Because it's not about features. It is about guarantees - of performance, of compatability, of deliverability.
This is the difference between open-source development and enterprise software development. Open source is flexible, nimble, able to add new features fast and often of good quality - but open source pays a steep cost in terms of backwards compatability and a lack of performance guarentees. An enterprise customer is very willing to give large piles of cash for Sun to say, "our next version of java will have feature Z and will be at least as fast as the current version on your workload". Open source is ripe with developers that say, "X has a problem, I'm taking it out and replacing it with Y. This will drop performance by 30% in the next version, but it includes new feature Z". (Don't think that's common? gcc3.0 and 4.0 both did that... sacrificed lots of performance and compatibility for new ABIs. Look how long 2.95 stayed around because of that.)
Java is a loss-leader that is Sun's entry ticket into the enterprise market - it is an enterprise-ready product. Don't suggest that giving up control of Java is a win until you factor in the loss of Sun's leadership position in enterprise middleware platforms; the cost of losing that is far more than what Sun spends on Java.
Sun has been very reticent to actually "Open Source" (note the caps) Java because of the problems they had with Microsoft. Had Microsoft not abused their contract with Sun all those years ago, Sun might still be releasing only a reference implementation for others to build their own JVMs against.
This is just complete and utter nonsense. Microsoft is free to implement whatever language they want. Whether it's based on Sun's code or not is completely beside the point. What they can't do is calling it Java unless it *is* Java, and that was the subject of the lawsuit.
Wow. You have managed to advance a technically correct, but completely and utterly worthless point. (Seriously, your argument is so good at arguing while avoiding the point entirely that you'd have a great future as a Microsoft lawyer).
Microsoft can fork Java - and has. Microsoft tried to fork Java once (back in the 1.1 days), and it took Sun years to shut down that fork (well into 1.4). And for all that time, Java writers were crippled by having to code to the 1.1 standard because that was all MS Java and Sun Java had in common (which delayed acceptance of Swing by several years). Then Microsoft came up with.NET and the CLR, which when you think about it, is a reimplementation of Java with different semantics. It's a de facto fork, just different enough to avoid lawsuits but attempting to do exactly the same thing. And given how thoroughly Mono has embraced that fork, Microsoft succeeded.
The name "Java" is something for lawyers to argue about - I could start a language called NotJava, which implements 100% of the Java spec but isn't certified, and (assuming my implementation were good) I guarantee you people would use it. After all, Linux isn't certified by anyone, and everyone uses Linux kernels. In terms of the market, it only matters that software be as easy to use as Java (i.e. have good documentation, easy to install, easy to write effective programs) and as powerful and featureful as Java. I repeat: the name "Java" does not matter.
There are already independent java implementations out there - look at IBM for one, GCJ for another (albeit poor) one. RMS is whining because he's found a java implementation to actually be hard, and he'd like Sun to do his work for him. He already has what his standard rhetoric asks for: an open standard upon which Free Software can compete fairly. But Free Software has done very poorly in comparison to Sun's Java. RMS is unable to understand that his Free Software dream has been beaten by an open and technically better competator without matching perfectly with his vision, so he starts ranting about how Sun isn't RMS-approved Open Source(tm) when they really are open source.
And RMS would like nothing better than to fork Java, insert some GPL-only technology, and see that technology adopted such that it kills off Sun's non-GPL Java and makes the world a "happy, GPL-only place". This is actually quite clever: instead of implementing the whole of the runtime, he only has to implement one feature. It's an insidious ploy, worthy of Microsoft, and RMS is complaining about how Sun isn't letting him do it?
So at the end, here's my point. Sun is worried about two forks. One is the Microsoft fork, and how guaranteed prolifieration can stall adoption of technologies essential for Java's success. Second is the GPL fork, which adopts good features but forces Sun out of the Java business because they can't match the features without GPLing their own code. (BTW, the OpenSolaris license is squarely aimed at that second concern).
I've seen McNealy talk, but never Schwartz. I was utterly impressed with McNealy's ability to sell out-of-the-box ideas and put together a cohesive picture. Schwartz... based on his blog, all I've seen is an ability to hype Sun without actually producing deliverables. Schwartz has been running the company for the past few years - anything interesting come out? Oh yeah, Solaris 10 and Opteron servers, both projects started via McNealy before Schwartz took over.
To an engineer, McNealy is the real deal, and Schwartz is a PHB. Maybe the Fortune500 CEO crowd would like this change, but I expect Sun would lose the last of their engineering talent were Schwartz to take over. They would become just another company cutting quality to make a buck.
Your first post was great. There's a very famous quote by Thomas Jefferson (I believe, and I'm probably not 100% accurate): "I despise what you have to say, but I will defend to the death your right to say it". For a few brief minutes, I respected you.
And then this crap. Have you read the "fire in a theater" decision? Apparently not, since Oliver rWendell Holmes spent a very long time explaining exactly why the government DID have a RESPONSIBILITY to protect against such speech. Why? Because all rights are in tension; the right to say "fire" (1st amendment) is outweighed by everyone else's right to be safe and secure in the possession of his or her life (4th and 14th amendment, due process clause). The government decided any reasonable person would expect that hearing "fire" in a theater meant an emergency, and the risk of a loss of life outweights the minor loss of freedom of not being able to say that word. And the premise is far from ridiculous: people DIED in the era before fire codes mandated evacuation schemes that work.
Libertarianism has a fundamental flaw: it assumes that anyone who should have responsibility exercises that responsibility rationally. If local governments were perfect, the Civil Rights Act would not have been necessary. Did you by chance notice that local governments aren't perfect? And that the whole point of a federal law (imperfect as it may be) was to give those who wanted a better world leverage against reactionary neandrethals who liked the status quo ante bellum?
I have no problem with communities (not necessarily meaning villages or towns but organizations of people with like apprecations and prejudices) joining together to chat about their beliefs.
Fine. This "community" called the United States of America joined together on principles of tolerance and freedom. And I find it rather arrogant that you waltz through claiming the freedoms you desire while dismissing the responsibilities that make those freedoms possible as "not your problem," that you dismiss laws you don't personally agree with as ridiculous and not worth obeying. If you think the Supreme Court and the laws of this country are wrong, then please leave the country.
Alas, illegal (at least in the US). Mac OS X has some trusted computing code that verifies that it is running on Apple hardware. Bypassing this code is a definite DMCA violation (hence no company will try it). AND Apple assumes a very narrow set of underlying hardware, which isn't what VMs provide - so there is another pile of effort to emulate chipset and so on.
In short, it's not going to happen until Apple wants it to happen.
Graph theory... oddly enough, I use more graph theory now than I ever did back in school.
Principle application: reasoning about multithreaded processes. It's HARD. Lots of state machines (yes, finite automata). Another problem, we wanted to prove a tree doesn't deadlock. (Yes, trees can deadlock when parallelism is insufficient. It sucks.) How do we avoid a deadlock?
In other words... Blizzard allows macros that don't fundamentally alter the game experience. This user used outside equipment (keyboard) to set up macros that do fundamentally alter the gaming experience - allow him to run automated attacks, from what I can read here - and Blizzard banned him for it.
I don't see this as any more novel than someone getting banned for inserting a graphics driver wallhack. He's violating the spirit of the game (no automated character improvements; you must invest your own time) while trying to weasel around the letter of the rules. Blizzard is vague precisely so that they can ban smartasses like this guy!
(All that said, I think Blizzard would be better served inventing some other "punishment" for this. Like, if your character spends 20 minutes attacking a critter it shouldn't be able to kill and it looks like you are a bot, some big SLOW nasty spawns and kills you. If you weren't a bot, you could outrun the nasty... but dumbass bot users die. And maybe take an experience penalty too.)
Xen doesn't do graphics cards (at this point, everyone - even VMware - virtualizes graphics cards). Network cards, you only get via a virtualized Xen network adapter with Xen drivers, so those will remain the same. Nobody does passthrough hardware - nobody.
You will find processor differences though. Move from AMD to Intel, or drop SSE extensions or some such, and things will break.
This is necessary - I did this about once a month for the past year. Ah, the joys of being in-dorm tech support for a hundred college students...
I only know of one problem. You really have to learn by removing a bunch of this crap yourself - new junk hides itself in new ways.
My five-step process:
1) Reboot in safe mode
2) Delete anything in C:\WINDOWS and C:\WINDOWS\SYSTEM32 (or whatever directories of choice) that has a hidden attribute and appeared since "problems began" (usually a month or so).
3) Wipe all temp directories. (that's C:\Documents and settings\username\local settings\temp and \temporary internet files, and maybe others I've forgotten).
4) Use regedit to remove strange Run, RunOnce, etc. entries. If in doubt, google, then destroy. Your user can always reinstall.
5) Reboot into normal Windows, then run a good antivirus and a good adware remover. BEFORE reconnecting to the network. (This may require having virus defs on a USB key).
The anti-spyware seem to get ~80% of what's out there. This gets 95%. Upgrade to the GP's PE environment instead of safe mode, you're probably at 99%. Anything else, transfer files off and reformat, because it's probably a rootkit. With practice, I got the above proceedure down to half an hour during "new computer" season.
I take some things back - this is a good discussion! And I apologize for some sloppiness of mixing P-M and P-D. I think Intel is deliberately coming up with confusing names to muddy the water.
7th: accepted, that's what I get for trying to go from memory. Having pulled out my references... PPro through PIII/Xenon are P6 cores, 6th gen. P4 is NetBurst core, 7th gen; NetBurst is the deeper pipeline (higher inflight) and register renaming for major improvements, and some tweaks (deeper btb, adaptive prefetch, increasing ALUs from 5 to 7). Athlon matches the renaming, and beefed up the ALUs to go as fast without the deeper pipeline - yes, all 7th gen. (And now I've found my old notes that say 7th gen is massively OOO, d'oh!)
Re: 8th gen... going from 3-issue to 4-issue isn't going to do much. P6 can't even keep a 3-issue fully saturated; a 4-issue will show some improvement, but far below the 33% theoretical gain. (Yes, Intel claims they'll make 4 IPC under load. I don't believe them. I suspect 4 IPC under ideal optimized load, <2 IPC under real load. How can you make 4 IPC when real code averages a branch every 5 instrs?). And I don't see convoying as a good thing. Itaniums were VLIW, but the compilers still aren't caught up; Merom can't expect compiler support (most compilers now target 5th gen Pentiums, no one will compile for Merom) except for their publicity benchmarks, and HPC users. Convoying looks good in benchmarks, but terrible outside HPC environments. It won't help on the desktop OR the server.
Adding 64-bit is not a mere port. 100+ registers to support renaming, major ISA changes (doubling GP registers and adding 4-level page tables) - the gain from implementing those is astounding. Easily more than a 4-issue will show. Opterons are 8th gen, no doubt about it. P4 seems to smear across 7th and 8th (they had a lot of cores, though no major arch. changes). Merom is a good 8th gen core designed from the ground up; Opteron is a so-so 8th gen core designed from the ground up (first system effect); P4 is a 7th gen core with 8th gen features bolted on.
I would actually claim the real 8th-gen features are x86-64 (the ISA extensions), MATURE register renaming (source of Opteron's performance gain in 32-bit), and good power management (selectively disabling unused parts of the chip). Merom is better than Opteron at these, but evolutionarily so.
Opteron shows up on every spec I've seen as "k8" (usually with some other letter indicating a revision). P-M (the modern descendant of the P3 core) probably crosses boundaries - some versions of it jump to 64-bit, it's a very long-lived core. But P4 is a joke - it's faster than a P-3 or P-M, but architecturally has very limited innovation. Opeterons are dramatically different from Athlons or anything in the Pentium line (except the most recent pentium revisions).
And if you claim that a Conroe/Merom is more advanced than a modern Opteron, you REALLY need to look at your processor architectures. It ain't - it's pulling even. Barely.
The next gen is VT - virtualization instructions. Which is actually coming up quite fast.
And this is one that looks to be messy for AMD. Intel's VT is more mature than AMD's Pacifica - not by as wide a margin as AMD's dual core beat Intel's, but Intel does have the virtualization edge. And once virtualization takes off - once VMware, Xen, and anyone else can run interesting stuff on generic chips - THEN we will need a whole new batch of processors.
Not immediately - I don't think the current generation of VT designs are good enough yet, I think it will take another iteration. But there is a wide market that Intel is going to hit just a hair before AMD. Which is a shame, because I really like AMD's chips.
Generational thought: Athlons / P3s and P4s were 7th generation, Opterons and 64-bit Pentiums are 8th generation. I don't know what 9th generation will look like yet - and I hope Intel / AMD don't try to claim it prematurely.
So this guy is talking to people saying "Im going to use a torch".... what is to keep other terrorists, connections, etc. from contacting him and saying "but I have access to bombs"?
You lost me at step two. If other terrorists, connections, etc. have access to bombs, why aren't they blowing up bridges already? Why do they need to find a kook to do the dirty work? Answer: they don't. This whole argument is a rational for harassing people who aren't involved in some sort of plot on the falicious assumption that it would block that same plot.
The argument that a net is not worthwhile only because it tends to catch small fish is silly. There just happen to be more small players, some are just kooks, but a kook with a good connection is a dangerous crazy person.
The problem is that the shrills trying to sell this net claim it catches big fish without killing too many other creatures that get caught in the net. The net doesn't catch big fish (because big fish see it and stay away from it), the owners of the net spend so much time dealing with small fish that they are less effective at catching large fish, and a net fine enough to catch small fish also catches all sorts of other creatures like squid and jellyfish and innocent citizens.
The administration claims this activity makes us safer. Which has just enough truth to be seductive: it does make us safer against the one-in-a-billion kook who could actually do damage. But the whole truth is scarier: protecting against that kook lets the really dangerous folks get away, AND harms bystanders. The administration is too busy patting itself on the back for doing something to notice that the something does more harm than good.
On the matter of 'pipelining': it should be rather obvious that if your pipeline isn't wide enough to automatically correct typos and (easily understandable) misspellings, you are clearly reading too fast for your apprehension skills.
Branch prediction, baby. With accurate branch prediction, CPI (cycles per instr., for those who don't architect processors) drops below 1, because branches are free!
Tongue-in-cheek. Now seriously: I only encounter problems when the wrong word is used. 'grammer' is OK because it isn't a word and thus is an obvious typo (and thus no pipeline stall); 'to'/'too' or 'than'/'then' is a problem. I refuse to correct those as typos because sometimes that word does make sense in context - I prefer to take an uncommon single stall for a wrong word over instead of taking two stalls for a more rare but grammatically correct usage (one stall realizing the corrected word doesn't parse, and a second stall from the pipeline hazard of overriding the correction).
Windows DOES NOT allow for that sort of thing. Yes, if you restrict yourself to just the API that's backwards compatable you'll be fine... but have you read the Win32 API docs? Notice how many of them behave differently on previous versions of windows - or simply aren't implemented there? Any non-trivial application (read: any real-word application) will have some of those version-dependent API calls, whether due to use by the program itself or by support libraries. Most server-class apps are written with a W2K or lower API target, regardless of what platform they actually run on, just to avoid these issues.
Supported configurations are a Big Deal. The real flaw of this study was that it studied supported configurations under Windows, but included unsupported configurations under Linux. Choosing products by market share is not a justification for such blatant administrative stupidity. An unsupported configuration means potentially more work; that is all this study proves, and we knew that already.
Oh it's quite simple, when you look at it with enough cynicism.
By blaming the lack of adequate equipment in the room for a lack of productivity, the faculty can claim that it is not their fault they are not being effective teachers. Which misses the fundamental point: all the IT-provided bells and whistles will not make someone a better teacher; it takes work, skill, and dedication, just like everything else in life. (I'm not banging on teachers specifically here... we're all just as guilty. I mean, if I had a third development box under my desk I'm sure I'd be much more productive...:-)
The best professors I had in college didn't care whether students were checking e-mail or whatnot during class - they cared about whether their lectures were coherent, sensible, and imparted the desired knowledge to their students. Some of my fellow students used laptops to goof off... some used laptops to be more productive... I honestly don't believe the grade distribution between laptop-users and non-laptop-users is statistically significant.
And by the way, after watching college IT bail out way too many clueless professors who didn't understand technology half as well as they pretended, I have a huge respect for you guys!
Think about it, you can get a decent new computer for around $500; do you really think that the cost of the operating system should be half of that?
Yes I do. I would quite happily pay $250 for an operating system that is (1) stable, (2) well-supported, and (3) part of the general market. Unfortunately, Windows fails at (1), Linux fails at (2) ["visit our forum, fix it yourself" isn't support in my book], and OSX & Solaris fail at (3). Which is a shame; I'm happy to spend my money, but don't see anything worth spending it on.
The students I know who actually use Photoshop happily pay the student cost. They are cartoonists, game hobbyists, design students, and so on. They need Photoshop because no other program serves their needs.
The students I hear complain about the cost of Photoshop just want a cheap paint program and should go buy Paint Shop Pro (
I look at Adobe and see one of the best pricing structures anywhere in corporate america. Free "viewers" on a non-secret format (PDF), deep discounts for powerful software to people who really need the discount, and reasonable prices for software that's a generation ahead of anything else out there (& don't anyone start a gimp flame-war, you KNOW it's not as powerful as Photoshop) - prices any company that actually needs those features can readily pay.
My experience with international students (while working as staff in a dorm) was that they tended to have pretty uniform educational and class backgrounds. High middle or upper class, often educated at private schools in England (for former British colonies) or at a very small cross-section of schools in their own country. Well-traveled, and mostly "americanized" - or at least very Westernized - to begin with. Apparently, the folks who aren't half americanized already stay in their own country for higher education. See, there's this catch the colleges don't mention that college is much more expensive for international students (less financial aid available), so their "diverse" international communities end up being pretty homogenous socio-economic communities.
So if by diverse, you mean people who take their other religions as seriously as educated American (i.e. not that seriously), speak with a sometimes-thick accent but act American already, then yes. They are diverse. But I saw more diversity and cultural enrichment with the university's efforts to vary socio-economic backgrounds of their American admits. (And for reference, I don't really see a need to change the system.)
Your statement implies that they therefore have no value beyond perhaps the cost of getting them set up.
Precisely - open source products have a (generally) high setup cost. Closed source products are all over the place, but even Windows has a lower setup cost than most open-source. (Hang on... I'll explain why...)
Your "apt-get install mozilla-firefox" example is convenient - and wrong. You've ignored two costs that are EXTREMELY high: learning to use apt-get and figuring out the name mozilla-firefox. (Cheeky, yes, but bear with me). apt-get is a complex and powerful tool that requires you to set up the right repositories and manually handle dependency conflicts (libc issues...). Simply issuing "install" works if the system is in perfect working order... it takes SERIOUS time-cost to get it there, or a uselessly uncustomized stock install. The name mozilla-firefox... gee, I thought the program was named firefox! Right now, go to google, start searching for how to install firefox on linux... first hit I get says download a tarball and run./firefox_installer. Certainly no apt-get instructions there... and wait, this differs from your instructions, who's right? (reaches for PANIC button). Multiple conflicting install paths, and firefox is one of the easiest Linux installs I've ever done (to be fair, it took me only two days. Because even though firefox was compatable, the installer had a libc conflict with my base install's libsafe. You expect a novice to figure that out?). Compare to MS Windows: I stick in the install CD, and click pretty obvious buttons.
Alright, I'm being facetious... but I do have a point. To someone who already knows what they need, open source is fine. To someone who doesn't know, who doesn't have mastery of every configuration option and tool they need... paying extra for closed source (and the polish that goes along with it) is far more economical.
What's the real value of open source alternatives... lets see there's the utility of the programs themselves, the value of having the source code that I can alter myself or pay someone else to alter, that value of having a community that is motivated for various reasons to give free or cheap support, and when necessary paid support. Often you get to talk directly to the developer of the software themselves. What's all that worth?
What's it worth to you? Quite a bit, I take it. To me, who doesn't need custom compilers or custom web browsers and who isn't interested in manually updating and patching the codebase, and who doesn't have the background in this codebase to follow free community-offered help NOR THE TIME TO LEARN THAT BACKGROUND, it has zero value.
And I'm rather offended at open source zealots who insist their code has value to me when I look at the time I'd have to invest to learn it and see negative value.
I don't complain about the existence of free, open-source products - heck, I use many of them and appreciate them! But I am deeply offended at this "open source is the one true way" and "you're either with us or against us" attitude that tells me the work I do is meaningless. Meaningless to you, maybe, but I'm not trying to sell to you. I'm trying to sell to people who appreciate the value of what I do.
Guess you don't have any kids that normally ride in the passenger's seat...
The passenger's seat is where your (1) significant other and (2) younger relations ride. So what if commuters usually don't have a passenger and it's safer for them... you'll never get any politician to claim a commuter is more important than a kid.:-)
Slashdot Mirror
This is the difference between open-source development and enterprise software development. Open source is flexible, nimble, able to add new features fast and often of good quality - but open source pays a steep cost in terms of backwards compatability and a lack of performance guarentees. An enterprise customer is very willing to give large piles of cash for Sun to say, "our next version of java will have feature Z and will be at least as fast as the current version on your workload". Open source is ripe with developers that say, "X has a problem, I'm taking it out and replacing it with Y. This will drop performance by 30% in the next version, but it includes new feature Z". (Don't think that's common? gcc3.0 and 4.0 both did that... sacrificed lots of performance and compatibility for new ABIs. Look how long 2.95 stayed around because of that.)
Java is a loss-leader that is Sun's entry ticket into the enterprise market - it is an enterprise-ready product. Don't suggest that giving up control of Java is a win until you factor in the loss of Sun's leadership position in enterprise middleware platforms; the cost of losing that is far more than what Sun spends on Java.
This is just complete and utter nonsense. Microsoft is free to implement whatever language they want. Whether it's based on Sun's code or not is completely beside the point. What they can't do is calling it Java unless it *is* Java, and that was the subject of the lawsuit.
Wow. You have managed to advance a technically correct, but completely and utterly worthless point. (Seriously, your argument is so good at arguing while avoiding the point entirely that you'd have a great future as a Microsoft lawyer).
Microsoft can fork Java - and has. Microsoft tried to fork Java once (back in the 1.1 days), and it took Sun years to shut down that fork (well into 1.4). And for all that time, Java writers were crippled by having to code to the 1.1 standard because that was all MS Java and Sun Java had in common (which delayed acceptance of Swing by several years). Then Microsoft came up with .NET and the CLR, which when you think about it, is a reimplementation of Java with different semantics. It's a de facto fork, just different enough to avoid lawsuits but attempting to do exactly the same thing. And given how thoroughly Mono has embraced that fork, Microsoft succeeded.
The name "Java" is something for lawyers to argue about - I could start a language called NotJava, which implements 100% of the Java spec but isn't certified, and (assuming my implementation were good) I guarantee you people would use it. After all, Linux isn't certified by anyone, and everyone uses Linux kernels. In terms of the market, it only matters that software be as easy to use as Java (i.e. have good documentation, easy to install, easy to write effective programs) and as powerful and featureful as Java. I repeat: the name "Java" does not matter.
There are already independent java implementations out there - look at IBM for one, GCJ for another (albeit poor) one. RMS is whining because he's found a java implementation to actually be hard, and he'd like Sun to do his work for him. He already has what his standard rhetoric asks for: an open standard upon which Free Software can compete fairly. But Free Software has done very poorly in comparison to Sun's Java. RMS is unable to understand that his Free Software dream has been beaten by an open and technically better competator without matching perfectly with his vision, so he starts ranting about how Sun isn't RMS-approved Open Source(tm) when they really are open source.
And RMS would like nothing better than to fork Java, insert some GPL-only technology, and see that technology adopted such that it kills off Sun's non-GPL Java and makes the world a "happy, GPL-only place". This is actually quite clever: instead of implementing the whole of the runtime, he only has to implement one feature. It's an insidious ploy, worthy of Microsoft, and RMS is complaining about how Sun isn't letting him do it?
So at the end, here's my point. Sun is worried about two forks. One is the Microsoft fork, and how guaranteed prolifieration can stall adoption of technologies essential for Java's success. Second is the GPL fork, which adopts good features but forces Sun out of the Java business because they can't match the features without GPLing their own code. (BTW, the OpenSolaris license is squarely aimed at that second concern).
To an engineer, McNealy is the real deal, and Schwartz is a PHB. Maybe the Fortune500 CEO crowd would like this change, but I expect Sun would lose the last of their engineering talent were Schwartz to take over. They would become just another company cutting quality to make a buck.
Your first post was great. There's a very famous quote by Thomas Jefferson (I believe, and I'm probably not 100% accurate): "I despise what you have to say, but I will defend to the death your right to say it". For a few brief minutes, I respected you.
And then this crap. Have you read the "fire in a theater" decision? Apparently not, since Oliver rWendell Holmes spent a very long time explaining exactly why the government DID have a RESPONSIBILITY to protect against such speech. Why? Because all rights are in tension; the right to say "fire" (1st amendment) is outweighed by everyone else's right to be safe and secure in the possession of his or her life (4th and 14th amendment, due process clause). The government decided any reasonable person would expect that hearing "fire" in a theater meant an emergency, and the risk of a loss of life outweights the minor loss of freedom of not being able to say that word. And the premise is far from ridiculous: people DIED in the era before fire codes mandated evacuation schemes that work.
Libertarianism has a fundamental flaw: it assumes that anyone who should have responsibility exercises that responsibility rationally. If local governments were perfect, the Civil Rights Act would not have been necessary. Did you by chance notice that local governments aren't perfect? And that the whole point of a federal law (imperfect as it may be) was to give those who wanted a better world leverage against reactionary neandrethals who liked the status quo ante bellum?
I have no problem with communities (not necessarily meaning villages or towns but organizations of people with like apprecations and prejudices) joining together to chat about their beliefs.
Fine. This "community" called the United States of America joined together on principles of tolerance and freedom. And I find it rather arrogant that you waltz through claiming the freedoms you desire while dismissing the responsibilities that make those freedoms possible as "not your problem," that you dismiss laws you don't personally agree with as ridiculous and not worth obeying. If you think the Supreme Court and the laws of this country are wrong, then please leave the country.
Alas, illegal (at least in the US). Mac OS X has some trusted computing code that verifies that it is running on Apple hardware. Bypassing this code is a definite DMCA violation (hence no company will try it). AND Apple assumes a very narrow set of underlying hardware, which isn't what VMs provide - so there is another pile of effort to emulate chipset and so on.
In short, it's not going to happen until Apple wants it to happen.
Principle application: reasoning about multithreaded processes. It's HARD. Lots of state machines (yes, finite automata). Another problem, we wanted to prove a tree doesn't deadlock. (Yes, trees can deadlock when parallelism is insufficient. It sucks.) How do we avoid a deadlock?
2) If they find something, it immediately becomes mail fraud, Al Capone can tell ya about how bad that one is...
Not being involved (mentally involved, not just physically!) is automated. And thus, botting. End of story.
(Being a judge would be so much fun...)
I don't see this as any more novel than someone getting banned for inserting a graphics driver wallhack. He's violating the spirit of the game (no automated character improvements; you must invest your own time) while trying to weasel around the letter of the rules. Blizzard is vague precisely so that they can ban smartasses like this guy!
(All that said, I think Blizzard would be better served inventing some other "punishment" for this. Like, if your character spends 20 minutes attacking a critter it shouldn't be able to kill and it looks like you are a bot, some big SLOW nasty spawns and kills you. If you weren't a bot, you could outrun the nasty... but dumbass bot users die. And maybe take an experience penalty too.)
You will find processor differences though. Move from AMD to Intel, or drop SSE extensions or some such, and things will break.
Hehe - I'm not the one waiting. It takes a month before somebody complains! (Yes, if I heard about it on day one, my life would be so much easier...).
I only know of one problem. You really have to learn by removing a bunch of this crap yourself - new junk hides itself in new ways.
My five-step process:
1) Reboot in safe mode
2) Delete anything in C:\WINDOWS and C:\WINDOWS\SYSTEM32 (or whatever directories of choice) that has a hidden attribute and appeared since "problems began" (usually a month or so).
3) Wipe all temp directories. (that's C:\Documents and settings\username\local settings\temp and \temporary internet files, and maybe others I've forgotten).
4) Use regedit to remove strange Run, RunOnce, etc. entries. If in doubt, google, then destroy. Your user can always reinstall.
5) Reboot into normal Windows, then run a good antivirus and a good adware remover. BEFORE reconnecting to the network. (This may require having virus defs on a USB key).
The anti-spyware seem to get ~80% of what's out there. This gets 95%. Upgrade to the GP's PE environment instead of safe mode, you're probably at 99%. Anything else, transfer files off and reformat, because it's probably a rootkit. With practice, I got the above proceedure down to half an hour during "new computer" season.
7th: accepted, that's what I get for trying to go from memory. Having pulled out my references... PPro through PIII/Xenon are P6 cores, 6th gen. P4 is NetBurst core, 7th gen; NetBurst is the deeper pipeline (higher inflight) and register renaming for major improvements, and some tweaks (deeper btb, adaptive prefetch, increasing ALUs from 5 to 7). Athlon matches the renaming, and beefed up the ALUs to go as fast without the deeper pipeline - yes, all 7th gen. (And now I've found my old notes that say 7th gen is massively OOO, d'oh!)
Re: 8th gen... going from 3-issue to 4-issue isn't going to do much. P6 can't even keep a 3-issue fully saturated; a 4-issue will show some improvement, but far below the 33% theoretical gain. (Yes, Intel claims they'll make 4 IPC under load. I don't believe them. I suspect 4 IPC under ideal optimized load, <2 IPC under real load. How can you make 4 IPC when real code averages a branch every 5 instrs?). And I don't see convoying as a good thing. Itaniums were VLIW, but the compilers still aren't caught up; Merom can't expect compiler support (most compilers now target 5th gen Pentiums, no one will compile for Merom) except for their publicity benchmarks, and HPC users. Convoying looks good in benchmarks, but terrible outside HPC environments. It won't help on the desktop OR the server.
Adding 64-bit is not a mere port. 100+ registers to support renaming, major ISA changes (doubling GP registers and adding 4-level page tables) - the gain from implementing those is astounding. Easily more than a 4-issue will show. Opterons are 8th gen, no doubt about it. P4 seems to smear across 7th and 8th (they had a lot of cores, though no major arch. changes). Merom is a good 8th gen core designed from the ground up; Opteron is a so-so 8th gen core designed from the ground up (first system effect); P4 is a 7th gen core with 8th gen features bolted on.
I would actually claim the real 8th-gen features are x86-64 (the ISA extensions), MATURE register renaming (source of Opteron's performance gain in 32-bit), and good power management (selectively disabling unused parts of the chip). Merom is better than Opteron at these, but evolutionarily so.
And if you claim that a Conroe/Merom is more advanced than a modern Opteron, you REALLY need to look at your processor architectures. It ain't - it's pulling even. Barely.
Generational effects I'm aware of: /P4): uOps-based architecture
3rd (386): virtual memory
4th (486): pipeline
5th (Pentium): superscalar (multiple execution units)
6th (PPro, K6): out-of-order execution
7th (Athlon, P3 (PM)
8th (Opteron, Pentium D): 64-bit
And this is one that looks to be messy for AMD. Intel's VT is more mature than AMD's Pacifica - not by as wide a margin as AMD's dual core beat Intel's, but Intel does have the virtualization edge. And once virtualization takes off - once VMware, Xen, and anyone else can run interesting stuff on generic chips - THEN we will need a whole new batch of processors.
Not immediately - I don't think the current generation of VT designs are good enough yet, I think it will take another iteration. But there is a wide market that Intel is going to hit just a hair before AMD. Which is a shame, because I really like AMD's chips.
Generational thought: Athlons / P3s and P4s were 7th generation, Opterons and 64-bit Pentiums are 8th generation. I don't know what 9th generation will look like yet - and I hope Intel / AMD don't try to claim it prematurely.
You lost me at step two. If other terrorists, connections, etc. have access to bombs, why aren't they blowing up bridges already? Why do they need to find a kook to do the dirty work? Answer: they don't. This whole argument is a rational for harassing people who aren't involved in some sort of plot on the falicious assumption that it would block that same plot.
The argument that a net is not worthwhile only because it tends to catch small fish is silly. There just happen to be more small players, some are just kooks, but a kook with a good connection is a dangerous crazy person.
The problem is that the shrills trying to sell this net claim it catches big fish without killing too many other creatures that get caught in the net. The net doesn't catch big fish (because big fish see it and stay away from it), the owners of the net spend so much time dealing with small fish that they are less effective at catching large fish, and a net fine enough to catch small fish also catches all sorts of other creatures like squid and jellyfish and innocent citizens.
The administration claims this activity makes us safer. Which has just enough truth to be seductive: it does make us safer against the one-in-a-billion kook who could actually do damage. But the whole truth is scarier: protecting against that kook lets the really dangerous folks get away, AND harms bystanders. The administration is too busy patting itself on the back for doing something to notice that the something does more harm than good.
Branch prediction, baby. With accurate branch prediction, CPI (cycles per instr., for those who don't architect processors) drops below 1, because branches are free!
Tongue-in-cheek. Now seriously: I only encounter problems when the wrong word is used. 'grammer' is OK because it isn't a word and thus is an obvious typo (and thus no pipeline stall); 'to'/'too' or 'than'/'then' is a problem. I refuse to correct those as typos because sometimes that word does make sense in context - I prefer to take an uncommon single stall for a wrong word over instead of taking two stalls for a more rare but grammatically correct usage (one stall realizing the corrected word doesn't parse, and a second stall from the pipeline hazard of overriding the correction).
Supported configurations are a Big Deal. The real flaw of this study was that it studied supported configurations under Windows, but included unsupported configurations under Linux. Choosing products by market share is not a justification for such blatant administrative stupidity. An unsupported configuration means potentially more work; that is all this study proves, and we knew that already.
By blaming the lack of adequate equipment in the room for a lack of productivity, the faculty can claim that it is not their fault they are not being effective teachers. Which misses the fundamental point: all the IT-provided bells and whistles will not make someone a better teacher; it takes work, skill, and dedication, just like everything else in life. (I'm not banging on teachers specifically here... we're all just as guilty. I mean, if I had a third development box under my desk I'm sure I'd be much more productive... :-)
The best professors I had in college didn't care whether students were checking e-mail or whatnot during class - they cared about whether their lectures were coherent, sensible, and imparted the desired knowledge to their students. Some of my fellow students used laptops to goof off... some used laptops to be more productive... I honestly don't believe the grade distribution between laptop-users and non-laptop-users is statistically significant.
And by the way, after watching college IT bail out way too many clueless professors who didn't understand technology half as well as they pretended, I have a huge respect for you guys!
Yes I do. I would quite happily pay $250 for an operating system that is (1) stable, (2) well-supported, and (3) part of the general market. Unfortunately, Windows fails at (1), Linux fails at (2) ["visit our forum, fix it yourself" isn't support in my book], and OSX & Solaris fail at (3). Which is a shame; I'm happy to spend my money, but don't see anything worth spending it on.
The students I hear complain about the cost of Photoshop just want a cheap paint program and should go buy Paint Shop Pro ( I look at Adobe and see one of the best pricing structures anywhere in corporate america. Free "viewers" on a non-secret format (PDF), deep discounts for powerful software to people who really need the discount, and reasonable prices for software that's a generation ahead of anything else out there (& don't anyone start a gimp flame-war, you KNOW it's not as powerful as Photoshop) - prices any company that actually needs those features can readily pay.
My experience with international students (while working as staff in a dorm) was that they tended to have pretty uniform educational and class backgrounds. High middle or upper class, often educated at private schools in England (for former British colonies) or at a very small cross-section of schools in their own country. Well-traveled, and mostly "americanized" - or at least very Westernized - to begin with. Apparently, the folks who aren't half americanized already stay in their own country for higher education. See, there's this catch the colleges don't mention that college is much more expensive for international students (less financial aid available), so their "diverse" international communities end up being pretty homogenous socio-economic communities.
So if by diverse, you mean people who take their other religions as seriously as educated American (i.e. not that seriously), speak with a sometimes-thick accent but act American already, then yes. They are diverse. But I saw more diversity and cultural enrichment with the university's efforts to vary socio-economic backgrounds of their American admits. (And for reference, I don't really see a need to change the system.)
Precisely - open source products have a (generally) high setup cost. Closed source products are all over the place, but even Windows has a lower setup cost than most open-source. (Hang on... I'll explain why...)
Your "apt-get install mozilla-firefox" example is convenient - and wrong. You've ignored two costs that are EXTREMELY high: learning to use apt-get and figuring out the name mozilla-firefox. (Cheeky, yes, but bear with me). apt-get is a complex and powerful tool that requires you to set up the right repositories and manually handle dependency conflicts (libc issues...). Simply issuing "install" works if the system is in perfect working order ... it takes SERIOUS time-cost to get it there, or a uselessly uncustomized stock install. The name mozilla-firefox... gee, I thought the program was named firefox! Right now, go to google, start searching for how to install firefox on linux ... first hit I get says download a tarball and run ./firefox_installer. Certainly no apt-get instructions there... and wait, this differs from your instructions, who's right? (reaches for PANIC button). Multiple conflicting install paths, and firefox is one of the easiest Linux installs I've ever done (to be fair, it took me only two days. Because even though firefox was compatable, the installer had a libc conflict with my base install's libsafe. You expect a novice to figure that out?). Compare to MS Windows: I stick in the install CD, and click pretty obvious buttons.
Alright, I'm being facetious ... but I do have a point. To someone who already knows what they need, open source is fine. To someone who doesn't know, who doesn't have mastery of every configuration option and tool they need ... paying extra for closed source (and the polish that goes along with it) is far more economical.
What's the real value of open source alternatives... lets see there's the utility of the programs themselves, the value of having the source code that I can alter myself or pay someone else to alter, that value of having a community that is motivated for various reasons to give free or cheap support, and when necessary paid support. Often you get to talk directly to the developer of the software themselves. What's all that worth?
What's it worth to you? Quite a bit, I take it. To me, who doesn't need custom compilers or custom web browsers and who isn't interested in manually updating and patching the codebase, and who doesn't have the background in this codebase to follow free community-offered help NOR THE TIME TO LEARN THAT BACKGROUND, it has zero value.
And I'm rather offended at open source zealots who insist their code has value to me when I look at the time I'd have to invest to learn it and see negative value.
I don't complain about the existence of free, open-source products - heck, I use many of them and appreciate them! But I am deeply offended at this "open source is the one true way" and "you're either with us or against us" attitude that tells me the work I do is meaningless. Meaningless to you, maybe, but I'm not trying to sell to you. I'm trying to sell to people who appreciate the value of what I do.
(Sorry, you pushed a hot-button here)
The passenger's seat is where your (1) significant other and (2) younger relations ride. So what if commuters usually don't have a passenger and it's safer for them... you'll never get any politician to claim a commuter is more important than a kid. :-)