Look, every time you introduce a pseudo-emulation layer, you introduce inefficiency over a direct execution engine. To quote another famous Scottish engineer, "Ye canna change the laws o' physics, Captain!"
The benefits offered by variable instruction set processors are illusory. You have a similar problem with the Java VM. Somewhere, somewhen, a hit has to occur with performance. You either do the work at compile time, at run time, or you fudge the issue with JITs and/or code-morphing (read: same thing).
It seems like a great idea which raises its head every time there is a new generation of engineers who forget the bitter lessons learnt by their forebears.
I'm perfectly serious. And what's inaccurate? Where's the crap? Soft machines like Crusoe share the same inherent design problem which I haven't seen change for more than 25 years.
I have a hunch that this fascination with Transmeta has more to do with the fact that they sponsored Linus's H1-B than anything else. It's called reflected glory (or something).
If I want a low power consumption 350MHz Pentium I can get one from Intel.
Soft instruction set computers have been around for > 25 years. I've never seen one that didn't end up hardwiring the instruction decode at least to get reasonable performance.
And who says the Army will use Linux? I don't see that happening on a wearable battlefield computer.
I can't think of a single UNIX-based commercial or government organization that uses GNOME. All the banks, telcos, cellular cos, NASA, NOAA, IRS, oil companies, CERN, the National Laboratories, CERN, weather orgs in the US and Europe use CDE/Motif. You don't have to like CDE, but it's there. Sun adopting GNOME? Let me tell you, Sun will not have that as their default desktop within 5 years, by which time they will have changed their corporate minds again.
I agree that consortia have a sad history. But then, what is the free/open software community if it's not the world's biggest ever consortium?
Yes, Bruce, it is. Anybody who bothered to read to the end of the article would observe the comment at the foot that it is satire intended to brighten your day. But it's well observed and astute. It highlights the self-indulgebt nature of the open source community with witheringly accurate fire.
Also, its predictions are likely to be accurate as well.
Multiple projects/forks? That's exactly how the UNIX companies handed the commodity desktop market to Microsoft, reacting with too little, too late with CDE (Common Desktop Environment). History will repeat itself with Linux on the desktop. In fact, it already has.
There is real scope for a massive industry split. Remember when IBM tried to reclaim the PC market with the PS/2 and "micro-channel architecture", a non-ISA bus which they had patented? (1987). The world of the PC just walked away from them and did its own thing, EISA and so on.
This could be the luckiest break that AMD will ever get.
Hang on, hang on, you're missing a thing or two about software copyright.
If your hypothetical company A produced a derivative product from Microsoft's source code, and Microsoft took action, the likeliest outcome (at least in the US) is that a court could order a comparison to be made by an independent expert of the two pieces of source code. If that expert found that there were "striking similarities" between the two, then the case would be part proven.
Secondly, how long did it take you to write this almost perfect clone of Windows? A week? Really? Can you show me the timekeeping records of the army of hackers you had working on this project to write a "new" Windows? Their names and addresses? No? (And so how long have those guys at Freedows been at it so far?) Your case is now in more trouble.
Finally, did you have access to the original source code? Can you prove that you didn't? Can all your army of programmers swear affidavits that they have never seen the Windows source code (and could not, therefore, have copied it)? Kinda tricky if it's been published for all to see on the Internet, don't you think.
Why is everybody so obsessed with source code, Microsoft's or anybody else's? Just what in the heck are you going to *do* with a glimpse of some of the source code to Office or Windows?
Laugh at banal commentary? Giggle at a misused pointer? Squirm over the indentation? Be mildly shocked at the local variable names?
Say you got the lot -- now what are you going to do? Fiddle around with n zillion lines of tired, structurally decaying code to make a version of Windows that doesn't work as well as the binary on the box you bought? What's the chances that you will have the least clue what you're doing? Or that it will be actually *worth* anything to anybody? What are you going to do? Spend your life rebuilding Windows? Please, feel free . . .
Don't you get it it? IT DOESN'T MATTER WHO SAW MICROSOFT'S SOURCE CODE.
Hmmm. I'm obviously an old timer, but a PS/2 was around in 1987 -- the IBM Personal System/2, when they tried (and failed) to make the architecture proprietary. What's left of it all is that little DIN socket on the back of your PC called a PS/2 mouseport.
My point -- oh yes -- it's darn confusing to those of us over the age of 25 if you start re-using industry TLAs. Talking of which, just who the heck are RMS and ESR?
You are, of course, quite right about the Polish involvement. They also gave it to the British because they had been somewhat rudely invaded. Forgive my oversimplified account which was for the sake of brevity, and because the majority of the kids that seem to read Slashdot either (a) don't read anything and/or (b) seem to think that computer programming began in about 1995 with them (or shortly before with Linus Torvalds). I haven't yet read Singh's book but have been told that it's excellent.
Of course the UK, NZ, Oz, Canadian and US governments are all doing the same thing -- they agreed to do so in 1947 when all 5 countries signed the UK-US Security Agreement (UKUSA for short). This formalized the cooperation that took place during WWII at Bletchley Park in the UK where the German Enigma code was broken and the world's first digital, programmable computer was built (Colossus).
They were real h/crackers. And they're still at it in the name of protecting freedom and democracy. You may disagree with that, of course, or disagree that this is necessary, but others are entitled to be concerned. For example, the Northern Ireland Omagh bombers are believed to have been identified by their used of cellphones and the British government's analysis of 15m (yes million) cellphone calls. This will have been traffic analysis, I doubt whether they bothered attempting to record/listen to every call.
I have to confess that I have never read so much nonsense in 30+ years of writing software as I've just read on this thread. Who, in the name of all that's Holy, gives a stuff what gender or color the members of the GNOME Foundation are? How does that affect the quality of the product? How does the GPLness or otherwise affect the quality of the software? Who decides who are the best talents to write the stuff? A democratic election?!?
This can't be real. We all know how well democracy works at selecting our political leaders -- we have to put up with it there for, as Winston Churchill said "Democracy is a bad form of Government, but all the others are worse".
This is not how to develop software. We can demand, must demand, and are entitled to demand, excellence. And what do we witness on these/. debates? Mindless juvenile rantings driven and possessed by Stallmanesque agendas which have little or nothing to do with quality of software but a very great deal to do with individual egos and an aimless, ill-defined, socialist agenda over communal ownership, colored by an overt desire to inflict the same on the unwilling.
Its a pretty sad state of affairs when RMS, GPL and ESR are the new holy trinity governing the judgement of software. Incidentally, just who in hell are they? (Let's see who is stupid enough to rise to that.)
This resulted from a misguided attempt to give employees the right to email privacy in the workplace. This caused such an uproar from employers (like me) that the government had to reverse the legislation. Employers pointed out that it's *their* email system, not the employees, and that they have no right to privacy over snail mail that they send on the company letterhead.
And now the best bit: the UK government has given itself the right to read *everybody's* email by forcing all major ISPs to instal interception equipment. How is that for good, old fashioned, British humbug?
. . . another bloody operating system. When will it penetrate the hacker mind that the vast majority of computer users really don't care about the OS? 90%+ of the great unwashed are using Windows and are likely to be still doing so in 10 years' time. The real success stories of the Internet/big computer scene are Solaris and FreeBSD. We have servers that stay up year-on-year running those systems.
The ISS is already equipped with 2-meter ham equipment -- don't know if they'll do data as well as voice though. The first three-man crew will all have ham licences. 9600 baud should be enough for anyone.
Don't be silly -- code morphing is simply a firmware form of just-in-time compiling of x86 instructions to native Transmeta code. If you reboot, you'll clear the "JIT cache". It's essentially the Java VM problem -- if you interpret, it will be slower, if you JIT it, it goes faster second time around -- and that's it. The first time is slower because the "compile" occurs then.
Er, not quite. IE came from NCSA Mosaic via Spyglass. Netscape's browser was "clean room" developed to avoid using any Mosaic code -- although Netscape employed ex-NCSA people.
Wrong bastard.
The Problem of RMS is that a generation of software writers seem to be being indoctrinated to believe that the only worthwhile test of "goodness" in software is whther it's released under the terms of the GPL or not. This is stifling innovation, and encouraging individuals to de-focus on the characteristics that actually matter in software, such as whether it works reliably, is easy to use, is innovative, is useful to the general public or some section thereof and so on.
In particular, innovation seems to be as short amongst the free software people as it is with Microsoft. Every piece of free software of note such as the GNU stuff, Linux (no, Dr, Stallman, it is *not* GNU/Linux however many times you repeat that calumny), the various tools and so on are all clones of a piece of "not free" software which in many cases was truly innovative. By "not free" I simply mean a software item that was released under a licence other than the GPL. Some of these "not free" licences are considerably less restrictive than the GPL. At least they don't seek to impose conditions on what other bits of software you use with them.
But worst of all, the death of innovation seems to be the result of the polarisation centring around the Microsofties and the free software folks.
Slashdot Mirror
Thanks for enlightening me -- that certainly makes more sense. It seems that we're subjected to a barrage of misreporting.
Still suspect, though.
So because it's Motif it's ugly? No wonder you post as an AC.
The world is full of Motif applications, far more than any other UNIX GUI. Trust me, I've seen uglier.
Your prejudice is showing.
Nonsense.
Look, every time you introduce a pseudo-emulation layer, you introduce inefficiency over a direct execution engine. To quote another famous Scottish engineer, "Ye canna change the laws o' physics, Captain!"
The benefits offered by variable instruction set processors are illusory. You have a similar problem with the Java VM. Somewhere, somewhen, a hit has to occur with performance. You either do the work at compile time, at run time, or you fudge the issue with JITs and/or code-morphing (read: same thing).
It seems like a great idea which raises its head every time there is a new generation of engineers who forget the bitter lessons learnt by their forebears.
I'm accepting bets on this.
Troll? Stupid expression.
I'm perfectly serious. And what's inaccurate? Where's the crap? Soft machines like Crusoe share the same inherent design problem which I haven't seen change for more than 25 years.
Linus' H1-B? Is he an illegal, then?
Let's hear you.
I have a hunch that this fascination with Transmeta has more to do with the fact that they sponsored Linus's H1-B than anything else. It's called reflected glory (or something).
If I want a low power consumption 350MHz Pentium I can get one from Intel.
Soft instruction set computers have been around for > 25 years. I've never seen one that didn't end up hardwiring the instruction decode at least to get reasonable performance.
And who says the Army will use Linux? I don't see that happening on a wearable battlefield computer.
"GNOME ate CDE's lunch . . ."
Where on earth did you get that idea from?
I can't think of a single UNIX-based commercial or government organization that uses GNOME. All the banks, telcos, cellular cos, NASA, NOAA, IRS, oil companies, CERN, the National Laboratories, CERN, weather orgs in the US and Europe use CDE/Motif. You don't have to like CDE, but it's there. Sun adopting GNOME? Let me tell you, Sun will not have that as their default desktop within 5 years, by which time they will have changed their corporate minds again.
I agree that consortia have a sad history. But then, what is the free/open software community if it's not the world's biggest ever consortium?
Yes, Bruce, it is. Anybody who bothered to read to the end of the article would observe the comment at the foot that it is satire intended to brighten your day. But it's well observed and astute. It highlights the self-indulgebt nature of the open source community with witheringly accurate fire.
Also, its predictions are likely to be accurate as well.
Multiple projects/forks? That's exactly how the UNIX companies handed the commodity desktop market to Microsoft, reacting with too little, too late with CDE (Common Desktop Environment). History will repeat itself with Linux on the desktop. In fact, it already has.
Read with humility.
--
Since when were compilers part of the OS?
Look, you and your pal Richard can scream GNU/Linux until you're blue in the face, but you're still wrong.
You really don't have a clue, do you?
I think you're pretty much on the money.
There is real scope for a massive industry split. Remember when IBM tried to reclaim the PC market with the PS/2 and "micro-channel architecture", a non-ISA bus which they had patented? (1987). The world of the PC just walked away from them and did its own thing, EISA and so on.
This could be the luckiest break that AMD will ever get.
Hang on, hang on, you're missing a thing or two about software copyright.
If your hypothetical company A produced a derivative product from Microsoft's source code, and Microsoft took action, the likeliest outcome (at least in the US) is that a court could order a comparison to be made by an independent expert of the two pieces of source code. If that expert found that there were "striking similarities" between the two, then the case would be part proven.
Secondly, how long did it take you to write this almost perfect clone of Windows? A week? Really? Can you show me the timekeeping records of the army of hackers you had working on this project to write a "new" Windows? Their names and addresses? No? (And so how long have those guys at Freedows been at it so far?) Your case is now in more trouble.
Finally, did you have access to the original source code? Can you prove that you didn't? Can all your army of programmers swear affidavits that they have never seen the Windows source code (and could not, therefore, have copied it)? Kinda tricky if it's been published for all to see on the Internet, don't you think.
Forget the reverse engineering, you're dead.
Why is everybody so obsessed with source code, Microsoft's or anybody else's? Just what in the heck are you going to *do* with a glimpse of some of the source code to Office or Windows?
Laugh at banal commentary? Giggle at a misused pointer? Squirm over the indentation? Be mildly shocked at the local variable names?
Say you got the lot -- now what are you going to do? Fiddle around with n zillion lines of tired, structurally decaying code to make a version of Windows that doesn't work as well as the binary on the box you bought? What's the chances that you will have the least clue what you're doing? Or that it will be actually *worth* anything to anybody? What are you going to do? Spend your life rebuilding Windows? Please, feel free . . .
Don't you get it it? IT DOESN'T MATTER WHO SAW MICROSOFT'S SOURCE CODE.
COMDEX/Fall, 1987.
Hmmm. I'm obviously an old timer, but a PS/2 was around in 1987 -- the IBM Personal System/2, when they tried (and failed) to make the architecture proprietary. What's left of it all is that little DIN socket on the back of your PC called a PS/2 mouseport.
My point -- oh yes -- it's darn confusing to those of us over the age of 25 if you start re-using industry TLAs. Talking of which, just who the heck are RMS and ESR?
You are, of course, quite right about the Polish involvement. They also gave it to the British because they had been somewhat rudely invaded. Forgive my oversimplified account which was for the sake of brevity, and because the majority of the kids that seem to read Slashdot either (a) don't read anything and/or (b) seem to think that computer programming began in about 1995 with them (or shortly before with Linus Torvalds). I haven't yet read Singh's book but have been told that it's excellent.
Of course the UK, NZ, Oz, Canadian and US governments are all doing the same thing -- they agreed to do so in 1947 when all 5 countries signed the UK-US Security Agreement (UKUSA for short). This formalized the cooperation that took place during WWII at Bletchley Park in the UK where the German Enigma code was broken and the world's first digital, programmable computer was built (Colossus).
They were real h/crackers. And they're still at it in the name of protecting freedom and democracy. You may disagree with that, of course, or disagree that this is necessary, but others are entitled to be concerned. For example, the Northern Ireland Omagh bombers are believed to have been identified by their used of cellphones and the British government's analysis of 15m (yes million) cellphone calls. This will have been traffic analysis, I doubt whether they bothered attempting to record/listen to every call.
Now, where's the balance, that's the question.
I have to confess that I have never read so much nonsense in 30+ years of writing software as I've just read on this thread. Who, in the name of all that's Holy, gives a stuff what gender or color the members of the GNOME Foundation are? How does that affect the quality of the product? How does the GPLness or otherwise affect the quality of the software? Who decides who are the best talents to write the stuff? A democratic election?!?
/. debates? Mindless juvenile rantings driven and possessed by Stallmanesque agendas which have little or nothing to do with quality of software but a very great deal to do with individual egos and an aimless, ill-defined, socialist agenda over communal ownership, colored by an overt desire to inflict the same on the unwilling.
This can't be real. We all know how well democracy works at selecting our political leaders -- we have to put up with it there for, as Winston Churchill said "Democracy is a bad form of Government, but all the others are worse".
This is not how to develop software. We can demand, must demand, and are entitled to demand, excellence. And what do we witness on these
Its a pretty sad state of affairs when RMS, GPL and ESR are the new holy trinity governing the judgement of software. Incidentally, just who in hell are they? (Let's see who is stupid enough to rise to that.)
MightyMicro
--
And why not? /.ers vote for software just because it's GPLed, not for any intrinsic quality. Don't start changing the rules now.
3 replies below my current contempt
Er . . . well you remembered it. Besides, what's so difficult about sunsite at department of computing at imperial college, academic, united kingdom?
See also ist.co.uk for imperial connections.
This resulted from a misguided attempt to give employees the right to email privacy in the workplace. This caused such an uproar from employers (like me) that the government had to reverse the legislation. Employers pointed out that it's *their* email system, not the employees, and that they have no right to privacy over snail mail that they send on the company letterhead.
And now the best bit: the UK government has given itself the right to read *everybody's* email by forcing all major ISPs to instal interception equipment. How is that for good, old fashioned, British humbug?
. . . another bloody operating system. When will it penetrate the hacker mind that the vast majority of computer users really don't care about the OS? 90%+ of the great unwashed are using Windows and are likely to be still doing so in 10 years' time. The real success stories of the Internet/big computer scene are Solaris and FreeBSD. We have servers that stay up year-on-year running those systems.
3 replies beneath my current contempt
The ISS is already equipped with 2-meter ham equipment -- don't know if they'll do data as well as voice though. The first three-man crew will all have ham licences. 9600 baud should be enough for anyone.
3 replies beneath my current contempt
Don't be silly -- code morphing is simply a firmware form of just-in-time compiling of x86 instructions to native Transmeta code. If you reboot, you'll clear the "JIT cache". It's essentially the Java VM problem -- if you interpret, it will be slower, if you JIT it, it goes faster second time around -- and that's it. The first time is slower because the "compile" occurs then.
Er, not quite. IE came from NCSA Mosaic via Spyglass. Netscape's browser was "clean room" developed to avoid using any Mosaic code -- although Netscape employed ex-NCSA people. Wrong bastard.
The Problem of RMS is that a generation of software writers seem to be being indoctrinated to believe that the only worthwhile test of "goodness" in software is whther it's released under the terms of the GPL or not. This is stifling innovation, and encouraging individuals to de-focus on the characteristics that actually matter in software, such as whether it works reliably, is easy to use, is innovative, is useful to the general public or some section thereof and so on. In particular, innovation seems to be as short amongst the free software people as it is with Microsoft. Every piece of free software of note such as the GNU stuff, Linux (no, Dr, Stallman, it is *not* GNU/Linux however many times you repeat that calumny), the various tools and so on are all clones of a piece of "not free" software which in many cases was truly innovative. By "not free" I simply mean a software item that was released under a licence other than the GPL. Some of these "not free" licences are considerably less restrictive than the GPL. At least they don't seek to impose conditions on what other bits of software you use with them. But worst of all, the death of innovation seems to be the result of the polarisation centring around the Microsofties and the free software folks.