ever"?
And get with the simes. "Boy"? As "Cool" as? At least use some modern slang, e.g. "I wish I was as pimp as you, dawg. Shoot."
This "technologist" is carrying on about bad programmers and security? Wow - I assume he's a seasoned professional with many large-scale projects under his belt?
With such trenchant insights as "Don't use C/C++"! "Don't use Windows!" "Watch out for user input"!.
Wow. How truly insightful. I'm not even going to bother pointing out the utter absurdity of claiming that using or not using C/C++ has anything to do with it, or the added security problems with using high level languages (do you trust the implementation?).
I'm just going to say I've had bloody poops with more useful information in them than this article.
Yet more nonsense. Unix [in general, including Unix-typical tools] has had the most pathetic security history of any operating system.
Cast ye not rocks from a precipice of cracked glass. Unix security is just as crappy as Windows, and has been for a lot longer.
Wow, someone lives in fairy land, with cute little translucently purple winged pixies flitting from nerd to nerd, granting their every drooling little fruity wish.
"Would you like World Peace? Abundant cash? And end to suffering in all forms?"
"N'Glaven, No! I want Microsoft to be crushed!!!"
Yeah, right, halfwit.
100 threads - maybe. 10 threads on a single CPU, nonsense. And the number of CPU's has no bearing. Threads in this situation are just a way to increase responsiveness and to ease development effort. Writing complex event loops is silly unless you're doing intense numerical computations or something else where the CPUs are going to be nearing 100% load.
The 4*NCPUS remark has no bearing in reality.
A lot of development models require threads that usually just sit there, but will occasionally wake up and act. Say you have a UI that needs updating, a logfile that needs written to, a few network sockets that need listening to, etc... There's no reason not to use as many threads as you need.
The ease in programming outwieghs the nonexistent issues in performance because most of your threads aren't doing anything most of the time anyway.
Now, if you're talking about hardcore computation in multiple threads, then obviously the fewer threads the better.
Again, real Unix operating systems have had working, stable M:N thread models for a long time. Why would it be any problem for Linux?
Does this whitepaper by a (relatve) newcomer somehow invalidate 10 years of real experience by the rest of the world and by large commercial companies which make real money selling Unix operating systems? Hardly.
There's nothing more "correct" about NPTL. Real Unix operating systems have used M:N thread semantics for a long, long time.
It was hubris for the developers to even release LinuxThreads ("Oh, our context switches are so fast we don't need user threads!") and the result has been a painful last few years. Oh what joy, every thread is a process! How, umm, clever! Thank God for the new choices (especially NGPT).
You're a filthy hypocrit. I'm sure if this was some evel corporation suing a mom-n-pop company over their use of some idiotic patent you'd be crying out against the horrible inhumanity of it all.
Pathetic.
Johnson, flag this message for section 721 investigation. Potential Conway incident - auth code 786C-1943.
Wow. What a load of crap.
a.) The EU is more interested in economics than racism. If you think their actions are motivated because of racism over economics you're clueless.
b.) What thesis are we supposed to invalidate? His idiotic comparison to Star Wars which is just stupid. Or his.. uhhh, what? What _IS_ his thesis, that the book doesn't impart some life-changing moral?? Good lord.
You are so cool, what with your non-selling outness. Are you that guy who's always droning on about all this original music he listens to?
"Bah. These sellouts today, I tell ya. I only listen to the Screaming Banshees and the Proletariat Meltdown!"
And I think it explains somewhat what the honored maitres were afraid of. At about 1/3 of the book it foreshadows some very powerful force that was sent out into the galaxy to germinate, and that force is probably what scares the shit out of everyone inthe last Dune book by papa Herbert.
The idiocy of your "point" is exceeded only its lack of grammatical and structural flaws.
Are you kidding? Historically Unix security has been absolutely horrible. Look at the number of bugs it's had. It _still_ has more flaws (when you take 'Unix' as a whole) than Windows, and Unix has had 30 years to improve!
Good lord, you must have been kidding.
I can agree with the FreeBSD issue, I guess... The rest I don't, but the point's moot.
I'm actually more thankful it isn't Perl. Good lord, I wish people would quit trying to develop "real" applications with that lumbering piece of merde. It's great for scripts, it's crap for real software development.
You've not done a lot of Java development, have you? Maybe..umm, none beyone a Hello World program or some trivial applet? I find that a lot with Java naysayers.
Your portability arguments were, I believe, thrown in for shock value? Of course Java is very portable. If you don't use JNI code you can run the vast, vast majority of code on any OS with a decent JVM.
As for usability, you're completely wrong there as well. Java is very quick to develop in - it's got a comprehensive, self-consistent set of standard libraries. It's got an amazing number of external tools and libraries, and there are several (several) free and commercial high quality servlet containers.
I think you probably equate Java with "web client Java", which is meaningless nowadays. Java is a full-blown language and runtime environment - the web is only tangentially relevent to this.
Also, Java has several scripting solutions available - Python, Tcl, Ruby, and countless others. They're even available through a semi-standard interface with the IBM bean scripting framework.
Java offers maintainability, rapid development, speed (are you on crack? Of course it's faster than PHP or Perl unless you're talking the JVM startup time, which is meaningless in real applications), extensive industry support, several nice IDE's.
The Java world is growing, not shrinking. I'll admit if they have that many FreeBSD machines and Java sucks on FreeBSD that may be a legitimate reason not to use it. Your points, however, show a decided lack of understanding.
As for the Britney Spears analogy, I've crapped more trenchant insights. Not funny, not applicable.
The real problem with Java is that people think they understand it because they briefly tried to write an applet 5 years ago, and decided it sucked. It's GOD in the client server arena, and in the servlet/JSP domain.
Their reasoning for throwing out Java seems like utter nonsense. What on Earth does FreeBSD have to do with anything?
They used a red herring argument abuot thread support on FreeBSD (they should change OS's anyway) to discount what's obviously the best choice - Java technology.
I hope more than this one PHP cheerleader is making the decisions on this.
That's strange. There are billions and billions of dollars who each say you're completely wrong, and your opinions insignificant.
Also - this "CISC" you mention, and the alleged violation of "Amdahl's law" (ooooh, important) seem to come to odds with the fact that they still make chips faster than the vastr, vast majority of CPU's out there. Strange, huh corky?
Sorry - bullshit. I call bullshit.
Duty bound in what world, your bullshit make-believe world? Once people give the money it's no longer theirs. End of story.
Way to go. Any binary that used the 'HZ' variable (a constant defined in a header file) will need to be recompiled for these new kernels. Way to go, Linux. Keep it up.
Case sensitive filesystems make no sense - Unix is only case sensitive because of impetus and because it was easier at the time.
In a desktop OS, users don't want to have "My Document" and "My document" be two different files. At least Apple is smart enough to know this.
Are you daft? How do you know he didn't steal the actual physical film or digital media with the film on it? Are you some kind of "super-insider" on the case - maybe you should start a little newsletter. Got any inside dirt on the Maryland sniper??