Rackmount cases are expensive. I've never seen one for less than $250. Meanwhile, you can get cheap minitower cases for maybe $50 (or less) and just stick them all on a $40 utility shelf system from Home Depot. (Granted, this might be a bad idea in, say, California:-) )
I think if I was building a cluster system that would be the way I'd go unless I had to impress people with photographs (though undoubtedly row after row of G4 boxen would do that equally easily:-) ). Racks cost big money.
No one ever did, and Mac users are a loyal bunch. The hockey puck was one of the worst mouse designs Apple ever cooked up, just as the current optical mouse is one of the best (I still won't use one, though -- I need two buttons).
I rather like your statement about old-line Mac users, though we Mac folks are inclined to consider IBM the good guys these days:-)
Those AppleSeeds are actually pretty nice hardware; I've read the technical report on the original and it's actually a better design than a Beowulf. MacOS Classic's cooperative multitasking is actually a plus in this arena as it means that your computation code can completely take over the system and grind away. If you really want to have some fun, though, write a lightweight bootloader that emulates just enough MacOS and MPI to get the job done; you don't need the GUI overhead for a cluster anyway.
I would make two changes to the design, though: I'd use OS X as a front-end system (since you're not doing much computation on the coordinating server, I'm sure) and I'd start calling the clusters Apple *Orchards*...
It's not the religion that anti-CO$ people are protesting, though it has more than its share of lunacies. It's the organization itself, which has been spared the full force of the law only by behaving in such a reflexively litigious manner (as per LRH's instructions years ago) that nobody really wants to screw with it.
Scientology investigators and lawyers are very good at their job (which mostly involves creating smear campaigns and running their opponents down in court; "black ops" are suspected to have occured but the evidence is sketchy). The organization is a hall of mirrors -- the real hierarchy doesn't much have anything to do with the way the CO$ and its related organizations present themselves, which I think would probably make declaring the Church a vexatious litigant (which it most certainly is) a bit pointless.
Essentially, what the Church's critics want is the elimination of the organization itself, with those still choosing to adhere to Hubbard's teachings practicing them on their own terms, and not the cult's. It would be a little like saying "Catholicism is a great religion, it's the Church I have a problem with" (I don't agree with that, btw) or something vaguely similar.
FWIW, however, it's not that easy to get someone away from the Church -- from the very beginning they hit you with hypnotic techniques to manufacture loyalty.
Not really -- "squander" means to waste something (usually money) in English, but "disipador" is directly cognate to "dissipate" which is what a heat sink does in English.
But you may have come up with one of the cooler mistranslations I've ever seen; I like the idea of hooking up a "heat squanderer" to my hardware. It's quite poetic. (Besides, it's not like you can do anything useful with the excess heat anyway:-) )
It occurs to me -- they do have those low-profile (NLX?) cases out there; I'm typing on a computer that uses one right now. Get one of those, spray-paint it black, and you can fit it into your entertainment system. Alternately, just cram it into a plain old external slimline case and plug it in...
It's a waste of time. Really. I've had one die on me (damaged cupholder track, nothing real interesting) and it's just not worth trying to find spare parts for something you can replace for $40.
BTW, there's an interesting choice of terminology (squanderer = heat sink? Como se dice "heat sink" en espan~ol?) in there... actually a pretty cool word for it, though./Brian
Well, as far as Sony and MS are concerned, we can only hope that the same market forces that destroyed DivX will hold in this case. I think it will certainly prevent software subscriptions and networked authentication from taking off, but I don't know if it will prevent trouble on the hardware level. But anyway...
Don't forget about cell phones that even if it seems like a good idea at the time, there are serious security risks in Open Source cell phone drivers; I'd say this is a pretty clear-cut (and rare) case of Open Source being more trouble than it's worth in some circumstances. (Cell phone fraud is a big industry that caters to some pretty shady characters, in case you have no clue why I have a problem with this.)
The problem with Nokia's idea is not that it's a bad one -- it's not -- but if it's not focused on games, than what is it focused on? They need to hire some very good technical marketing people to make this the killer app it needs to be; focusing on the hack value of an open source games^h^h^h^h^hmedia console is going to sell a lot of boxes to the/. crowd, but I promise you that you are not going to get a critical mass of Joe Sixpacks just by slapping Tux onto the front of the box and flashing Quake into ROM (or however it might get in there).
Certainly Nokia is in a better position to do this than Indrema could ever dream of being. They're a big company, based in one of the most wired countries in the world, and they know enough about consumer marketing. I say I'll buy one when it hits the light of day. Question is, who else can they sign up, because at the end of the day it's still about games over everything else. (And they better have the best damn Snake game ever created on this thing:-) )
See, you're looking at three of the scariest companies out there. Sony and AOLTW make massive amounts of money off of being gigantic conflicts of interest, and Microsoft is... Microsoft. I honestly don't know which I find scarier.
(And no, I don't like Sony at all. I might buy a PlayStation2 from them, but I have no plans to ever buy, say, a Walkman again.)
The thing is that you can plug Linux all you want but a cute penguin isn't going to be much of a selling point for Joe Blow in his dorm room (unless Mr. Blow happens to be a geek).
Nokia does, I think, have the clout to pull off what Indrema couldn't. They've got money and an immensely popular product (and one could say they're already in the electronic games industry -- how many expert Snake players we got here?). Here's what Nokia *could* deliver:
-A decidedly bad-ass game system with included crackability. I'd be amazed if the technology involved was radically different from the Xbox.
-A ready-to-network cheap processing node a la that PS2-ish thing that Sony's marketing as a cheap supercomputer (I'd love to see a... Oh, wait, it already is...)
-A system with easy wireless connectivity -- just patch it into your cell phone. No worries about Bluetooth or 802.11...
But they need to have a killer app for it, and a cell-phone base station won't be it. Since the system will be open-source, it's a safe bet they won't be making their money off of runtime licensing. They can't market it like an ordinary game console because they won't be able to loss-leader it. Ever. I think it can be done, but I'm not quite sure how.
See, whatcha do is download the TiVo source code and then figure out how to get it to deal with two video-in streams at once. Then you've got UltimateTV for Linux.
Depends on whose definition of PC you use. Are you talking about a personal computer (the current Jobsian definition, which is not too uncommon and includes pretty much everything desktoppish and related/developed from), or a PC, which is indeed x86-powered.
The PC in PowerPC indicated that it was a POWER chip designed for personal computing platforms like the Mac.
What could be done is to just take one of the BSDs, shave it down to a microkernel (wait, isn't this how Avie and co. created Mach way back when?), and slap a Win32 interface on it. Use the Qt framebuffer, and you're good to go...
Motorola has a mobo called the Sandpoint that seems to be the closest thing to a buildable system, but it's bare-bones by any standard -- not even USB. That's fine, you can add the stuff with a PCI card, but...
The problem with Sandpoint is expandability -- it takes a processor daughtercard, of which something like half a dozen are available, all with extremely limited (*one* SO-DIMM) memory expansion. You can put a fairly nice processor in them (G4 boards are available), and it's ATX format, so you can put it in a standard case. But you can't expand the memory without swapping what's already in there. It's put out as an evaluation board, so *maybe* that excuses the lack of current features, but I would think it's a bit difficult to evaluate a processor on such limited system.
Okay, how do I know all this? I'm thinking of building a Sandpoint box myself (can't find @!#$* pricing) to do a little OS hacking (Darwin, if you must know;-) ) and I thought the fruits of my research thus far would be useful. Sandpoint will probably suit my purposes (it better; it's the only game in town as far as I can tell), but it's not the system the market needs.
I hate to say it, but what we really need is Mac cloning again. No one, and I mean NO ONE, is going to be crazy enough to come out with a commodity PPC motherboard without it being able to boot up (at the very least) MacOS X. For a while, until the clones were Steved, Motorola sold a 603e board called Tanzania. It was a bit flaky by Mac hardware standards (Apple sold only one Tanzania machine, the 4400/7250 by name, depending on where you bought it), but it did make clone sublicensing and building your own Mac possible.
IMHO the only way to get commodity PPC hardware is to force Apple's hand on the cloning issue by porting Darwin to systems like Sandpoint and RS/6000. Once that works (after which point we can all start buying the IBM chips, which run faster than the Motorola chips), we'll have what's needed. But we need the clones, either from the top down (which ain't gonna happen soon) or the bottom up.
Okay... patenting something that was originally developed under GPL.
Is that legal? (Okay, it probably is, but it makes no sense.)
More to the point, I can't see as how they'd have grounds for it, since GPLing something does seem to essentially place the algorithms in question in the public domain...
So what it really comes down to is whether we're dealing with a fraudulent patent or not.
First off, I can't believe there isn't prior art here -- IANAPL but I don't think there's anything too original about a CD database, or even a CD database remotely accessible (you're going to tell me that there isn't a copy of an LP database from 1979 floating around on a tape in Murray Hill or someplace like that somewhere? I don't bloody think so...) Hell, this is something somebody could have written using Filemaker or Access ten years ago and it would have infringed on the patent!
Let's be at least marginally realistic: this isn't capitalism, this is economic fascism, where companies take advantage of privilege to lock out competition. At the risk of invoking Godwin's Law, isn't this how the German economy ran under the Nazis?
Slashdot Mirror
Rackmount cases are expensive. I've never seen one for less than $250. Meanwhile, you can get cheap minitower cases for maybe $50 (or less) and just stick them all on a $40 utility shelf system from Home Depot. (Granted, this might be a bad idea in, say, California :-) )
:-) ). Racks cost big money.
I think if I was building a cluster system that would be the way I'd go unless I had to impress people with photographs (though undoubtedly row after row of G4 boxen would do that equally easily
/Brian
No one ever did, and Mac users are a loyal bunch. The hockey puck was one of the worst mouse designs Apple ever cooked up, just as the current optical mouse is one of the best (I still won't use one, though -- I need two buttons).
/Brian
I'd love to see a Beowulf of... oh, wait. Never mind.
/Brian
I rather like your statement about old-line Mac users, though we Mac folks are inclined to consider IBM the good guys these days :-)
Those AppleSeeds are actually pretty nice hardware; I've read the technical report on the original and it's actually a better design than a Beowulf. MacOS Classic's cooperative multitasking is actually a plus in this arena as it means that your computation code can completely take over the system and grind away. If you really want to have some fun, though, write a lightweight bootloader that emulates just enough MacOS and MPI to get the job done; you don't need the GUI overhead for a cluster anyway.
I would make two changes to the design, though: I'd use OS X as a front-end system (since you're not doing much computation on the coordinating server, I'm sure) and I'd start calling the clusters Apple *Orchards*...
/Brian
It's not the religion that anti-CO$ people are protesting, though it has more than its share of lunacies. It's the organization itself, which has been spared the full force of the law only by behaving in such a reflexively litigious manner (as per LRH's instructions years ago) that nobody really wants to screw with it.
Scientology investigators and lawyers are very good at their job (which mostly involves creating smear campaigns and running their opponents down in court; "black ops" are suspected to have occured but the evidence is sketchy). The organization is a hall of mirrors -- the real hierarchy doesn't much have anything to do with the way the CO$ and its related organizations present themselves, which I think would probably make declaring the Church a vexatious litigant (which it most certainly is) a bit pointless.
Essentially, what the Church's critics want is the elimination of the organization itself, with those still choosing to adhere to Hubbard's teachings practicing them on their own terms, and not the cult's. It would be a little like saying "Catholicism is a great religion, it's the Church I have a problem with" (I don't agree with that, btw) or something vaguely similar.
FWIW, however, it's not that easy to get someone away from the Church -- from the very beginning they hit you with hypnotic techniques to manufacture loyalty.
/Brian
Not really -- "squander" means to waste something (usually money) in English, but "disipador" is directly cognate to "dissipate" which is what a heat sink does in English.
:-) )
But you may have come up with one of the cooler mistranslations I've ever seen; I like the idea of hooking up a "heat squanderer" to my hardware. It's quite poetic. (Besides, it's not like you can do anything useful with the excess heat anyway
/Brian
Silly me, responding to my own post, but...
It occurs to me -- they do have those low-profile (NLX?) cases out there; I'm typing on a computer that uses one right now. Get one of those, spray-paint it black, and you can fit it into your entertainment system. Alternately, just cram it into a plain old external slimline case and plug it in...
/Brian
I knew this had to be *possible*; I never knew anyone had tried it, though...
:-)
Unfortunately you can't just wire up a CD-ROM to a CD-RW without that pesky motherboard getting in the way
/Brian
It's a waste of time. Really. I've had one die on me (damaged cupholder track, nothing real interesting) and it's just not worth trying to find spare parts for something you can replace for $40. BTW, there's an interesting choice of terminology (squanderer = heat sink? Como se dice "heat sink" en espan~ol?) in there... actually a pretty cool word for it, though. /Brian
Black paint.
HTH,
/Brian
Well, as far as Sony and MS are concerned, we can only hope that the same market forces that destroyed DivX will hold in this case. I think it will certainly prevent software subscriptions and networked authentication from taking off, but I don't know if it will prevent trouble on the hardware level. But anyway...
/. crowd, but I promise you that you are not going to get a critical mass of Joe Sixpacks just by slapping Tux onto the front of the box and flashing Quake into ROM (or however it might get in there).
:-) )
Don't forget about cell phones that even if it seems like a good idea at the time, there are serious security risks in Open Source cell phone drivers; I'd say this is a pretty clear-cut (and rare) case of Open Source being more trouble than it's worth in some circumstances. (Cell phone fraud is a big industry that caters to some pretty shady characters, in case you have no clue why I have a problem with this.)
The problem with Nokia's idea is not that it's a bad one -- it's not -- but if it's not focused on games, than what is it focused on? They need to hire some very good technical marketing people to make this the killer app it needs to be; focusing on the hack value of an open source games^h^h^h^h^hmedia console is going to sell a lot of boxes to the
Certainly Nokia is in a better position to do this than Indrema could ever dream of being. They're a big company, based in one of the most wired countries in the world, and they know enough about consumer marketing. I say I'll buy one when it hits the light of day. Question is, who else can they sign up, because at the end of the day it's still about games over everything else. (And they better have the best damn Snake game ever created on this thing
/Brian
Uh... MS is pretty obsessive about that digital-signing thing. I suspect they'd probably do everything they possibly can to lock out the unauthorized.
/Brian
I still don't quite get WinCE on the Dreamcast anyway -- it seems it wasn't widely used, so did it have any benefits to Average Joe Dreamcast Hacker?
/Brian
See, you're looking at three of the scariest companies out there. Sony and AOLTW make massive amounts of money off of being gigantic conflicts of interest, and Microsoft is... Microsoft. I honestly don't know which I find scarier.
(And no, I don't like Sony at all. I might buy a PlayStation2 from them, but I have no plans to ever buy, say, a Walkman again.)
/Brian
I can't see how MS intends to get any serious penetration anyway -- is it still only available with a satellite dish?
/Brian
The thing is that you can plug Linux all you want but a cute penguin isn't going to be much of a selling point for Joe Blow in his dorm room (unless Mr. Blow happens to be a geek).
Nokia does, I think, have the clout to pull off what Indrema couldn't. They've got money and an immensely popular product (and one could say they're already in the electronic games industry -- how many expert Snake players we got here?). Here's what Nokia *could* deliver:
-A decidedly bad-ass game system with included crackability. I'd be amazed if the technology involved was radically different from the Xbox.
-A ready-to-network cheap processing node a la that PS2-ish thing that Sony's marketing as a cheap supercomputer (I'd love to see a... Oh, wait, it already is...)
-A system with easy wireless connectivity -- just patch it into your cell phone. No worries about Bluetooth or 802.11...
But they need to have a killer app for it, and a cell-phone base station won't be it. Since the system will be open-source, it's a safe bet they won't be making their money off of runtime licensing. They can't market it like an ordinary game console because they won't be able to loss-leader it. Ever. I think it can be done, but I'm not quite sure how.
But I'd buy one.
/Brian
See, whatcha do is download the TiVo source code and then figure out how to get it to deal with two video-in streams at once. Then you've got UltimateTV for Linux.
/Brian
You want to have some real fun, try Ron Nicholson's Chipmunk Basic. It has OO features; I even tried to build a class library with it once.
/Brian
Losing Douglas Adams is like losing Jim Henson. Too soon, too great a person to keep around.
See you at Milliways...
/Brian
Depends on whose definition of PC you use. Are you talking about a personal computer (the current Jobsian definition, which is not too uncommon and includes pretty much everything desktoppish and related/developed from), or a PC, which is indeed x86-powered.
The PC in PowerPC indicated that it was a POWER chip designed for personal computing platforms like the Mac.
/Brian
What could be done is to just take one of the BSDs, shave it down to a microkernel (wait, isn't this how Avie and co. created Mach way back when?), and slap a Win32 interface on it. Use the Qt framebuffer, and you're good to go...
/Brian
Motorola has a mobo called the Sandpoint that seems to be the closest thing to a buildable system, but it's bare-bones by any standard -- not even USB. That's fine, you can add the stuff with a PCI card, but...
;-) ) and I thought the fruits of my research thus far would be useful. Sandpoint will probably suit my purposes (it better; it's the only game in town as far as I can tell), but it's not the system the market needs.
The problem with Sandpoint is expandability -- it takes a processor daughtercard, of which something like half a dozen are available, all with extremely limited (*one* SO-DIMM) memory expansion. You can put a fairly nice processor in them (G4 boards are available), and it's ATX format, so you can put it in a standard case. But you can't expand the memory without swapping what's already in there. It's put out as an evaluation board, so *maybe* that excuses the lack of current features, but I would think it's a bit difficult to evaluate a processor on such limited system.
Okay, how do I know all this? I'm thinking of building a Sandpoint box myself (can't find @!#$* pricing) to do a little OS hacking (Darwin, if you must know
I hate to say it, but what we really need is Mac cloning again. No one, and I mean NO ONE, is going to be crazy enough to come out with a commodity PPC motherboard without it being able to boot up (at the very least) MacOS X. For a while, until the clones were Steved, Motorola sold a 603e board called Tanzania. It was a bit flaky by Mac hardware standards (Apple sold only one Tanzania machine, the 4400/7250 by name, depending on where you bought it), but it did make clone sublicensing and building your own Mac possible.
IMHO the only way to get commodity PPC hardware is to force Apple's hand on the cloning issue by porting Darwin to systems like Sandpoint and RS/6000. Once that works (after which point we can all start buying the IBM chips, which run faster than the Motorola chips), we'll have what's needed. But we need the clones, either from the top down (which ain't gonna happen soon) or the bottom up.
/Brian
Okay... patenting something that was originally developed under GPL.
Is that legal? (Okay, it probably is, but it makes no sense.)
More to the point, I can't see as how they'd have grounds for it, since GPLing something does seem to essentially place the algorithms in question in the public domain...
/Brian
So what it really comes down to is whether we're dealing with a fraudulent patent or not.
First off, I can't believe there isn't prior art here -- IANAPL but I don't think there's anything too original about a CD database, or even a CD database remotely accessible (you're going to tell me that there isn't a copy of an LP database from 1979 floating around on a tape in Murray Hill or someplace like that somewhere? I don't bloody think so...) Hell, this is something somebody could have written using Filemaker or Access ten years ago and it would have infringed on the patent!
Let's be at least marginally realistic: this isn't capitalism, this is economic fascism, where companies take advantage of privilege to lock out competition. At the risk of invoking Godwin's Law, isn't this how the German economy ran under the Nazis?
/Brian
Troll.
Besides, how else would you do it? That's the Unix way is to cruft things together.
/Brian