Their stated goal is often times to have a islamic government, like Saudi Arabia, or Iran. I would argue that these are definately fascist governments. Fascists typically are authoritarian (check), highly nationalistic (in an islamic state the nation is suposed to represents the religion - so check), and anti-communist (see the Taliban.)
Agreed; +1. And don't forget racism! They'd throw every Jew in Israel into ovens, except they can't afford the infrastructure.
Most of those comparisons are flawed, it's true: switching from tape to optical media is a significant difference with an obvious difference in features.
But when it comes to dial-up vs. broadband, I'm not so sure. Broadband is just like dial-up, only faster. I guess there's the "always-on" feature, but you could have that if you paid for an extra land line (which is frequently just as expensive as broadband).
Now, you might say: "But with broadband comes all these crazy features that you could never have squeezed down a 56.6 baud pipe!" Yeah, well, the BR/HD folks say the same thing about their new formats: you can cram all kinds of crazy stuff into the dozens of GB these disks can offer. Not to mention that Blu-ray supports a full Java stack, blah blah blah.
And, despite the fact that broadband hasn't universally replaced dial-up, I'd suspect (without evidence) that this is mostly due to limited availablity of broadband in many parts of the US and the world.
If (in a magical fantasy land) the formats didn't get cracked, no one would buy in, and the formats would rot, which would be a good thing.
Why would it be a good thing?
I think you're forgetting that this story begins "in a magical fantasy land"...
With that said, suppose in a magical fantasy land, Blu-Ray and HD-DVD are uncrackable, fulfilling Hollywood's wildest DRM dreams. In this fantasy land, nobody buys them; they're a financial disaster. Hollywood loses a lot of money, and can no longer blame piracy for their financial woes.
That's nice for schadenfreude alone, but also because it teaches them a lesson about what markets will/won't accept. In this world of fantasy, they see the light, learn their lesson, and start offering less draconian formats.
Personally I hope that Blu-Ray and HD-DVD _never_ get cracked, or at least if they do it's never ported to Windows in an easy to use fashion. It's hard to think of any other way to get the formats dropped faster.
Besides, I still don't understand your thinking. If it isn't cracked I ain't buying in.
That was the grandparent post's whole point. If (in a magical fantasy land) the formats didn't get cracked, no one would buy in, and the formats would rot, which would be a good thing.
With that said, I think everybody agrees that the formats certainly will be cracked, so, meh.
It sounds like you agree that 4-6 (various degrees of Locking/Semi-Protection) are all "less wiki" than 7 (mostly free-for-all, some version-flagging). I think you also agree that 3 (everything version-flagged) is less wiki than 7 (only a few things version-flagged).
Since the proposed move is to 7, IMO, it doesn't matter which of the other options is more/less wiki... 7 is wikier than them all.:-)
With that said, you raise an interesting point re: whether 3 is actually more wiki than 4-6. Do we measure it by the most restricted pages? If so, then any Locking, even just a few pages, is less wiki than when everything is version-flagged. But I think it makes more sense to measure it by how much of the wiki is free-for-all, (which is the real "wiki" way,) and then comparing the restrictions on what's left. *shrug*
If you think this is "less wiki", then you've missed the point of Wales' response. Here's a ranking from least wiki to most wiki:
1) Every page is locked: only modifiable by the admins (this is almost every page on the web right now!) 2) Every page is semi-protected: only modifiable by logged in non-new users 3) Every page is version-flagged: where anyone can make a modification, but only non-new users can "bless" the page to make it public. 4) Most pages are a wiki-like free-for-all, but some pages are Locked 5) Most pages are a wiki-like free-for-all, but some pages are Locked and some are Semi-protected (today's status quo) 6) Most pages are a wiki-like free-for-all, but some pages are Semi-protected (and none are locked) 7) Most pages are a wiki-like free-for-all, but some pages are Version-Flagged (and none are locked or semi-protected) 8) All pages are a wiki-like free-for all
Clearly, moving from 5 to 3 would make Wikipedia substantially less wiki, but that's not what they're proposing. But that's not what's happening; they're going to 7. Moving from 5 to 7 makes Wikipedia substantially more wiki.
Wish I'd looked at the Eclipse documentation more closely before posting... there's some fun gotchas buried in the EclipseCon 2006 CDT PowerPoint Presentation available on the Wiki. Here's some delightful reminders of what's wrong with the CDT:
"Turn off Build Automatically" -- In Eclipse/Java, you'd never need to tell someone to do this, even in the largest of projects, because the build runs quickly and incrementally (using the built-in Eclipse Java compiler). But in CDT, the only way to build is to run your entire toolchain using a Makefile. (So instead of fixing this, they provide features to auto-generate the Makefile!)
"The CDT full indexer is very expensive on large C++ projects (Recommendation: Don't use it on such projects)" Gee, thanks! That's the thing that makes Eclipse (in Java) so Eclipse-y, you know? So make sure you turn that off on large projects.
A personal fave is that the debugger integration in eclipse is second to none.
Are you using the same CDT I used? Debugger integration doesn't work. The wrong line of the wrong source file is highlighted. (No, this is not the FAQ question. I'm talking about a totally wrong file.
As I mentioned elsewhere in this thread, the first thing you notice when using the CDT is that Code Navigation doesn't work (most of the time). It works for a few seconds, and then stops working until you restart Eclipse. A feature like that is so important that to have it work only some of the time makes it more frustrating than not having it at all. Then you hit the debugger problems and you realize you're in IDE hell.
The key here is that in every release of the CDT, they've advertised "more accurate" indexing as a feature in the release notes. Get it? More accurate? Indexing source code is hard, I'll admit, but they're nowhere near 100% accuracy, and it really shows. (Part of the problem is that they don't get to use clever Java bytecode analysis tricks to auto-discover function declarations.)
If the CDT were even half as nice as the JDT, I'd use it all the time, really!
As my anonymous peer just remarked, it's "in there" as a feature, supposedly, but it doesn't work (most of the time) in the CDT. This is something the CDT guys fully acknowledge and hope to resolve in future releases of the CDT. (Remember, Eclipse's magic tricks are much easier in Java than in C, thanks to bytecode introspection.)
I'm a game developer. What exactly is Manifesto supposed to do for me? He's obviously not going to put my game in a box and sell it retail. He's not going to get me a development kit for the major consoles. (As you know, only large-scale publishers can purchase those dev kits, at any price.) He's basically going to sell my game for me online, and take a cut. Gee, thanks pal. I can accept PayPal on my own, thank-you-very-much.
The Manifesto Manifesto is nothing but a list of complaints, not a list of services that Manifesto offers.
But then, suppose I'm intrigued, so I want to sign up and sell my game through Manifesto. How do I do that? The "Help" page is busted.
Hello, I'm a sleazy SEO with poor grammar and spelling. It's my job to trick Google (using link spam) into thinking that a web site is more important than it actually is.
Now, many of my customers think that the Google Toolbar will tell them their PageRank, and that this will tell them how good a job I'm doing. I wish they would stop looking at this number, because using that they can see how useless my services are and how effectively Google is combatting my tactics.
Here, let me quote a few irrelevant remarks out of context: "If Chewbacca lives on Endor, you must acquit."
In short, don't pay me to raise your PageRank, [because only improving site quality can do that,] but instead pay me for "targetted traffic", which you can't measure, but is really much more important! Yuk yuk yuk!
... because there is a really good reason to support NTFS in the kernel: so you can boot off of an NTFS drive. That would eliminate the need for Windows users to re-partition their drives when installing Linux, and allow for an easier dual-boot.
Many Java application development frameworks (especially those that were developed prior to Java 1.5) required you to explicitly declare a lot of information about your application in XML. Java Server Faces (JSF) can be pretty bad about this, as can numerous other Java-based frameworks that do "dependency injection" like Spring.
You quickly find that you have almost as many lines of code in XML than you do in Java, (or sometimes even MORE XML than Java) and that keeping your Java and your XML in sync becomes an enormous chore. (Especially since the errors can't normally be detected at compile time.)
This really doesn't have anything specifically to do with Java, as much as the culture that develops around Java. And there's some hope that Java's 1.5 annotations will allow new versions of those frameworks to be developed that use considerably less XML... but most such frameworks are vapor for now.
As I've complained before, the problem here is that you can't develop any games for the major consoles unless you are an authorized developer.
Before you can develop games for consoles you need to apply to the particular console company to become an approved developer. The exact process varies but it generally means proving that you are an experienced game developer with a financially stable company. The console companies won't approve hobby/inexperienced teams to work on their consoles.
Got a great idea for a video game? Well, too bad. You have to get approval first... and BTW, Nintendo doesn't like your idea.
Sure, you can code up homebrew games, but doing so requires hacking your console (which is something most people don't know how to do, and in many cases requires expensive/pseudo-legal mod chips or other hardware).
Next time you see an article about "indie developers", don't get them confused with the ordinary shareware/freebie game developers you see online. The consoles have no indie developers, and probably never will have indie developers. Nintendo, MS and Sony are all afraid that people will code up porn games (which they will) and will disable the console's anti-piracy features (which they will).
I'm astounded the game industry has gotten as far as it has, in light of the fact that you can't develop any games for them unless you are an authorized developer.
Before you can develop games for consoles you need to apply to the particular console company to become an approved developer. The exact process varies but it generally means proving that you are an experienced game developer with a financially stable company. The console companies won't approve hobby/inexperienced teams to work on their consoles.
Sure, you can code up homebrew games, but doing so requires hacking your console (which is something most people don't know how to do, and in many cases requires expensive mod chips or other hardware).
The best way to get creative folks working on game software is to let people develop their ideas without having to get approval first. Hell, I'd even shell out two grand for a development kit if I knew I could actually use it and distribute my games to my friends.
Of course, Nintendo, MS and Sony are all afraid that people will code up porn games (which they will) and will disable the console's anti-piracy features (which they will). Which is why, unfortunately, this will never happen.
In a setting like this, I think you'll want clarity of communication over clarity of writing per se. One of the best ways to do that is constant peer review, and one of the best ways to get that is in a public speaking course, instead of a writing course.
One of the best classes I've ever had in my life was a Voice, Speech and Debate class, with emphasis on the Speech. The premise of the class? You're selected randomly to speak for three to five minutes on a topic selected randomly out of a hat. You have a minute (later 30 seconds, later about five seconds) to come up with your speech. Don't write anything down, and don't use notes or props or cards. You give these speeches in front of the class, and your peers give comments and review you. You're graded on the clarity and organization of your presentation.
Granted, it's not the writing course you were looking for, but I think in teaches what you really wanted to teach: how to organize your thoughts, and how other people do it.
The Chinese people do have the right to privacy guaranteed under the Chinese Constitution. I direct your attention to Article 40. While you're there, you might check out articles 35, 36 and 41. (Most people aren't even aware that China has a Constitution.)
I had originally picked up a copy of this book because it appeared to be the only book on the market that spent any time discussing automated testing of AJAX components.
As I'm sure you all know, testing AJAX stuff in multiple browsers is really important if you want to guarantee cross-browser compatibility; it's also really tedious. JsUnit seemed like it would be a promising tool for AJAX automation.
In fact, I'm sad to say, JsUnit can't be used to validate AJAX components at all; in fact, it can't it be used to validate *any* command that requires a callback, including XmlHttpRequest, event handlers, pop-up windows, etc.
This is because browsers (IE/Firefox both) interpret JavaScript in a single thread, but actions you perform may have asynchronous side effects OUTSIDE of your own thread. So when you attempt one of those fancy asynchronous XmlHttpRequests, you can't just sleep/wait until your request finishes, because it will *never* finish until you completely return from your current thread. Only then will the interpreter begin working on the next item in the event queue.
That means, among other things, that it's impossible to wrap an AJAX request in a "try/catch" block:
Because this will never work, JsUnit's strategy of emulating JUnit or the other *Unit frameworks is fundamentally unsuitable for testing AJAX in multiple browsers.
If you *are* interested in testing AJAX applications in multiple browsers, I recommend looking into Selenium, which basically works around the problem by constantly scheduling timers to re-invoke itself every 10ms... that gives the interpreter enough time to do other work, and allows Selenium to implement a simple "pause" action that actually works.
This question was asked in the Q&A session after Richard's presentation. Richard answered that he will consult a lawyer, and he will do everything possible to fix this, and if you have suggestions they are welcome - and in the end he might fail to find a solution within the GPLv3 to this problem.
This is the primary issue preventing Linus from accepting GPL3. Linus writes:
Notice how the current GPLv3 draft pretty clearly says that Red Hat would have to distribute their private keys so that anybody sign their own versions of the modules they recompile, in order to re-create their own versions of the signed binaries that Red Hat creates. That's INSANE.
The FSF has in turn claimed that Linus has "misread" the license, but it's not at all obvious to me that he has. The GPL3 states:
Complete Corresponding Source Code also includes any encryption or authorization codes necessary to install and/or execute the source code of the work, perhaps modified by you, in the recommended or principal context of use[...].
If DRM'd hardware is the "recommended or principal context of use", then you apparently do have to distribute your private key. If non-DRM'd hardware is the principal context, but the software happens to be used on a DRM'd machine, then you get the Man-in-the-middle attack that the grandparent complains about. The FSF really needs to go back and think about this more carefully.
Remember when you said to just subtract SHR from VIRT, "Problem Solved"? That's not true. LinuxThreads, device memory mapping, and other gotchas mean that there's no easy way to tell just how much memory you'll get back when you kill a given process.
I'm perfectly happy to blame the kernel if it comes to that, for not giving ps and top enough information to do their job. But it makes no sense to blame the user for believing the deceptive information ps reveals.
That's just not true, as someone else has swillden points out in this comment to the current story. Nobody should follow your suggestion.
Based on your over-simplified claim (which I'll call "wrong") the 43 java threads on my Tomcat box are using 3.0GB of RAM total, minus 426MB shared, which is impossible on a box with 256MB of RAM and 512MB swap.
More generally, the problem with ps (and top) is that they fail to highlight the most important piece of information: the amount of unshared memory each process is using, or, as TFA calls it, the "marginal cost" of each process.
Instead, they give you the total memory available to each process. That number is irrelevant to a user of that process. It won't tell you, for example, how much memory you'd save if you killed off any given process. It won't even tell you how much total memory (shared+unshared) that process is using... as others have pointed out, ps's number includes unused copy-on-write device-mapped memory.
"MFN" stands for most favored nation status. The US normally calls it normal trade relations, which is probably a better name for it. As an earlier poster pointed out, China received permanent NTR in December of 2001. (I believe the Bush administration was hoping that China would tend liberalize its country if it started to trade more with us. It has liberalized substantially relative to what it used to be, but it's still a very repressive regime, so I suppose the jury is still out on this question.)
Slashdot Mirror
Their stated goal is often times to have a islamic government, like Saudi Arabia, or Iran. I would argue that these are definately fascist governments. Fascists typically are authoritarian (check), highly nationalistic (in an islamic state the nation is suposed to represents the religion - so check), and anti-communist (see the Taliban.)
Agreed; +1. And don't forget racism! They'd throw every Jew in Israel into ovens, except they can't afford the infrastructure.
Most of those comparisons are flawed, it's true: switching from tape to optical media is a significant difference with an obvious difference in features.
But when it comes to dial-up vs. broadband, I'm not so sure. Broadband is just like dial-up, only faster. I guess there's the "always-on" feature, but you could have that if you paid for an extra land line (which is frequently just as expensive as broadband).
Now, you might say: "But with broadband comes all these crazy features that you could never have squeezed down a 56.6 baud pipe!" Yeah, well, the BR/HD folks say the same thing about their new formats: you can cram all kinds of crazy stuff into the dozens of GB these disks can offer. Not to mention that Blu-ray supports a full Java stack, blah blah blah.
And, despite the fact that broadband hasn't universally replaced dial-up, I'd suspect (without evidence) that this is mostly due to limited availablity of broadband in many parts of the US and the world.
I think you're forgetting that this story begins "in a magical fantasy land"...
With that said, suppose in a magical fantasy land, Blu-Ray and HD-DVD are uncrackable, fulfilling Hollywood's wildest DRM dreams. In this fantasy land, nobody buys them; they're a financial disaster. Hollywood loses a lot of money, and can no longer blame piracy for their financial woes.
That's nice for schadenfreude alone, but also because it teaches them a lesson about what markets will/won't accept. In this world of fantasy, they see the light, learn their lesson, and start offering less draconian formats.
With that said, I think everybody agrees that the formats certainly will be cracked, so, meh.
I agree, and that's a very legitimate concern. I guess we'll just have to see what happens...
It sounds like you agree that 4-6 (various degrees of Locking/Semi-Protection) are all "less wiki" than 7 (mostly free-for-all, some version-flagging). I think you also agree that 3 (everything version-flagged) is less wiki than 7 (only a few things version-flagged).
:-)
Since the proposed move is to 7, IMO, it doesn't matter which of the other options is more/less wiki... 7 is wikier than them all.
With that said, you raise an interesting point re: whether 3 is actually more wiki than 4-6. Do we measure it by the most restricted pages? If so, then any Locking, even just a few pages, is less wiki than when everything is version-flagged. But I think it makes more sense to measure it by how much of the wiki is free-for-all, (which is the real "wiki" way,) and then comparing the restrictions on what's left. *shrug*
If you think this is "less wiki", then you've missed the point of Wales' response. Here's a ranking from least wiki to most wiki:
1) Every page is locked: only modifiable by the admins (this is almost every page on the web right now!)
2) Every page is semi-protected: only modifiable by logged in non-new users
3) Every page is version-flagged: where anyone can make a modification, but only non-new users can "bless" the page to make it public.
4) Most pages are a wiki-like free-for-all, but some pages are Locked
5) Most pages are a wiki-like free-for-all, but some pages are Locked and some are Semi-protected (today's status quo)
6) Most pages are a wiki-like free-for-all, but some pages are Semi-protected (and none are locked)
7) Most pages are a wiki-like free-for-all, but some pages are Version-Flagged (and none are locked or semi-protected)
8) All pages are a wiki-like free-for all
Clearly, moving from 5 to 3 would make Wikipedia substantially less wiki, but that's not what they're proposing. But that's not what's happening; they're going to 7. Moving from 5 to 7 makes Wikipedia substantially more wiki.
Thanks!
Wish I'd looked at the Eclipse documentation more closely before posting... there's some fun gotchas buried in the EclipseCon 2006 CDT PowerPoint Presentation available on the Wiki. Here's some delightful reminders of what's wrong with the CDT:
"Turn off Build Automatically" -- In Eclipse/Java, you'd never need to tell someone to do this, even in the largest of projects, because the build runs quickly and incrementally (using the built-in Eclipse Java compiler). But in CDT, the only way to build is to run your entire toolchain using a Makefile. (So instead of fixing this, they provide features to auto-generate the Makefile!)
"The CDT full indexer is very expensive on large C++ projects (Recommendation: Don't use it on such projects)" Gee, thanks! That's the thing that makes Eclipse (in Java) so Eclipse-y, you know? So make sure you turn that off on large projects.
Oh, and there's my personal favorite FAQ: Can I debug Java and C++ at the same time? Answer? "If you can get this to work, please let the cdt-dev mailing list know!"
The Eclipse CDT is a joke. Even Visual Studio can handle reference searches on large projects.
As I mentioned elsewhere in this thread, the first thing you notice when using the CDT is that Code Navigation doesn't work (most of the time). It works for a few seconds, and then stops working until you restart Eclipse. A feature like that is so important that to have it work only some of the time makes it more frustrating than not having it at all. Then you hit the debugger problems and you realize you're in IDE hell.
The key here is that in every release of the CDT, they've advertised "more accurate" indexing as a feature in the release notes. Get it? More accurate? Indexing source code is hard, I'll admit, but they're nowhere near 100% accuracy, and it really shows. (Part of the problem is that they don't get to use clever Java bytecode analysis tricks to auto-discover function declarations.)
If the CDT were even half as nice as the JDT, I'd use it all the time, really!
As my anonymous peer just remarked, it's "in there" as a feature, supposedly, but it doesn't work (most of the time) in the CDT. This is something the CDT guys fully acknowledge and hope to resolve in future releases of the CDT. (Remember, Eclipse's magic tricks are much easier in Java than in C, thanks to bytecode introspection.)
I'm a game developer. What exactly is Manifesto supposed to do for me? He's obviously not going to put my game in a box and sell it retail. He's not going to get me a development kit for the major consoles. (As you know, only large-scale publishers can purchase those dev kits, at any price.) He's basically going to sell my game for me online, and take a cut. Gee, thanks pal. I can accept PayPal on my own, thank-you-very-much.
The Manifesto Manifesto is nothing but a list of complaints, not a list of services that Manifesto offers.
But then, suppose I'm intrigued, so I want to sign up and sell my game through Manifesto. How do I do that? The "Help" page is busted.
http://msdn.microsoft.com/directx/xna/faq/
That was my main criticism of The Sims as well. (Shameless plug) If you're like me, you may prefer Alter Ego instead.
"What if you could live your life over again?"
... because there is a really good reason to support NTFS in the kernel: so you can boot off of an NTFS drive. That would eliminate the need for Windows users to re-partition their drives when installing Linux, and allow for an easier dual-boot.
Many Java application development frameworks (especially those that were developed prior to Java 1.5) required you to explicitly declare a lot of information about your application in XML. Java Server Faces (JSF) can be pretty bad about this, as can numerous other Java-based frameworks that do "dependency injection" like Spring.
You quickly find that you have almost as many lines of code in XML than you do in Java, (or sometimes even MORE XML than Java) and that keeping your Java and your XML in sync becomes an enormous chore. (Especially since the errors can't normally be detected at compile time.)
This really doesn't have anything specifically to do with Java, as much as the culture that develops around Java. And there's some hope that Java's 1.5 annotations will allow new versions of those frameworks to be developed that use considerably less XML... but most such frameworks are vapor for now.
Sure, you can code up homebrew games, but doing so requires hacking your console (which is something most people don't know how to do, and in many cases requires expensive/pseudo-legal mod chips or other hardware).
Next time you see an article about "indie developers", don't get them confused with the ordinary shareware/freebie game developers you see online. The consoles have no indie developers, and probably never will have indie developers. Nintendo, MS and Sony are all afraid that people will code up porn games (which they will) and will disable the console's anti-piracy features (which they will).
The best way to get creative folks working on game software is to let people develop their ideas without having to get approval first. Hell, I'd even shell out two grand for a development kit if I knew I could actually use it and distribute my games to my friends.
Of course, Nintendo, MS and Sony are all afraid that people will code up porn games (which they will) and will disable the console's anti-piracy features (which they will). Which is why, unfortunately, this will never happen.
In a setting like this, I think you'll want clarity of communication over clarity of writing per se. One of the best ways to do that is constant peer review, and one of the best ways to get that is in a public speaking course, instead of a writing course.
One of the best classes I've ever had in my life was a Voice, Speech and Debate class, with emphasis on the Speech. The premise of the class? You're selected randomly to speak for three to five minutes on a topic selected randomly out of a hat. You have a minute (later 30 seconds, later about five seconds) to come up with your speech. Don't write anything down, and don't use notes or props or cards. You give these speeches in front of the class, and your peers give comments and review you. You're graded on the clarity and organization of your presentation.
Granted, it's not the writing course you were looking for, but I think in teaches what you really wanted to teach: how to organize your thoughts, and how other people do it.
The Chinese people do have the right to privacy guaranteed under the Chinese Constitution. I direct your attention to Article 40. While you're there, you might check out articles 35, 36 and 41. (Most people aren't even aware that China has a Constitution.)
As I'm sure you all know, testing AJAX stuff in multiple browsers is really important if you want to guarantee cross-browser compatibility; it's also really tedious. JsUnit seemed like it would be a promising tool for AJAX automation.
In fact, I'm sad to say, JsUnit can't be used to validate AJAX components at all; in fact, it can't it be used to validate *any* command that requires a callback, including XmlHttpRequest, event handlers, pop-up windows, etc.
This is because browsers (IE/Firefox both) interpret JavaScript in a single thread, but actions you perform may have asynchronous side effects OUTSIDE of your own thread. So when you attempt one of those fancy asynchronous XmlHttpRequests, you can't just sleep/wait until your request finishes, because it will *never* finish until you completely return from your current thread. Only then will the interpreter begin working on the next item in the event queue.
That means, among other things, that it's impossible to wrap an AJAX request in a "try/catch" block:Because this will never work, JsUnit's strategy of emulating JUnit or the other *Unit frameworks is fundamentally unsuitable for testing AJAX in multiple browsers.
If you *are* interested in testing AJAX applications in multiple browsers, I recommend looking into Selenium, which basically works around the problem by constantly scheduling timers to re-invoke itself every 10ms... that gives the interpreter enough time to do other work, and allows Selenium to implement a simple "pause" action that actually works.
Remember when you said to just subtract SHR from VIRT, "Problem Solved"? That's not true. LinuxThreads, device memory mapping, and other gotchas mean that there's no easy way to tell just how much memory you'll get back when you kill a given process.
I'm perfectly happy to blame the kernel if it comes to that, for not giving ps and top enough information to do their job. But it makes no sense to blame the user for believing the deceptive information ps reveals.
That's just not true, as someone else has swillden points out in this comment to the current story. Nobody should follow your suggestion.
Based on your over-simplified claim (which I'll call "wrong") the 43 java threads on my Tomcat box are using 3.0GB of RAM total, minus 426MB shared, which is impossible on a box with 256MB of RAM and 512MB swap.
More generally, the problem with ps (and top) is that they fail to highlight the most important piece of information: the amount of unshared memory each process is using, or, as TFA calls it, the "marginal cost" of each process.
Instead, they give you the total memory available to each process. That number is irrelevant to a user of that process. It won't tell you, for example, how much memory you'd save if you killed off any given process. It won't even tell you how much total memory (shared+unshared) that process is using... as others have pointed out, ps's number includes unused copy-on-write device-mapped memory.
ps is at best deceptive, if not actually wrong.
"MFN" stands for most favored nation status. The US normally calls it normal trade relations, which is probably a better name for it. As an earlier poster pointed out, China received permanent NTR in December of 2001. (I believe the Bush administration was hoping that China would tend liberalize its country if it started to trade more with us. It has liberalized substantially relative to what it used to be, but it's still a very repressive regime, so I suppose the jury is still out on this question.)