> And why not? If you want people to do work for free, move to Russia and time travel about half a century back.
Just for the record, I have several former-Soviet-citizen-friends, and they report that they did, indeed, get paid. Ironically, the engineers got paid less than the brick-layers. But they still got paid.
Another way to look at it, though, would be that many got paid for doing no work. It's the quality of work that goes away when there's no incentive.
What's the exception? The enthusiast. In this environment (Slashdot), we are afflicted by a tremendous bias. Because *we* are hobbyists who "would do it anyway", "even if we didn't get paid", we feel that the Open Source movement has a chance. And it does, because it's an industry that can be driven by the few who "would do it anyway".
Bad model for the whole society though. We are *not* representative.
Ok, I've read through the comments, and I'm frustrated with something: Please, everyone, remember that many viruses do not exploit bugs in the operating system! For everyone who says, "I think this is a conflict of interest for MS", please take note that if Microsoft writes a perfect OS, but some user double-clicks on the ZIP file he received through email, it's still doom for his data.
AV solves not just the problems with bug-attacking viruses, but also idiocy-attacking ones. I.e., if your user can't be trusted to have perfect browsing habits, the AV software can help pick up some of the slack.
(a) There's a quote from the article (b) The poster is in the same industry and clearly has a valid opinion (c) The poster lists candidate 'safe' tests, vs the 'unsafe' 'tests' for which the guy got busted (d) The poster expresses his lack of sympathy for what the guy in the article did (e) The poster mentions the conditions under which such activity is valid...
Is there anything about the post that's *not* on-topic?
Whoever meta-moderates this one, I hope you check the context...
Ok, so am I the only one who tried to sign up for a download? No, of course I didn't use my *real* email address. No one's that dumb (my apologies if you have suddenly become a member of the set of dumb people). I guess my "exclusive invite code" of "giveittomenow" just *happened* to be a valid code (I'm such a l33t h4x0r, eh?).
But then, shock of all horrors, it's the most defaultiest rails login app I've ever seen in production! Seems to me they could at least have changed some colors or added a logo (oh, right, they don't have a logo yet... or is it that blue rorschach?)
Well, my apologies if this is a little straightforward, but here goes...
I live in Sweden. Our mainstream media sure talk about piracy alot, but I have never seen them talk about "illegal" trading etc, even if it is against the law.
"illegal P Pronunciation Key (-lgl)
adj.
Prohibited by law.
Prohibited by official rules: an illegal pass in football.
Unacceptable to or not performable by a computer: an illegal operation."
Do they write about "illegal murder", "illegal robbery" etc too?
I assume you are talking about redundancy in these phrases. I.e., that all murder is illegal, so why refer to it as illegal murder? Well... If you look farther up in your post...
...Washington Post etc... "illegal filesharing" this and "illegal piracy" that...
It bears noting that there is filesharing that is completely legal. Regarding piracy -- good point. But don't complain too much. It sounds like you're trying to find something to post about, and you're stretching.
2 or 3 hours max while playing something like WoW?
Well... for WoW there're A/C power adapters. I suppose that would be annoying if you wanted to play WoW at a picnic down by the lake. For me, however, when I was moving and had no phone line (let alone an Internet connection), I was mooching off of Panera and friends for a couple weeks. I was very pleased to find that Panera does provide access to the power grid.
I know that the percentage of Americans who own computers crossed the half-way mark in 2000 -- I don't know what the number is now. But I sure as heckfire know that nowhere near 50% of Americans play WoW (only 1.5M worldwide). It's almost as if there are other things to do with a computer...
Ant Slayer
Re:as usual, blame the users for trying
on
Too Many Passwords
·
· Score: 1
I'd submit the entire organization would function more effectively were they all allowed access to the various systems sans passwords once they'd entered the building.
That's a cool idea except that you are only considering the impact to day-to-day operation. I worked in incident response and computer forensics for a fortune 500 for a few years, and I can provide for you the part you're missing. The vast majority of cases where we got involved were internal folks abusing the system.
If everyone was as honorable as you and I are, then there would be no problems. Heck, then we wouldn't need security on the perimeter. But an alarmingly high number of security compromises happen after someone who works there gets ticked or interested or greedy, etc. I'm not even talking about "espionage" where someone hires on with your company to get at the gooey chocolate center.
You can only let go of access controls on the inside when you can trust everyone on the inside. You may feel offended that you, as an employee, should not be trusted, but alas -- history, case law, and the whole industry of internal investigations indicates that employees cannot be trusted outside of reasonable bounds. A cool enough lawsuit can take a company down.
Most people think that password security is to keep some "criminal" from getting into the company's secrets. Unfortunately, most of the time the "criminal" doesn't come from outside.
I have owned a Compaq R3200 series with a 3200+ for about a year now. One thing I find frustrating is that reviewers rarely run the machine the way I do. I use my box for a LAN party machine (or did, before the integrated GeForce 4 lost sight of the curve), but it's always plugged in. I use it for some computational work, sound processing, and Blendering as a hobby -- but it's always plugged in.
Sure, it's warm, but it's also sitting on my coffee table, my kitchen table, my Panera table, my desk, or my mattress. The heat doesn't really bother me then. And it kicks every Pentium M's butt I've ever played with.
Yet, these reviewers don't do what I do when I unplug -- instead of running Windows benchmarks and all that crap, I run Linux. A simple echo out to/sys/... and I clock it where I want. I turn the brightness down, etc. Lo and behold, a year+ after purchase, I still get about 2.5 hours of battery life (vast approaching the 3 hours I got with a dopey 1.4ghz Pentium M I had at my last company from Dell).
Who needs 2 ghz to pound away in emacs on the train?
What do you mean the premise is all wrong? A couple quotes from the article:
--- " The original question was "Can a spam filter play chess?". Clearly, the answer to this is yes, but making it play well is not so easy."
"Perhaps most interestingly overall, it should be remembered that dbacl doesn't think ahead like most chess engines do. Its successes and failures are almost entirely based on the historical record of the game as it develops and mimicry of training games, not at all on calculating moves and countermoves in the future." ---
It seems to me that the author addresses this issue directly. I.e., the premise is fine -- he taught a bayesian filter to play chess, but not well, and the reason it sucks is because it only plays with history (no prediction).
I think you're mistaken in your assumption that user-friendliness tends to insecurity. You say the following in your post:
"If you make Linux easier to use and more accessible to the general public, it must lose either some of its security lustre, some of its flexibility, or some of both."
Insecurity comes from poor development practices and bad design. User-unfriendliness comes from two things -- one, the assumption that all users are power users, or bad development practices and bad design.
Ultimately, if we were all perfect programmers, no buffer overflows would ever happen because we'd always check our bounds, and input would always be validated on the front, back, and even the middle end. But these decisions have nothing to do with whether or not you can use a discoverable GUI to figure out how to configure something. It doesn't impact whether or not you can copy and paste something -- those are not security problems.
I submit as evidence MacOS X. I recently switched jobs and found myself in a Mac shop, as opposed to my former UNIX and Linux shops. MacOS has, arguably, the user-friendliest interface available. Yet, I can still pull up a terminal and run my for loops in bash. I can compile any geeky code I want.
Hmm. Relative security (not perfect, but certainly comparable to Linux and better than Windows), flexibility, and user-friendliness.
Good design. Good development practices.
Can Linux get there? Sure -- we just need some better thought going into the GUIs and novice-accessibility and discoverability features Linux lacks. Personally, I think a bunch of folks out there are doing a great job with Gnome and friends. Just a few hundred more man-years to go;-).
Then you'll get people trying to plug their RJ-45 cable into the "voice" port of their internal modem -- oops. Or maybe they'll try to go buy an adapter so they can plug it into the wall. Or maybe they'll complain because they have to *unplug* their computer from the cable modem so they can plug in.
Ultimately, we, the Slashdot crowd (egad, am I including myself in the Slashdot crowd?!?), must be careful not to assume that we are a representative sample. Not everyone has broadband Internet connections... even then, what percentage do you think have little routers with RJ-45 ports?
All that said -- I say do both, but hey, I'm not the one trying to fit all those cool chips in that little case =).
My apologies if I am way off base here, but it sounds like web programming means much less to you than it does to me. The great advantage of using a CMS is that I, the 1337 developer-dude, am not responsible for all the content on the site. If I wanted to spend 8 times the time on every project so that there are spiffy, user-friendly ways for the tech-unsavvy in my organization to be able to put up new content, I could do it.
But... somehow... I don't seem to have that sort of motivation... Actually, it's not a motivation thing. Ultimately, my time is worth more than building maintenance infrastructure that eventually starts to look an awful lot like a CMS.
I would be curious to know what this clear evidence is. I have read the literature on both sides of this issue intensively for many years, and it's humorous to me that people say things like "clear evidence" without ever making reference to the actual evidence.
Usually, people will point out the fossil record that it's "clear evidence" that ID can't be correct, yet an intelligent ID proponent will simply point out that the presence of old dead things doesn't mean that there was no designer. The fossil record is not a successful argument.
So what are the real "clear evidences"? This isn't an age-of-the-earth problem, so uranium dating won't help. Instead, this is an information content problem -- consider this helpful analogy: The DNA structure that defines your being is impossibly complex. Why? It's a chain of about 6 billion base pairs. Imagine that each is a bit (when it's actually more like a trit). The odds of generating yours in one shot are 1/2^6E9. If all the atoms in the universe (10^80) were candidate base pairs to produce your DNA, combining into new candidate strands to be you at a rate of one per second (not the 20 years that evolution would require for a generation), the odds of generating your DNA randomnly in impossible conditions in 20 billion years are roughly 1/(2^6E9)/20E9/10E80/6E9, or 1 in 16025317071114407339428943471579329170526278863496 21253942602733418923567123267050964426391033155045 79327369273824029716303716841360696927185673872692 48346434819911785259196081621432543164646371472077 5519919216616685
Ever tried to GDB the kernel? D-Trace is very cool -- its an interface to add "probes" to watch what happens in the kernel. This is most useful for people writing drivers (there's a case study that sort of demonstrates the coolness of it in this context) and for people working on performance problems.
Having developed a bunch of tools close to the kernel, I understand right away what benefit there is in being able to do this stuff in kernel-space.
It's not actually useful to compare it to GDB because it's a completely different world. The bottom line is that no, none of the other OSs you mentioned have anything quite like this. Your third question is quite relevant though -- it's not easy to integrate 86,000 lines of code bastardized from another operating system into another kernel =).
I'm thinking that you may not have read the article. The "proposed" architecture uses what is essentially a G6 (same architecture as the G4/G5) as its "Processing Unit". Adding the vector processing units doesn't change that. OS X should basically run on the G6 and offload streamed processing threads to the vector units.
As the author points out, multimedia operations are already passed off to vector processing units in SSE and Altiva style multimedia instruction pathways with today's architectures. This occurs both on the main CPUs and within the GPUs from Nvidia and ATI.
The new twist is that IBM, Toshiba, and Sony are moving this power into the CPU and taking out a bunch of the old architectural assumptions that have hindered our aging x86 architecture. By pulling memory in larger blocks, addressing it differently, killing the cache and moving its power into private memories for the vector processing units, there's a lot going on to make this move.
My gut feel is that the response to the questions you raise as well as many of the afore-mentioned beefs about how "hard" it is to program for a parallel environment is that it won't be that tough. We already have a multiplicity of processing streams running on our machines -- it's just that one is on the CPU, one is on the GPU, one is in the sound card, etc.
The impact for the games developer or the flashy-GUI-front-end developer for the OS, or the multimedia-content-generation-software developer could conceivably be small. Regular code is serialized in the main processing unit, while well-written libraries spool vectorizable and compartmentalizable tasks off to the vector CPUs.
How many of us actually use our multi-gigahertz CPUs to their potential anyway? The big, demanding stuff we do with our PCs is *already* threaded off to other processors, abstracted in our software world so that we aren't controlling those individual threads of execution. The whole point of the GPU is that we don't have to micromanage the processing of it -- I don't think it's impossible for this technology to eventually filter into many of the computing markets.
It's not a panacea, just a new way of looking at things. Will we be seeing 150 GFLOPS as the author predicts? I *really* doubt it. But we may see some really powerful additions to the flashy and computationally intensive components of our computing experience.
Remember -- vector coprocessors already exist, just in a handicapped form. Cell is just the faith-healer for vector coprocessing units;-).
A fellow Gentoo user, I don't see this fitting into the Gentoo paradigm. Gentoo, from its very beginning, is the opposite of packaged binary software. The initial geek factor with Gentoo was that you could throw in all the wacky compiler optimizations you wanted and build your Linux system from scratch -- very attractive to a small percentage of humans.
What I'm seeing now is that Portage has a couple really cool advantages over other packaging systems, and with those features come a horde of less wacky enthusiasts. Those features are, namely, ease of removal and upgrade and dependancy bliss. Nothing like issuing an "emerge world" and coming back 10 hours later with no hitches.
Even the binary packages you can install through the portage tree are relegated to/opt -- an obvious attempt at some modicum of castigation. I suspect that there will be little impetus at the developer level to move Gentoo in the direction of more precompiled binary applications and away from the "compile everything from source with your own optimizations" model.
It just seems to break the foundational philosophy of the distribution to me.
Hmm... I always found appending 'init=/bin/bash' to be an excellent technique for password recovery. That is, of course, unless someone has hardened LILO or Grub accordingly. Though, based on your narrative, it sounds like this "guy" wasn't very "knowledgeable" in general, so I'm guessing it would work just fine...
The OS is fine for a PDA as it is. What about
hardware improvements? Every Palm I have purchased
since the Palm Pro has been heading in the
direction of crap. My Tungsten T3 isn't nearly as
robust as my T2 was.
Note that PalmSource is the Palm OS software company -- PalmOne is the hardware company. Essentially, you're asking the wrong question. If the PalmOne hardware platform(s) suck, that bodes ill for PalmSource, who is a completely independent company now (read the article).
This move is in their best interests because it frees them from Palm OS's former addiction to a single hardware platform. Now, a hardware device developer will be able to build a mobile device with ARM, MIPS, M68k, Geode, etc. -- anything that runs the Linux kernel.
As previously pointed out, there's even a potential for running this hybrid OS on ipaqs and axims and other PocketPC devices -- beautiful hardware, but why Windows?
Pop that into an OpenOffice.org spreadsheet and look at the graph.
That is not linear, in fact, it's non-linear in the direction that *helps* more and more processors. If the decline from 4 CPUs to 64 CPUs is a mere 1.7% efficiency compared to the 17.5% drop from 2 to 4, then, by golly, I'm going to cram hundreds of CPUs in there and see it tail off. Hello amazing performance.
Instead, reality is that the dynamics change. You can't evaluate "equivalent performance" to a single processor system. There is no reasonable metric with which to do so.
Slashdot Mirror
Somehow I expected you to say that "affordable" was "less than an HD TV".
> And why not? If you want people to do work for free, move to Russia and time travel about half a century back.
Just for the record, I have several former-Soviet-citizen-friends, and they report that they did, indeed, get paid. Ironically, the engineers got paid less than the brick-layers. But they still got paid.
Another way to look at it, though, would be that many got paid for doing no work. It's the quality of work that goes away when there's no incentive.
What's the exception? The enthusiast. In this environment (Slashdot), we are afflicted by a tremendous bias. Because *we* are hobbyists who "would do it anyway", "even if we didn't get paid", we feel that the Open Source movement has a chance. And it does, because it's an industry that can be driven by the few who "would do it anyway".
Bad model for the whole society though. We are *not* representative.
-Ant Slayer-
Ok, I've read through the comments, and I'm frustrated with something: Please, everyone, remember that many viruses do not exploit bugs in the operating system! For everyone who says, "I think this is a conflict of interest for MS", please take note that if Microsoft writes a perfect OS, but some user double-clicks on the ZIP file he received through email, it's still doom for his data.
AV solves not just the problems with bug-attacking viruses, but also idiocy-attacking ones. I.e., if your user can't be trusted to have perfect browsing habits, the AV software can help pick up some of the slack.
-Ant Slayer-
Offtopic? Huh?
(a) There's a quote from the article
(b) The poster is in the same industry and clearly has a valid opinion
(c) The poster lists candidate 'safe' tests, vs the 'unsafe' 'tests' for which the guy got busted
(d) The poster expresses his lack of sympathy for what the guy in the article did
(e) The poster mentions the conditions under which such activity is valid...
Is there anything about the post that's *not* on-topic?
Whoever meta-moderates this one, I hope you check the context...
-Ant Slayer-
$ find . -type f | xargs perl -p -i.bak -e "s/(openoffice)\.org/\1/ig;"
Ant Slayer
Ok, so am I the only one who tried to sign up for a download? No, of course I didn't use my *real* email address. No one's that dumb (my apologies if you have suddenly become a member of the set of dumb people). I guess my "exclusive invite code" of "giveittomenow" just *happened* to be a valid code (I'm such a l33t h4x0r, eh?).
But then, shock of all horrors, it's the most defaultiest rails login app I've ever seen in production! Seems to me they could at least have changed some colors or added a logo (oh, right, they don't have a logo yet... or is it that blue rorschach?)
-Ant Slayer-
Well, my apologies if this is a little straightforward, but here goes...
I live in Sweden. Our mainstream media sure talk about piracy alot, but I have never seen them talk about "illegal" trading etc, even if it is against the law.
From American Heritage:
"illegal P Pronunciation Key (-lgl)
adj.
Prohibited by law.
Prohibited by official rules: an illegal pass in football.
Unacceptable to or not performable by a computer: an illegal operation."
Do they write about "illegal murder", "illegal robbery" etc too?
I assume you are talking about redundancy in these phrases. I.e., that all murder is illegal, so why refer to it as illegal murder? Well... If you look farther up in your post...
It bears noting that there is filesharing that is completely legal. Regarding piracy -- good point. But don't complain too much. It sounds like you're trying to find something to post about, and you're stretching.
Ant Slayer
Correction... 4 million.
Ant Slayer
2 or 3 hours max while playing something like WoW?
Well... for WoW there're A/C power adapters. I suppose that would be annoying if you wanted to play WoW at a picnic down by the lake. For me, however, when I was moving and had no phone line (let alone an Internet connection), I was mooching off of Panera and friends for a couple weeks. I was very pleased to find that Panera does provide access to the power grid.
I know that the percentage of Americans who own computers crossed the half-way mark in 2000 -- I don't know what the number is now. But I sure as heckfire know that nowhere near 50% of Americans play WoW (only 1.5M worldwide). It's almost as if there are other things to do with a computer...
Ant Slayer
I'd submit the entire organization would function more effectively were they all allowed access to the various systems sans passwords once they'd entered the building.
That's a cool idea except that you are only considering the impact to day-to-day operation. I worked in incident response and computer forensics for a fortune 500 for a few years, and I can provide for you the part you're missing. The vast majority of cases where we got involved were internal folks abusing the system.
If everyone was as honorable as you and I are, then there would be no problems. Heck, then we wouldn't need security on the perimeter. But an alarmingly high number of security compromises happen after someone who works there gets ticked or interested or greedy, etc. I'm not even talking about "espionage" where someone hires on with your company to get at the gooey chocolate center.
You can only let go of access controls on the inside when you can trust everyone on the inside. You may feel offended that you, as an employee, should not be trusted, but alas -- history, case law, and the whole industry of internal investigations indicates that employees cannot be trusted outside of reasonable bounds. A cool enough lawsuit can take a company down.
Most people think that password security is to keep some "criminal" from getting into the company's secrets. Unfortunately, most of the time the "criminal" doesn't come from outside.
-Ant Slayer-Let us tie police records and social security numbers to mac addresses.
Alright... people will finally start *noticing* my MAC addresses then...
00:de:ad:be:ef:00
be:dd:ed:00:ba:be
etc.
-Ant Slayer-
I have owned a Compaq R3200 series with a 3200+ for about a year now. One thing I find frustrating is that reviewers rarely run the machine the way I do. I use my box for a LAN party machine (or did, before the integrated GeForce 4 lost sight of the curve), but it's always plugged in. I use it for some computational work, sound processing, and Blendering as a hobby -- but it's always plugged in.
/sys/... and I clock it where I want. I turn the brightness down, etc. Lo and behold, a year+ after purchase, I still get about 2.5 hours of battery life (vast approaching the 3 hours I got with a dopey 1.4ghz Pentium M I had at my last company from Dell).
Sure, it's warm, but it's also sitting on my coffee table, my kitchen table, my Panera table, my desk, or my mattress. The heat doesn't really bother me then. And it kicks every Pentium M's butt I've ever played with.
Yet, these reviewers don't do what I do when I unplug -- instead of running Windows benchmarks and all that crap, I run Linux. A simple echo out to
Who needs 2 ghz to pound away in emacs on the train?
-Ant Slayer-
What do you mean the premise is all wrong? A couple quotes from the article:
---
" The original question was "Can a spam filter play chess?". Clearly, the answer to this is yes, but making it play well is not so easy."
"Perhaps most interestingly overall, it should be remembered that dbacl doesn't think ahead like most chess engines do. Its successes and failures are almost entirely based on the historical record of the game as it develops and mimicry of training games, not at all on calculating moves and countermoves in the future."
---
It seems to me that the author addresses this issue directly. I.e., the premise is fine -- he taught a bayesian filter to play chess, but not well, and the reason it sucks is because it only plays with history (no prediction).
-Ant Slayer-
All I need is a DVI recording device, eh? Sounds good to me.
-Ant-Slayer-
I think you're mistaken in your assumption that user-friendliness tends to insecurity. You say the following in your post:
;-).
"If you make Linux easier to use and more accessible to the general public, it must lose either some of its security lustre, some of its flexibility, or some of both."
Insecurity comes from poor development practices and bad design. User-unfriendliness comes from two things -- one, the assumption that all users are power users, or bad development practices and bad design.
Ultimately, if we were all perfect programmers, no buffer overflows would ever happen because we'd always check our bounds, and input would always be validated on the front, back, and even the middle end. But these decisions have nothing to do with whether or not you can use a discoverable GUI to figure out how to configure something. It doesn't impact whether or not you can copy and paste something -- those are not security problems.
I submit as evidence MacOS X. I recently switched jobs and found myself in a Mac shop, as opposed to my former UNIX and Linux shops. MacOS has, arguably, the user-friendliest interface available. Yet, I can still pull up a terminal and run my for loops in bash. I can compile any geeky code I want.
Hmm. Relative security (not perfect, but certainly comparable to Linux and better than Windows), flexibility, and user-friendliness.
Good design. Good development practices.
Can Linux get there? Sure -- we just need some better thought going into the GUIs and novice-accessibility and discoverability features Linux lacks. Personally, I think a bunch of folks out there are doing a great job with Gnome and friends. Just a few hundred more man-years to go
-Ant Slayer-
Then you'll get people trying to plug their RJ-45 cable into the "voice" port of their internal modem -- oops. Or maybe they'll try to go buy an adapter so they can plug it into the wall. Or maybe they'll complain because they have to *unplug* their computer from the cable modem so they can plug in.
Ultimately, we, the Slashdot crowd (egad, am I including myself in the Slashdot crowd?!?), must be careful not to assume that we are a representative sample. Not everyone has broadband Internet connections... even then, what percentage do you think have little routers with RJ-45 ports?
All that said -- I say do both, but hey, I'm not the one trying to fit all those cool chips in that little case =).
Ant Slayer
My apologies if I am way off base here, but it sounds like web programming means much less to you than it does to me. The great advantage of using a CMS is that I, the 1337 developer-dude, am not responsible for all the content on the site. If I wanted to spend 8 times the time on every project so that there are spiffy, user-friendly ways for the tech-unsavvy in my organization to be able to put up new content, I could do it.
But... somehow... I don't seem to have that sort of motivation... Actually, it's not a motivation thing. Ultimately, my time is worth more than building maintenance infrastructure that eventually starts to look an awful lot like a CMS.
-Ant Slayer-
I would be curious to know what this clear evidence is. I have read the literature on both sides of this issue intensively for many years, and it's humorous to me that people say things like "clear evidence" without ever making reference to the actual evidence.
6 21253942602733418923567123267050964426391033155045 79327369273824029716303716841360696927185673872692 48346434819911785259196081621432543164646371472077 5519919216616685
Usually, people will point out the fossil record that it's "clear evidence" that ID can't be correct, yet an intelligent ID proponent will simply point out that the presence of old dead things doesn't mean that there was no designer. The fossil record is not a successful argument.
So what are the real "clear evidences"? This isn't an age-of-the-earth problem, so uranium dating won't help. Instead, this is an information content problem -- consider this helpful analogy: The DNA structure that defines your being is impossibly complex. Why? It's a chain of about 6 billion base pairs. Imagine that each is a bit (when it's actually more like a trit). The odds of generating yours in one shot are 1/2^6E9. If all the atoms in the universe (10^80) were candidate base pairs to produce your DNA, combining into new candidate strands to be you at a rate of one per second (not the 20 years that evolution would require for a generation), the odds of generating your DNA randomnly in impossible conditions in 20 billion years are roughly 1/(2^6E9)/20E9/10E80/6E9, or 1 in 1602531707111440733942894347157932917052627886349
That's dang improbable.
-Ant Slayer-
Ever tried to GDB the kernel? D-Trace is very cool -- its an interface to add "probes" to watch what happens in the kernel. This is most useful for people writing drivers (there's a case study that sort of demonstrates the coolness of it in this context) and for people working on performance problems.
Having developed a bunch of tools close to the kernel, I understand right away what benefit there is in being able to do this stuff in kernel-space.
It's not actually useful to compare it to GDB because it's a completely different world. The bottom line is that no, none of the other OSs you mentioned have anything quite like this. Your third question is quite relevant though -- it's not easy to integrate 86,000 lines of code bastardized from another operating system into another kernel =).
-Ant Slayer-
I'm thinking that you may not have read the article. The "proposed" architecture uses what is essentially a G6 (same architecture as the G4/G5) as its "Processing Unit". Adding the vector processing units doesn't change that. OS X should basically run on the G6 and offload streamed processing threads to the vector units.
;-).
As the author points out, multimedia operations are already passed off to vector processing units in SSE and Altiva style multimedia instruction pathways with today's architectures. This occurs both on the main CPUs and within the GPUs from Nvidia and ATI.
The new twist is that IBM, Toshiba, and Sony are moving this power into the CPU and taking out a bunch of the old architectural assumptions that have hindered our aging x86 architecture. By pulling memory in larger blocks, addressing it differently, killing the cache and moving its power into private memories for the vector processing units, there's a lot going on to make this move.
My gut feel is that the response to the questions you raise as well as many of the afore-mentioned beefs about how "hard" it is to program for a parallel environment is that it won't be that tough. We already have a multiplicity of processing streams running on our machines -- it's just that one is on the CPU, one is on the GPU, one is in the sound card, etc.
The impact for the games developer or the flashy-GUI-front-end developer for the OS, or the multimedia-content-generation-software developer could conceivably be small. Regular code is serialized in the main processing unit, while well-written libraries spool vectorizable and compartmentalizable tasks off to the vector CPUs.
How many of us actually use our multi-gigahertz CPUs to their potential anyway? The big, demanding stuff we do with our PCs is *already* threaded off to other processors, abstracted in our software world so that we aren't controlling those individual threads of execution. The whole point of the GPU is that we don't have to micromanage the processing of it -- I don't think it's impossible for this technology to eventually filter into many of the computing markets.
It's not a panacea, just a new way of looking at things. Will we be seeing 150 GFLOPS as the author predicts? I *really* doubt it. But we may see some really powerful additions to the flashy and computationally intensive components of our computing experience.
Remember -- vector coprocessors already exist, just in a handicapped form. Cell is just the faith-healer for vector coprocessing units
-Ant Slayer-
A fellow Gentoo user, I don't see this fitting into the Gentoo paradigm. Gentoo, from its very beginning, is the opposite of packaged binary software. The initial geek factor with Gentoo was that you could throw in all the wacky compiler optimizations you wanted and build your Linux system from scratch -- very attractive to a small percentage of humans.
/opt -- an obvious attempt at some modicum of castigation. I suspect that there will be little impetus at the developer level to move Gentoo in the direction of more precompiled binary applications and away from the "compile everything from source with your own optimizations" model.
What I'm seeing now is that Portage has a couple really cool advantages over other packaging systems, and with those features come a horde of less wacky enthusiasts. Those features are, namely, ease of removal and upgrade and dependancy bliss. Nothing like issuing an "emerge world" and coming back 10 hours later with no hitches.
Even the binary packages you can install through the portage tree are relegated to
It just seems to break the foundational philosophy of the distribution to me.
-Ant Slayer-
Hmm... I always found appending 'init=/bin/bash' to be an excellent technique for password recovery. That is, of course, unless someone has hardened LILO or Grub accordingly. Though, based on your narrative, it sounds like this "guy" wasn't very "knowledgeable" in general, so I'm guessing it would work just fine...
-Ant Slayer-
"Intelegent"?
-Ant Slayer-
Note that PalmSource is the Palm OS software company -- PalmOne is the hardware company. Essentially, you're asking the wrong question. If the PalmOne hardware platform(s) suck, that bodes ill for PalmSource, who is a completely independent company now (read the article).
This move is in their best interests because it frees them from Palm OS's former addiction to a single hardware platform. Now, a hardware device developer will be able to build a mobile device with ARM, MIPS, M68k, Geode, etc. -- anything that runs the Linux kernel.
As previously pointed out, there's even a potential for running this hybrid OS on ipaqs and axims and other PocketPC devices -- beautiful hardware, but why Windows?
-Ant Slayer-
My apologies, but I couldn't help but think that you'd be *really* lucky to get 50x out of 64 CPUs. Examine the following:
1 CPU @ 1.00x -> 1.00 / 1 = 1.000
2 CPUs @ 1.95x -> 1.95 / 2 = 0.975
4 CPUs @ 3.20x -> 3.20 / 4 = 0.800
64 CPUs @ 50.0x -> 50.0 / 64 = 0.783
Pop that into an OpenOffice.org spreadsheet and look at the graph.
That is not linear, in fact, it's non-linear in the direction that *helps* more and more processors. If the decline from 4 CPUs to 64 CPUs is a mere 1.7% efficiency compared to the 17.5% drop from 2 to 4, then, by golly, I'm going to cram hundreds of CPUs in there and see it tail off. Hello amazing performance.
Instead, reality is that the dynamics change. You can't evaluate "equivalent performance" to a single processor system. There is no reasonable metric with which to do so.
-Ant Slayer-