IBM is committed to selling linux. This means what IBM can do is somewhat limited by the constraints of the GPL.
The GPL says that either everyone can distribute this freely to anyone else, and everyone who recieves it has the same right to distribute under the GPL as everyone else, with no restrictions (such as license fees). If not, no one has the right to distribute the file at all unless you get some alternate form of permission from everyone who contributed to writing it. This means one of two things would be the case if IBM owned the UNIX copyrights.
Linux is not in any way an infringement of the UNIX source code.
Linux is an infringement of the UNIX source code, and if IBM does not grant a free and open license to everyone under the GPL to use the infringing parts of Linux, IBM (and, technically, everyone else in the world) must cease to distribute Linux until those infringing parts are removed.
SCO has been effectively "untouchable" because they have no need for a long term strategy. They know that once the repercussions for their actions-- the failures of their lawsuits, etc-- finally come to pass, they will have long cashed in their inflated stock and run away. IBM, meanwhile, will still be here in 10 years time. That means no kamikaze attacks and no attacking your own revenue streams.
Of course, if what you're saying is that IBM probably won't be GPLing SysV in its entirity, then no, probably not, but who would want it if they did?
IBM giving an asset away isn't poetic. I'd call it heart-stoppingly unimaginable.
Irony is generally considered to be inherently poetic.
Not necessarily... it really just all depends on what exactly the number's significance was. You seem to be assuming the number was some sort of ratio, like Pi or e or phi. It doesn't have to be.
One thing that seems significant to me is that the number was 216 digits in our number system, but was also 216 digits in hebrew. This doesn't seem to make sense-- I don't totally understand hebrew numbers as far as I'm aware it seems like you aren't going to get the same number of digits in both systems for a single value. If I'm right about this, and this isn't just a plot hole, this (plus the way in which the movie implies the location of the digits within the number was as important as the value itself) suggests the number was not so much a value as a pattern.
Stephen Wolfram surmised that if you took the physical laws of the universe down to the absolute most bare postulates, you could describe the entire universe as a cellular automata that could be described in maybe two or three lines of "code". Apply this code to whatever the universe's initial conditions were, and you'd have a perfect simulation of the universe from the beginning. Perhaps the Number was an encoding of that sort of automata, or something of that sort. There are many possibilities.
I had always figured all the really fundamental numbers were irrational. After thinking about it and looking up on the internet it seems there are actually only 6: pi, e, i, 1, 0, and phi
1, 0 and (depending on definitions) i are rational. Moreover, you can easily *describe* e and pi in a single, rational integer such as the 216 digit number: just take the binary representation of a perl program which calculates the digits of e or pi.
...who think the slashbots are overreacting when they say Microsoft is about to start targetting Google for destruction as if it's priority number one...
Perhaps I should have been more clear. What I mean is that the contrasting design decisions of the PPC and x86 make it very very easy to emulate x86 on PPC and very difficult to go the other way around. Here are the reasons why.
This is a vast oversimplification, but when you are emulating you want two things. First off, you want it to be easy to efficiently rephrase the source instruction set into the target instruction set. Second off, you want to be able to hold all the emulated registers in hardware registers (since machine code is often very carefully optimized such that the registers are being used to their full potential).
The first issue has to do with RISC vs CISC-- Reduced vs Complex instruction set computing, two differing design philosophies. Again I'm probably oversimplifying here, but the latter-- the philosophy x86 follows-- has to do with having a vocabulary of lots of instructions which do complicated things; the former, the philosophy PPC follows, has to do with having a small number of instructions which do simple things. The idea here is that the RISC chip does less work per instruction, but it is hopefully able to execute so many more instructions in the same amount of time that overall it does work faster. Neither x86 nor PPC perfectly follows their philosophies, but for the sake of this argument it comes out about the same.
This makes RISC chips happen to be unnaturally good in the specific case of emulation. Since RISC chips very rarely have instruction sets that that dramatically differ from one another, and since conceptually all CISC instructions can be represented as a series of RISC primitives, all that a RISC chip has to do when emulating another arbitrary instruction set is break the source instructions down into piecies. A CISC set meanwhile has to do a lot of wierd molding in order to remain optimized. (As a random thing to consider, emulators that run on a CISC/x86 machine often must involve assembly to run well; emulators for PPC machines are almost universally written in plain old c.) When you look at the specific case of emulating a RISC machine from a CISC machine, this goes from being a quirky advantage of RISC to a major pain.
Basically, look at it this way: let's say you want to pull two values from memory, add them, and store the result back to memory. A CISC chip might have a single instruction that does this; on RISC, you would issue two load instructions, an add instruction, and a store instruction. If you're emulating CISC from RISC, no problem, you just look up what the loadloadaddstore instruction means and how to break it down into RISC instructions. But if you're emulating RISC from CISC, you have the difficult and expensive task of constantly looking at the incoming stream of instructions and trying to analyze them to figure out how you can reorder them and possibly merge them together into bigger instructions... not something you want to be doing in realtime.
The second issue is the registers, which has already been discussed-- hitting memory is relatively VERY expensive. This isn't so much a problem in normal compiled programs because you can carefully arrange the register usage to be efficient, but with emulation it's a massive pain because you don't have any higher-level information about the program and so have very little basis from which to optimize.
These mismatches between the designs of the two chips just happen to serve to create a case where while both x86 and PPC do many things well, the specific job of emulating PPC is something of a perverse case for x86. This is all I was trying to say.
Um. First off, 68k emulation is not expected to go terribly fast.
Second off, the 68000 may have twice the number of registers the x86 has, but the PPC has twice the number of registers the 68000 does! Emulating 16 registers on 8 registers is a problem to solve, emulating 32 registers on 8 registers seems a little bit daunting. This is even before you take into consideration that 4 of the x86 registers are not really general purpose, 8 of the 68000 registers are not really general purpose, and all 32 of the PPC's registers are potentially general purpose...
I don't think you can totally discount the register issue.
Microsoft's Home Entertainment division threw $348 MILLION away in the last QUARTER for which numbers are available, due to their policy of trying to grab a hold of the console market by selling their console at such a massive loss that licensing fees don't begin to make up for it. Think about this. This is about $100 million PER MONTH.
If MS will cheerfully spend $100 million a month to *potentially* expand their monopoly into a new market-- basically gaining customers by largely paying for the customers' products for them-- how exactly is $100 million going to make a difference as a fine? Isn't the idea of antitrust remedy to do something to convince the company to not perform their anticompetitive actions again? $100 million isn't just something MS would happily pay to maintain their monopoly, it's LESS than they're ALREADY spending to maintain their monopoly.
If this does turn out to be more than just rumors, this isn't a penalty for monopolistic status and anticompetitive action; it's a tax, a "ok, go on as you have, but give us some money for the privilidge to do so", and a measly one at that.
What I heard frequently repeated was that the venture capitalists who provided Google's initial investment would be expecting Google to IPO at some point in order to maximize the VCs' return-on-investment, and that Google was to some degree under obligation to eventually do as they wished.
Is there some basis to this or was this just another Slashdot urban legend?
Microsoft has been making constant noises about their new, upcoming search engine and how great it's going to be, basically doing the full-tilt vaporware thing, almost certainly with the purpose of adding uncertainty to Google's IPO. Microsoft would do something like release, with a huge media blitz, their new search product the day before Google's IPO. Waiting either pushes MS to actually release their product or allows Google, on the not wholly unrealistic possibility that MS drops the subject and then suddenly starts making noises again when Google reschedules their IPO, to accuse MS-Search of being vaporware. .
SCO is being the computer industry loose cannon, until the money from their donation from Microsoft (and, hypothetically, Sun) runs out; they are under no obligation to do anything at all to make more money; and they seem to be desperately, greedily obsessed with doing exactly two things: Hurting linux in any way, public relations or otherwise, and getting press attention. They've been making noises about suing Google; this is probably because making such noises keeps them in the press and because they know Google's about to IPO so they want to push Google to pay SCO some money so they'll go away and stop making uncertainty. However, it's not inconcievable they may well actually sue Google; their case with IBM may be about to unravel, and if so they need to start up a new lawsuit of some sort to keep up the illusion they have some sort of revenue model; and doing so on a nearly-IPOing Google would greatly help MS, which, how shall I put this, SCO at least seems to feel grateful toward, what with the fact their only documented profit ever happened solely because if MS's donation. SCO's a minor consideration, but still delaying IPO at least gives them time to die or deflates their lawsuit threats. .
The economy is changing, and since the president may well change or nearly change in about six months it may be set to change a lot more in a very unpredictable and drastic way. This is a big deal. IPOs are of much less good if the economy changes drastically shortly afterward. (Can you say "Andover"?)
I'm sure you can think of more reasons if you think about it.
It is a fact of a turing machine that any one can emulate any other
It is a fact that they can. However that does not mean that it will be easy.
It'd be slow, but it'd work just fine.
This is exactly the problem: it would be slow. And up until a certain point, it's slow enough you might as well not do it at all. No, there's no commercial demand for PPC emulation on x86; there doesn't really need to be. People write emulators just because they can. Do you think there is any "demand" for an emulator for the Amstrad CPC? In the meantime, there's some hobbyist demand from people who are "curious" about OS X; there's the guarantee of instant infamy for anyone who succeeds. People have really tried, put a lot of effort into trying to, emulate the PPC on an x86. I've never seen anyone succeed. As it turns out, though, writing a PPC emulator that runs on the x86 just happens to be unbelievably difficult to do with anything even remotely approaching an acceptable speed of emulation due to the neatly mismatching design philosophies of the two instruction sets. Yeah, if there was a real commercial *NEED* for someone to emulate, an acceptable emulator could probably be created. But the issue is a little more complicated than "oh, no one wants it".
If it's Mac emulation you are looking at, well that's a problem. The Mac ROMs are not available outside of Mac hardware, nor is the OS, and without those, it is useless. So to run the emulator, someone would need a legit copy of the ROMs and OS, meaning they'd need to own a Mac. Well if you own the hardware an emulator is worthless.
Not only is this not the hard part, this is the part that has already been solved. Modern macintoshes no longer have anything significant in ROM. The ROM is just a tiny kickstart thing and the OS is booted entirely using the openly documented Open Firmware protocol. This part is a non-issue.
Since the internals of an apple machine aren't that public, virtualizing the hardware might be a little bit difficult.. but, well, not that difficult, as practically all of the work has already been done for you in the form of the mac-on-linux project, a VMWare-like virtual machine for macintosh hardware that will let you boot OS X within a virtual machine on top of Linux. I am uncertain how much extra work needs to be done on top of that when emulating on the PC platform since I don't know what the internals of mac-on-linux look like. However, at the very least, the hardest and most voodoo-y part, actually getting it to boot, has already been done.
As far as the OS goes, you can buy a copy of the Mac OS without buying an actual mac. As in, you can go to a store and buy a copy of Mac OS X 10.3 in a box. This is not unrealistic; just because someone is emulating doesn't mean they aren't willing to actually buy the OS. Case in point, everyone who emulates Windows on the Mac does in fact actually have to buy a copy of Windows.
BTW, just out of curiousity, where are these PPC systems which you say are "available from IBM for reasonable prices"? I may just be going about it wrong, but I'm looking at IBM's website and the cheapest POWER-based system I can find is nearly $6000.
Netscape v. Microsoft would have turned out very differently if it weren't for Netscape literally self-destructing. Netscape did a better job of tearing itself apart than MS did. All MS did was hasten the process along by cutting the legs out from under Netscape's revenue model.
This won't work with google. Google's already on a very minimal advertising-based revenue model and they don't seem to be set to be making any huge mistakes. They've been having trouble of late but those troubles aren't due to internal mistakes, they're due to people purposefully trying to cheat at pagerank. If MS becomes even the slightest bit popular, people will be cheating at MS-Search as well so that won't happen.
Remember, ATM MS-Search is just vaporware. And google seems to have the sense that they will be ready for whatever MS throws at them.
Questioning of credentials is only valid sometimes.
If someone is making claims which they do not directly back up-- for example saying, say, "Iraq has 20,000 tons of chemical weapons" but not giving a direct explanation of where that number came from-- then saying something like "well, this person's biased, so we can't necessarily trust their words." is a totally fair thing to argue.
If someone is giving evidence or research-- for example saying "Iraq has 20,000 tons of chemical weapons, here is our exact evidence of what they've been doing with it and how we know they have them"-- this no longer works, and claims of "bias" become merely a distraction tactic, an invalid ad hominem attack. Once someone presents actual, hard evidence, you can no longer attack them to discredit their argument, you must attack the actual evidence, by either disproving it or showing why it is not important or showing why its veracity cannnot be trusted.
Which has groklaw done? Well, to be brutally honest I can't get their site to load right now (probably because, say, slashdot's linking to them?) but in the past they have consistently presented reasoned, well-researched and backed up statements just about every time, and clearly labelled and demarcated off any opinions, conjecture or op-ed. So I'm going to guess they're presenting facts and evidence, meaning speaking of groklaw's "bias" doesn't matter; the facts will be doing the speaking, not groklaw.
Ya know, at some point Pixar is going to hook up with a firm from India and cut their labor costs about 80%...
Yeah... instead of having to pay all that money to buy and maintain all those Dells to do their animation, they'll just outsource to the third world and have a cramped, unsafe room full of Koreans working for three cents a day in sweatshop conditions frantically scribbling down vector calculus formulas...
One of the theorized reasons for the GC price drop to $100 was just to hurt MS and Sony. Not just because more people would buy Gamecubes, but because MS and Sony would feel obligated, eventually, to price drop their consoles to reciprocate.
A price drop is something that Nintendo could afford, since they were making a profit on the Gamecube-- and they still are. Sony meanwhile is having financial difficulties and the XBox is losing obscene amounts of money.
The logic goes though that by forcing everyone into a price war, Nintendo is hurt little, Sony is hurt, and Microsoft goes from pissing away huge amounts of money with the XBox to pissing away massive amounts of money with the XBox. I'm not sure MS cares, but it's possible that this is what Nintendo had in mind.
I just wonder if at some point people are going to go from "the XBox has a marginally larger installed base in the US, the XBox is winning" to "Nintendo is making sizeable profits and the MS Games division is losing money like a huge gaping gut wound and looks to continue to do so for the concievable future, Nintendo is winning"...
The problem here is that patents are increasingly and increasingly not about ways to solve a problem and about problems themselves. The patent office is unable to tell anymore what is or is not a good patent, so it's just a huge land grab where each grab covers an infinite space of implementations. It's gone from patenting a specific implementation of plug-ins to a browser to patenting the idea of plug-ins to a browser itself.
It doesn't matter if you come up with a new way to do something. Very likely, your new way to do something is already covered by someone else's overbroad patent just by nature of what it does. Even more likely, someone else will independently come up with the same great new idea a year after you do, and patent it. And unless you are a very large company with the capacity to initiate and fight a protracted patent ownership battle in court, they will get to keep the patent, not you.
In the meanwhile, *maybe* you will be able to dance carefully around the huge holes created by the patents on what programming techniques techniques you can use. However this will mean careful knowledge of the patents out there, detailed lawyerlike scrutiny of every single line of code you write, and the preparedness to spend lots of money defending yourself against frivolous patent lawsuits whether you violate a patent or no. If you have to sanitize *everything* you do against umpteen million patents, that is a huge undertaking for a program of any size *ON TOP* of writing the program itself and it creates a major barrier to entry.
And all it would take to reach a point like that would be for the patents the patent office has *already granted* to be enforced.
Because, as we all know, the Free Market will always end up providing the best solution for everybody, where consumers can choose the best product at the best price and everything sorts itself out, magically, until we're all rich and free and happy.
The problem here is that while that theory actually does work, it requires the free market to actually be in correct operation.
If you do not actually have the free market running, the theory breaks down.
This is a problem in the specific case of music because of the existence of organizations like the RIAA and the BPI. These are called "cartels" and they are organizations that exist for the one and sole purpose of preventing the free market from operating-- or, more specifically, to prevent the natural consumer-protecting forces of the free market from coming into effect against producers.
In the presence of a cartel, the rules of economics change entirely; if the cartel becomes large enough, as it had has with the RIAA and BPI, the forces of capitalism as most people think of it cease to function..
Are there actually going to be any repercussions for this, and will the media actually pick this up and turn this into a major incident which the 'average person' is aware of?
Or will this just wind up with a bunch of "liberals" being very outraged, followed by a short flurry of page 7 articles in newspapers, followed by this story dropping off the face of the planet forever with around election time the only indication this ever happened being an occational mention in maybe a Tom Tomorrow strip or two?
eh, who am I kidding. Like with so many other things that republicans have done in the last four years, it will probably be the latter.
What do the slander/false claims laws work like in Australia? Are we about to see one more country crossed off the list of places SCO can go around claiming people owe them money for other people's work?
So that's what now, Germany and tentatively Australia? Well, maybe not a long list, but at least it's two countries longer than the list of countries in which SCO has successfully prosecuted a UNIX IP case...
What would be interesting is if what Google did was release spam blocking software.
It seems to me that blocking spam, and weeding out google-exploit spam search results, are the same sort of text processing / arms race sort of problem. Research on the latter, which is what Google is working on right now, will probably lead to techniques helpful in the former. So if they're looking at expanding into email, it seems like that would be a likely area for them to expand into...
Of course, given, they aren't right now doing a good JOB of filtering out the google-exploit spam results, but I expect they'll unveil some kind of brandnamed technology attempting to deal with the problem sometime shortly before MSN's search engine is released...
I just hope if they offer email addresses, they offer some, you know, better domans. I'm sorry, I don't want to be "mcc@google.com".
(1) I don't want a portal. (2) Historically, when search services become portals, their search services suffer as a result, or else try to force you to jump over all their portal "features" to use the search features you came to use. (3) I have multiple times in the past found myself having to stop using a search engine (for example, altavista) because they just couldn't keep their frigging portal-ness out of my face.
If google added portal features, I'd be OK with that as long as I could just keep using the search and not have to think about their portal. However I just have trouble trust that anyone, even google, could start "being a portal" and yet not have their core service lose focus or otherwise suffer as a result.
Yah, that's the story, though I must have seen an earlier version. In this version they seem to have combined the two articles I mentioned into one and correctly labelled the Dean bus pic.
Slashdot Mirror
IBM is committed to selling linux. This means what IBM can do is somewhat limited by the constraints of the GPL.
The GPL says that either everyone can distribute this freely to anyone else, and everyone who recieves it has the same right to distribute under the GPL as everyone else, with no restrictions (such as license fees). If not, no one has the right to distribute the file at all unless you get some alternate form of permission from everyone who contributed to writing it. This means one of two things would be the case if IBM owned the UNIX copyrights.
- Linux is not in any way an infringement of the UNIX source code.
- Linux is an infringement of the UNIX source code, and if IBM does not grant a free and open license to everyone under the GPL to use the infringing parts of Linux, IBM (and, technically, everyone else in the world) must cease to distribute Linux until those infringing parts are removed.
SCO has been effectively "untouchable" because they have no need for a long term strategy. They know that once the repercussions for their actions-- the failures of their lawsuits, etc-- finally come to pass, they will have long cashed in their inflated stock and run away. IBM, meanwhile, will still be here in 10 years time. That means no kamikaze attacks and no attacking your own revenue streams.Of course, if what you're saying is that IBM probably won't be GPLing SysV in its entirity, then no, probably not, but who would want it if they did? IBM giving an asset away isn't poetic. I'd call it heart-stoppingly unimaginable. Irony is generally considered to be inherently poetic.
Not necessarily... it really just all depends on what exactly the number's significance was. You seem to be assuming the number was some sort of ratio, like Pi or e or phi. It doesn't have to be.
One thing that seems significant to me is that the number was 216 digits in our number system, but was also 216 digits in hebrew. This doesn't seem to make sense-- I don't totally understand hebrew numbers as far as I'm aware it seems like you aren't going to get the same number of digits in both systems for a single value. If I'm right about this, and this isn't just a plot hole, this (plus the way in which the movie implies the location of the digits within the number was as important as the value itself) suggests the number was not so much a value as a pattern.
Stephen Wolfram surmised that if you took the physical laws of the universe down to the absolute most bare postulates, you could describe the entire universe as a cellular automata that could be described in maybe two or three lines of "code". Apply this code to whatever the universe's initial conditions were, and you'd have a perfect simulation of the universe from the beginning. Perhaps the Number was an encoding of that sort of automata, or something of that sort. There are many possibilities.
I had always figured all the really fundamental numbers were irrational. After thinking about it and looking up on the internet it seems there are actually only 6: pi, e, i, 1, 0, and phi
1, 0 and (depending on definitions) i are rational. Moreover, you can easily *describe* e and pi in a single, rational integer such as the 216 digit number: just take the binary representation of a perl program which calculates the digits of e or pi.
...who think the slashbots are overreacting when they say Microsoft is about to start targetting Google for destruction as if it's priority number one...
Perhaps I should have been more clear. What I mean is that the contrasting design decisions of the PPC and x86 make it very very easy to emulate x86 on PPC and very difficult to go the other way around. Here are the reasons why.
This is a vast oversimplification, but when you are emulating you want two things. First off, you want it to be easy to efficiently rephrase the source instruction set into the target instruction set. Second off, you want to be able to hold all the emulated registers in hardware registers (since machine code is often very carefully optimized such that the registers are being used to their full potential).
The first issue has to do with RISC vs CISC-- Reduced vs Complex instruction set computing, two differing design philosophies. Again I'm probably oversimplifying here, but the latter-- the philosophy x86 follows-- has to do with having a vocabulary of lots of instructions which do complicated things; the former, the philosophy PPC follows, has to do with having a small number of instructions which do simple things. The idea here is that the RISC chip does less work per instruction, but it is hopefully able to execute so many more instructions in the same amount of time that overall it does work faster. Neither x86 nor PPC perfectly follows their philosophies, but for the sake of this argument it comes out about the same.
This makes RISC chips happen to be unnaturally good in the specific case of emulation. Since RISC chips very rarely have instruction sets that that dramatically differ from one another, and since conceptually all CISC instructions can be represented as a series of RISC primitives, all that a RISC chip has to do when emulating another arbitrary instruction set is break the source instructions down into piecies. A CISC set meanwhile has to do a lot of wierd molding in order to remain optimized. (As a random thing to consider, emulators that run on a CISC/x86 machine often must involve assembly to run well; emulators for PPC machines are almost universally written in plain old c.) When you look at the specific case of emulating a RISC machine from a CISC machine, this goes from being a quirky advantage of RISC to a major pain.
Basically, look at it this way: let's say you want to pull two values from memory, add them, and store the result back to memory. A CISC chip might have a single instruction that does this; on RISC, you would issue two load instructions, an add instruction, and a store instruction. If you're emulating CISC from RISC, no problem, you just look up what the loadloadaddstore instruction means and how to break it down into RISC instructions. But if you're emulating RISC from CISC, you have the difficult and expensive task of constantly looking at the incoming stream of instructions and trying to analyze them to figure out how you can reorder them and possibly merge them together into bigger instructions... not something you want to be doing in realtime.
The second issue is the registers, which has already been discussed-- hitting memory is relatively VERY expensive. This isn't so much a problem in normal compiled programs because you can carefully arrange the register usage to be efficient, but with emulation it's a massive pain because you don't have any higher-level information about the program and so have very little basis from which to optimize.
These mismatches between the designs of the two chips just happen to serve to create a case where while both x86 and PPC do many things well, the specific job of emulating PPC is something of a perverse case for x86. This is all I was trying to say.
Um. First off, 68k emulation is not expected to go terribly fast.
Second off, the 68000 may have twice the number of registers the x86 has, but the PPC has twice the number of registers the 68000 does! Emulating 16 registers on 8 registers is a problem to solve, emulating 32 registers on 8 registers seems a little bit daunting. This is even before you take into consideration that 4 of the x86 registers are not really general purpose, 8 of the 68000 registers are not really general purpose, and all 32 of the PPC's registers are potentially general purpose...
I don't think you can totally discount the register issue.
Microsoft's Home Entertainment division threw $348 MILLION away in the last QUARTER for which numbers are available, due to their policy of trying to grab a hold of the console market by selling their console at such a massive loss that licensing fees don't begin to make up for it. Think about this. This is about $100 million PER MONTH.
If MS will cheerfully spend $100 million a month to *potentially* expand their monopoly into a new market-- basically gaining customers by largely paying for the customers' products for them-- how exactly is $100 million going to make a difference as a fine? Isn't the idea of antitrust remedy to do something to convince the company to not perform their anticompetitive actions again? $100 million isn't just something MS would happily pay to maintain their monopoly, it's LESS than they're ALREADY spending to maintain their monopoly.
If this does turn out to be more than just rumors, this isn't a penalty for monopolistic status and anticompetitive action; it's a tax, a "ok, go on as you have, but give us some money for the privilidge to do so", and a measly one at that.
What I heard frequently repeated was that the venture capitalists who provided Google's initial investment would be expecting Google to IPO at some point in order to maximize the VCs' return-on-investment, and that Google was to some degree under obligation to eventually do as they wished.
Is there some basis to this or was this just another Slashdot urban legend?
- Microsoft has been making constant noises about their new, upcoming search engine and how great it's going to be, basically doing the full-tilt vaporware thing, almost certainly with the purpose of adding uncertainty to Google's IPO. Microsoft would do something like release, with a huge media blitz, their new search product the day before Google's IPO. Waiting either pushes MS to actually release their product or allows Google, on the not wholly unrealistic possibility that MS drops the subject and then suddenly starts making noises again when Google reschedules their IPO, to accuse MS-Search of being vaporware.
- SCO is being the computer industry loose cannon, until the money from their donation from Microsoft (and, hypothetically, Sun) runs out; they are under no obligation to do anything at all to make more money; and they seem to be desperately, greedily obsessed with doing exactly two things: Hurting linux in any way, public relations or otherwise, and getting press attention. They've been making noises about suing Google; this is probably because making such noises keeps them in the press and because they know Google's about to IPO so they want to push Google to pay SCO some money so they'll go away and stop making uncertainty. However, it's not inconcievable they may well actually sue Google; their case with IBM may be about to unravel, and if so they need to start up a new lawsuit of some sort to keep up the illusion they have some sort of revenue model; and doing so on a nearly-IPOing Google would greatly help MS, which, how shall I put this, SCO at least seems to feel grateful toward, what with the fact their only documented profit ever happened solely because if MS's donation. SCO's a minor consideration, but still delaying IPO at least gives them time to die or deflates their lawsuit threats.
- The economy is changing, and since the president may well change or nearly change in about six months it may be set to change a lot more in a very unpredictable and drastic way. This is a big deal. IPOs are of much less good if the economy changes drastically shortly afterward. (Can you say "Andover"?)
I'm sure you can think of more reasons if you think about it..
.
It is a fact of a turing machine that any one can emulate any other
It is a fact that they can. However that does not mean that it will be easy.
It'd be slow, but it'd work just fine.
This is exactly the problem: it would be slow. And up until a certain point, it's slow enough you might as well not do it at all. No, there's no commercial demand for PPC emulation on x86; there
doesn't really need to be. People write emulators just because they can. Do you think there is any "demand" for an emulator for the Amstrad CPC? In the meantime, there's some hobbyist demand from people who are "curious" about OS X; there's the guarantee of instant infamy for anyone who succeeds. People have really tried, put a lot of effort into trying to, emulate the PPC on an x86. I've never seen anyone succeed. As it turns out, though, writing a PPC emulator that runs on the x86 just happens to be unbelievably difficult to do with anything even remotely approaching an acceptable speed of emulation due to the neatly mismatching design philosophies of the two instruction sets. Yeah, if there was a real commercial *NEED* for someone to emulate, an acceptable emulator could probably be created. But the issue is a little more complicated than "oh, no one wants it".
If it's Mac emulation you are looking at, well that's a problem. The Mac ROMs are not available outside of Mac hardware, nor is the OS, and without those, it is useless. So to run the emulator, someone would need a legit copy of the ROMs and OS, meaning they'd need to own a Mac. Well if you own the hardware an emulator is worthless.
Not only is this not the hard part, this is the part that has already been solved. Modern macintoshes no longer have anything significant in ROM. The ROM is just a tiny kickstart thing and the OS is booted entirely using the openly documented Open Firmware protocol. This part is a non-issue.
Since the internals of an apple machine aren't that public, virtualizing the hardware might be a little bit difficult.. but, well, not that difficult, as practically all of the work has already been done for you in the form of the mac-on-linux project, a VMWare-like virtual machine for macintosh hardware that will let you boot OS X within a virtual machine on top of Linux. I am uncertain how much extra work needs to be done on top of that when emulating on the PC platform since I don't know what the internals of mac-on-linux look like. However, at the very least, the hardest and most voodoo-y part, actually getting it to boot, has already been done.
As far as the OS goes, you can buy a copy of the Mac OS without buying an actual mac. As in, you can go to a store and buy a copy of Mac OS X 10.3 in a box. This is not unrealistic; just because someone is emulating doesn't mean they aren't willing to actually buy the OS. Case in point, everyone who emulates Windows on the Mac does in fact actually have to buy a copy of Windows.
BTW, just out of curiousity, where are these PPC systems which you say are "available from IBM for reasonable prices"? I may just be going about it wrong, but I'm looking at IBM's website and the cheapest POWER-based system I can find is nearly $6000.
Film at 11
Netscape v. Microsoft would have turned out very differently if it weren't for Netscape literally self-destructing. Netscape did a better job of tearing itself apart than MS did. All MS did was hasten the process along by cutting the legs out from under Netscape's revenue model.
This won't work with google. Google's already on a very minimal advertising-based revenue model and they don't seem to be set to be making any huge mistakes. They've been having trouble of late but those troubles aren't due to internal mistakes, they're due to people purposefully trying to cheat at pagerank. If MS becomes even the slightest bit popular, people will be cheating at MS-Search as well so that won't happen.
Remember, ATM MS-Search is just vaporware. And google seems to have the sense that they will be ready for whatever MS throws at them.
Questioning of credentials is only valid sometimes.
If someone is making claims which they do not directly back up-- for example saying, say, "Iraq has 20,000 tons of chemical weapons" but not giving a direct explanation of where that number came from-- then saying something like "well, this person's biased, so we can't necessarily trust their words." is a totally fair thing to argue.
If someone is giving evidence or research-- for example saying "Iraq has 20,000 tons of chemical weapons, here is our exact evidence of what they've been doing with it and how we know they have them"-- this no longer works, and claims of "bias" become merely a distraction tactic, an invalid ad hominem attack. Once someone presents actual, hard evidence, you can no longer attack them to discredit their argument, you must attack the actual evidence, by either disproving it or showing why it is not important or showing why its veracity cannnot be trusted.
Which has groklaw done? Well, to be brutally honest I can't get their site to load right now (probably because, say, slashdot's linking to them?) but in the past they have consistently presented reasoned, well-researched and backed up statements just about every time, and clearly labelled and demarcated off any opinions, conjecture or op-ed. So I'm going to guess they're presenting facts and evidence, meaning speaking of groklaw's "bias" doesn't matter; the facts will be doing the speaking, not groklaw.
Have you seen one single quote that Darl McBride has put out on this entire subject?
If so, I do not think you would necessarily be accusing the linux side of being the "emotional" ones.
Ya know, at some point Pixar is going to hook up with a firm from India and cut their labor costs about 80%...
Yeah... instead of having to pay all that money to buy and maintain all those Dells to do their animation, they'll just outsource to the third world and have a cramped, unsafe room full of Koreans working for three cents a day in sweatshop conditions frantically scribbling down vector calculus formulas...
One of the theorized reasons for the GC price drop to $100 was just to hurt MS and Sony. Not just because more people would buy Gamecubes, but because MS and Sony would feel obligated, eventually, to price drop their consoles to reciprocate.
A price drop is something that Nintendo could afford, since they were making a profit on the Gamecube-- and they still are. Sony meanwhile is having financial difficulties and the XBox is losing obscene amounts of money.
The logic goes though that by forcing everyone into a price war, Nintendo is hurt little, Sony is hurt, and Microsoft goes from pissing away huge amounts of money with the XBox to pissing away massive amounts of money with the XBox. I'm not sure MS cares, but it's possible that this is what Nintendo had in mind.
I just wonder if at some point people are going to go from "the XBox has a marginally larger installed base in the US, the XBox is winning" to "Nintendo is making sizeable profits and the MS Games division is losing money like a huge gaping gut wound and looks to continue to do so for the concievable future, Nintendo is winning"...
Q: How many Microsoft engineers does it take to change a light bulb?
A: They don't, they just redefine darkness as the new standard.
It's all neat and stuff that they've done this, but "the world's first singing synthesis software"? Not by a long shot.
Perhaps next month we will get to see an article by Slashdot on the subject of Mitsubishi releasing the world's first car.
The problem here is that patents are increasingly and increasingly not about ways to solve a problem and about problems themselves. The patent office is unable to tell anymore what is or is not a good patent, so it's just a huge land grab where each grab covers an infinite space of implementations. It's gone from patenting a specific implementation of plug-ins to a browser to patenting the idea of plug-ins to a browser itself.
It doesn't matter if you come up with a new way to do something. Very likely, your new way to do something is already covered by someone else's overbroad patent just by nature of what it does. Even more likely, someone else will independently come up with the same great new idea a year after you do, and patent it. And unless you are a very large company with the capacity to initiate and fight a protracted patent ownership battle in court, they will get to keep the patent, not you.
In the meanwhile, *maybe* you will be able to dance carefully around the huge holes created by the patents on what programming techniques techniques you can use. However this will mean careful knowledge of the patents out there, detailed lawyerlike scrutiny of every single line of code you write, and the preparedness to spend lots of money defending yourself against frivolous patent lawsuits whether you violate a patent or no. If you have to sanitize *everything* you do against umpteen million patents, that is a huge undertaking for a program of any size *ON TOP* of writing the program itself and it creates a major barrier to entry.
And all it would take to reach a point like that would be for the patents the patent office has *already granted* to be enforced.
Because, as we all know, the Free Market will always end up providing the best solution for everybody, where consumers can choose the best product at the best price and everything sorts itself out, magically, until we're all rich and free and happy.
The problem here is that while that theory actually does work, it requires the free market to actually be in correct operation.
If you do not actually have the free market running, the theory breaks down.
This is a problem in the specific case of music because of the existence of organizations like the RIAA and the BPI. These are called "cartels" and they are organizations that exist for the one and sole purpose of preventing the free market from operating-- or, more specifically, to prevent the natural consumer-protecting forces of the free market from coming into effect against producers.
In the presence of a cartel, the rules of economics change entirely; if the cartel becomes large enough, as it had has with the RIAA and BPI, the forces of capitalism as most people think of it cease to function..
Are there actually going to be any repercussions for this, and will the media actually pick this up and turn this into a major incident which the 'average person' is aware of?
Or will this just wind up with a bunch of "liberals" being very outraged, followed by a short flurry of page 7 articles in newspapers, followed by this story dropping off the face of the planet forever with around election time the only indication this ever happened being an occational mention in maybe a Tom Tomorrow strip or two?
eh, who am I kidding. Like with so many other things that republicans have done in the last four years, it will probably be the latter.
Was the first one I attempted to download.
It was an incomplete download.
The first mp3 I successfully downloaded was Blur - Song 2
What do the slander/false claims laws work like in Australia? Are we about to see one more country crossed off the list of places SCO can go around claiming people owe them money for other people's work?
So that's what now, Germany and tentatively Australia? Well, maybe not a long list, but at least it's two countries longer than the list of countries in which SCO has successfully prosecuted a UNIX IP case...
What would be interesting is if what Google did was release spam blocking software.
It seems to me that blocking spam, and weeding out google-exploit spam search results, are the same sort of text processing / arms race sort of problem. Research on the latter, which is what Google is working on right now, will probably lead to techniques helpful in the former. So if they're looking at expanding into email, it seems like that would be a likely area for them to expand into...
Of course, given, they aren't right now doing a good JOB of filtering out the google-exploit spam results, but I expect they'll unveil some kind of brandnamed technology attempting to deal with the problem sometime shortly before MSN's search engine is released...
I just hope if they offer email addresses, they offer some, you know, better domans. I'm sorry, I don't want to be "mcc@google.com".
My answer to this is
(1) I don't want a portal.
(2) Historically, when search services become portals, their search services suffer as a result, or else try to force you to jump over all their portal "features" to use the search features you came to use.
(3) I have multiple times in the past found myself having to stop using a search engine (for example, altavista) because they just couldn't keep their frigging portal-ness out of my face.
If google added portal features, I'd be OK with that as long as I could just keep using the search and not have to think about their portal. However I just have trouble trust that anyone, even google, could start "being a portal" and yet not have their core service lose focus or otherwise suffer as a result.
Yah, that's the story, though I must have seen an earlier version. In this version they seem to have combined the two articles I mentioned into one and correctly labelled the Dean bus pic.