Linus, Transmeta, Proprietary Code and Metcalfe

smeng58 writes "InfoWorld's Bob Metcalfe asks why, if Linus Torvalds is truly a believer in Open Source, Transmeta Corp. has seen fit to make Crusoe, or at least its VLIW "code morphing", proprietary. The column goes on to say that, since the processor will run Windows code, there must be some thing wrong with Linux. Sad when a computer pundit appears not understand what x86 code is. "

Comments

Excellent job Mr. Metcalfe

[mosch]

Web columnists get paid for bringing readers, and Mr. Metcalfe has succeeded with aplumb in doing that with this latest irrational comparison. Yes, Mr. Metcalfe, I swear that all Open Source software users are exactly like the unfortunately typical flaming slashdroids that sometimes show their face here. Really. I swear that I'm not being sarcastic.

Mr. Metcalfe just got a whole lot of hits from us, therefore proving to his employers his competance as a desirable writer. Odd, isn't it, that one can do that today solely by writing bizarre gibberish and merging it with an seventh grade paper on animal farm?

Mr. Metcalfe, you're a real nowhere man.
----------------------------

A steady diet of his own words.

[Wakko Warner]

Metcalfe is a fucking idiot who, were it not for ethernet, would have done us all a favor by being stillborn. I can't believe people actually give him money to write the kind of crap he writes. - A.P.
--

"One World, one Web, one Program" - Microsoft promotional ad

Hm, he's rich.

[Wakko Warner]

That must make him smart.

- A.P.
--

"One World, one Web, one Program" - Microsoft promotional ad

Re:Who cares what Bob Metcalf thinks anyway?

[Alex Belits]

s/what/if/

The Metcalf Lesson

[Skyshadow]

Bob Metcalf provides a valuable lesson to those of us in the tech industry: Just because someone comes up with something cool does not mean that they have a clue.

Personally, I don't see why Metcalf even gets the time of day from anyone. Where does his credibility hail from? Coming up with a bit of tech, even one which sees so wide a deployment, does not necessarily mean that the developer is qualified to make comments on other areas. It doesn't even mean that they're a big thinker -- Bob Metcalf is obviously no RMS. Technical ability does not mean you are immune from being an idiot.

In any event, Metcalf hasn't earned any sort of consideration or credibility in my mind. On the other hand, Linus has built up a fair amount of credibility thanks to his general ability to stay "above" the petty arguements in OSS land and to almost always make a great deal of sense. So, in a Bob Metcalf v. Linus Torvolds arguement, who gets the benefit of the doubt?

----

Re:Bob Metcalfe joins the tabloid press

[Surazal]

Now lets be facing it, Bob Metcalfe is not a stupid man.

An alternative way of putting all this into perspective is that Bob here has done a great service to Open Source and Linux by making its detractors look like a bunch of stark-raving lunatics, primarily because Bob writes like one.

Bob fails to see the point of course, obviated by his constant references to "Open Sores". The guy has gone down to being nothing more than the equivalent of a USENET troll. He still has a slightly better reputation than your average /. troll, but barely. I expect this to change soon, and not necessarily for the better.

Bob, if you're listening, give up the journalism gig. You're no good at it. Your ideas might be considered good points if you didn't write like such an idiot. However, you are a smart guy and all. There's gotta be something more worthwhile for you to do!

Bob Metcalfe is GUILTY! Score him as -1, Troll!

[Kurt Gray]

Mr. Metcalfe's article is blatantly flamebait, therefore he is a troll! (jury gasps) Here is the evidence:

1. Guilty of making unreasonable suggestions: Metcalfe says...how about bundling Crusoe chips with Transmeta shares? This the trademark of a real troll: trying to stir up the sentiment that any businesses remotely related to Open Source must give away their capital to the public or they are being dishonest to Open Source. Guilty!
2. Guilty of making unreasonable assumptions about a person's role at the company they work for: ...if Torvalds "get's it" then why isn't Crusoe open source Another troll trademark: assuming every employee at company X sets the policy for every aspect of the company X's business. Guilty!
3. A bold-face confession that this article was engineered to stir up resentment from Slashdot: ...why is the sanctimonious open-source press still cheering him on? Are the likes of Slashdot.org, just gobbled up by VA Linux, also porking out in Orwell's farmhouse? This, I submit to you ladies and gentlemen of the jury, is a plain confession that this article was meant to stir up the masses of Slashdot readership and cause unhappiness in these here web pages! The true goal of every professional troll! TROLL, I SAY!
4. Guilty of trying to insite a Slashdot riot: Metcalfe blatantly asks, Where is the outrage? Oh there is outrage, that you sir are a troll!
5. Guilty of dragging Microsoft into the conversation: ...Crusoe is touted for running X86 Intel software, in particular Microsoft Windows. Doesn't the open-source community say Windows is beneath contempt? Yet again, the true trademark of a troll: when all else fails, mention Microsoft.

So you see ladies and gentlemen of the jury, I submit to you that Bob Metcalfe is a troll! He has engineered this article to flame Slashdot! What is his next article going to be about? Natalie Portman pouring hot grits down his pants?! I say to you, ignore him!! Let him feel your defeaning silence and disinterest! Please, as with all trolls, DON'T FEED THE TROLLS.

Thank you ladies and gentelmen of the jury for your time. The prosecution rests.

infoworld and doubleclick

[sjames]

Interesting, I have disabled Doubleclick on my machine by setting up fake zones in my DNS server. The page will not display. If I temporarily re-enable doubleclick, the page displays.

Final proof, re-disable doubleclick, restart netscape (to flush it's cache of lookups) and the article won't display again.

I guess that's the last article I read on their site.

As for the article itself, perhaps Mr. Metcalf is under the impression that Transmeta is a sole proprietorship owned by Linus?

As for the windows crack, it probably also runs CP/M 86, so what?

I have a great idea.

[jd]

If Bob's reading this, how about he takes a nice, looong vacation, puts his feet up and stops trying to think about all those itsy bitsy CPUs and their microcode. It's ok, I'm sure everyone on Slashdot would love to fill in for him, until he gets better. All he has to do is relax, take whatever the doctors give him, and not listen to any strange voices.

I'm almost serious, when I say that. Anyone who can't understand the difference between a CPU instruction set and an application is in serious trouble, and is in urgent need of help.

I'll accept, for the sake of argument, that newbies MIGHT get confused. But there are some important points in just that. First, is that word "might". Most newbies are smart. They may not know much, which is why they make so many errors, but that doesn't make them stupid. A newbie can figure out that if they can read English, they can read any book written in English. It doesn't take Einstein to figure out that if the computer can read x86, then it can read anything written in x86.

Then there's the "confused" part. It's understandable if a newbie can't quite figure out exactly what part the OS plays and what part the processor plays. That takes at least a little experience, which (by definition) a newbie doesn't have. However, that doesn't mean they can't guess. Sometimes they'll guess right, sometimes they won't, and sometimes they'll look things up. And everything's fine with the world.

My point? Bob Metcalfe isn't a newbie. Not even close. And yet he's making errors that even a newbie would be hard pushed to make, even deliberately. Errors such as "it runs Windows, so there's a problem with Linux" are beyond ignorance and enter a realm of surrealism and unreality that I would consider a fairly good symptom of insanity.

In many ways, this is worse than the article trying to push the blame for the DDOS attacks onto Linux and Solaris. At least, in that, there was some attempt at reasoning, however distorted and delusionary. Here, there isn't.

Lastly, Linus and the virtual instruction sets. EXCUSE ME, MR METCALFE!! Linus doesn't run Transmeta. There's no evidence Linus even WORKED on the translation layers. For all Mr Metcalfe knows, Linus could have been working on Mobile Linux (which might be construed as a little irregular, given his promise). He could have been working on the intelligent caching. Or the instruction set itself. We Don't Know! To throw the blame at someone, for something you don't even know he did/didn't do, is childish, immature and pathetic.

A new phrase

[Effugas]

People, I have this awful feeling.

Somehow, someday, I'm going to start slipping down that horrible slope that Metcalfe has.

If this ever happens, please. Inform me that I am Metcalfing. Bludgeon it into my skull if you have to.

That's not to say that every controversial opinion I'll ever hold is automatically suspect--not that my opinions are particularly huge deal by any measurement, but I actually believe *gasp* that Microsoft has made some rather valuable technical advances in their time. But if I <i>ever</i> come up with something as *brain dead, credibility destroying, and obviously flaccid flamebait* as "Open Sores", please!

Help me! Remind me of Mr. Metcalfe. If that fails, go grab something from ESR's stash ;-)

Yours Truly,

Dan Kaminsky
DoxPara Research
http://www.doxpara.com

Why does anyone care what he thinks?

[tgd]

"Technology pundit Bob Metcalfe walks in the valley of death, Open Sourcerers to his left and Microsofties to his right. He needs all the encouragement he can get at metcalfe@infoworld.com. "

I said this last time an article of his ramblings showed up on here, and its worth saying again. Just because he "invented" some technologies thirty years ago that made him marginally relavent in the industry back then doesn't mean he's at all relavent now. He's walking through a valley of his own ego, reality to his left and the world as he believes it is on his right.

Linus' financial situation

[Bruce Perens]

Well, Linus stands to make a good deal from Transmeta's IPO, whenever that happens. At the moment, I suspect he's getting a reasonable salary, not an over-the-wall one. We'll find out more about his deal when Transmeta files its S-1, but that might be a while.

Be conscious that Linus has held himself aloof from Red Hat and other companies directly involved in Linux who would gladly have paid him movie-star salaries.

If I were to be a worshipper in the "church of Linus" :-), it would not be for what he's done with software, but for how nice he and Tove are and what beautiful children they are bringing up. Having a life is #1, free software comes in somewhere after that.

Linux jobs abound these days. Can you program? If so, maybe you won't be poor as a church mouse for long. Remember, free software does not require you to take an oath of poverty.

Thanks

Bruce

Re:Linus' financial situation

[twit]

Not to be too blunt, but there's nothing wrong with giving away free software - even free as in beer, rather than free as in speech - by definition. I don't know where you got the idea that there was. What is wrong is collusion, dumping (and Open Source isn't dumping, by definition), and/or unfair business practises. IE would have been a fair competitor to Netscape if it hadn't been bundled with Windows, &c, &c, &c.

Certainly, in a capitalist society there are winners and losers. The GPL is in no way responsible for this. The GPL does not present an "unfair advantage" to anyone using it; they must give as they receive, and the advantage is given to anyone who wants to give.

Let's look at the example you gave, the small software house who lives in fear of a GPL'ed product taking away their niche. I can't see the downside of this, really: if users will be satisfied with a free product which does 75% of what the commercial product does, and satisfies their requirements, then that's all for the good of the user.

However, total cost of ownership being what it is, software cost is a minuscule part of deploying and running an application. If a vendor cannot compete with a free product offering no support and no training and no promise of future development, no corporate commitment at all, they were quite literally doomed from the very start. Unfortunately, this kind of arrogance runs rampant through the software industry: why should be offer support, training, and all those other things that let a company successfully use our product?

Likewise, Be wouldn't be giving away its OS without the existence of Linux or the BSD's, which are not licensed under the GPL. Are they now recouping their investment? Let's be reasonable, please: if they had an ounce of sense, they wouldn't have expected to recoup their investment for many years. That they're not recouping it presently, for a year-old product, is an indictment of nothing except capitalism.

For that matter, BeOS uses a number of GPL'ed tools and utilities, such as bash. Is one operating system which uses GPL'ed utilities good, and the other bad? It certainly isn't the GPL that makes one good and the other bad, one succeed and the other fail.

Believe it or not, the market, or rather any free or pseudo-free demand economy, exists for the benefit of the buyers, not the sellers. The buyers move the markets through present consumption, and they exist only to promote further consumption. If the GPL is harmful to prospective sellers, don't forget that capitalism is the root of this evil. You might as well blame the Microsoft EULA, as Microsoft has driven far more companies out of business than any number of free software ventures.

--

Re:Linus' financial situation

[Crixus]

Well, Linus stands to make a good deal from Transmeta's IPO, whenever that happens. At the moment, I suspect he's getting a reasonable salary, not an over-the-wall one. We'll find out more about his deal when Transmeta files its S-1, but that might be a while.

I read an interview with Linus a few months ago where he described his salary as generous, (I'm paraphrasing here) and probably more than he deserves.

I didn't bookmark the interview.. I'll see if I can find it. Not that it's actually relevent to the topic of the main thread.

Re:Ugh, this is even more Ill-Informed

[Bruce Perens]

Yes, I know, but I don't agree with that. I want to make instruction sets. I can change my microcode when the chip changes, too. As far as this policy is concerned, we have not yet heard the final word and I am more optimistic than you.

Thanks

Bruce

Re:Change the subject...

[Bruce Perens]

Oh, blow it out of your anonymous orifice. I said he has an old car and a rented home, and if you've looked at what homes cost around here, a "reasonable salary" just won't cut it. If you have agression to take out, school's in session again on Monday, sonny.

Bruce

Compatibility and microcode

[Bruce Perens]

Let's go through this again, just to make sure you have all of my points.

I want to the Open Source world to use this CPU to write microcode and implement new instruction sets. I actually have personal experience in writing microcode, by the way, so I know what I'm talking about.

I don't care if the chip changes. I will provide new microcode when that happens, and programs written in my instruction set will continue to run independent of which chip they are on - all chips have different microcode to implement the same instruction set.

Transmeta can accomodate this without giving away their prized advantages. Whatever patents they have are licensed for the chips they sell, thus there's a license wherever the microcode runs.

Proprietary vendors can even establish an aftermarket in microcode, selling optimized implementations of an instruction set that might even be available publicly in a sub-optimal form.

Transmeta makes money from sales of their hardware into this market, and protects those hardware sales with their patents. What software people do after the chip is sold isn't of concern to them.

So, does this make sense or am I missing something?

Thanks

Bruce

Re:Compatibility and microcode

[Bruce Perens]

Yes, but the instruction set I write might not be related to the PPC, ARM, or 68k. It might be something new entirely. Why not go through all of the operations GCC generates at its processor-independent level and write instructions that implement those operations most optimally?

Regarding the 2 years to write a bios, I'd imagine the code morphing and bootstrapping an instruction set onto the CPU is in there, but still it sounds like a long time.

Thanks

Bruce

Re:Ugh, this is even more Ill-Informed

[Bruce Perens]

No, that's not what I want. See message-id 387, below, that makes it more clear.

Thanks

Bruce

You got it backwards, Brett!

[Bruce Perens]

Oops, here comes another representative of the bitter and hurtful camp!

You got it backwards, Brett. Can you guess why I am working on the problem of modifying a license (not necessarily the GPL) for the situation of ASPs? Because some ASPs asked me to. Why did they ask me? To protect their work and their competetiveness. They want to put out Open Source software and benefit from the collaboration of other ASPs and the community. They want to create a commons in which no one party to has an unfair advantage over the other in this collaboration. So, they will put their own code behind whatever license I come up with. And you can bet that will be compliant with the OSD. It's part of their business strategy.

The GPL treats commercial software developers the same way as everyone else. In fact, only paragraph 3(c) of the GPL even has a reference to commercial vs. non-commercial distribution, and not in a way that would impair commercial distribution.

You can't understand this sharing thing, can you? Want to use my GPL code? Fine! Share it with this excellent set of sharing rules we've cooked up called the GPL. Return value equal to the value I put in, and treat me as I treat you. But you're saying no share! gimmie! I want to use the code any way I want, and not give a thing back if that's how I feel, and who cares how much work you put into it! My work is more important to me than your work, so give me your work on my terms! Go over to freshmeat.net and look at all the GPLs on new code. Every one is offering you partnership in that code, if you can just learn to share.

And it happens that yes, I am on some corporate boards. I'd rather have me there than someone who doesn't believe in Open Source, and I bet a lot of people out there feel the same way. Why am I there? I tell them how they can make money while being good citizens of the Open Source community. I do it for some of them for nothing, and others have kindly offered me some shares in compensation. No cash or sale of shares has happened to date, nor does one appear to be close.

And what's all of this stuff about depriving programmers of their livelyhoods? I've got several nice positions advertised on my company web site for people who want to code Free Software. And I'm hardly the only one with open positions. The only reason you don't have one of those jobs is your attitude - unless you can't do the work? I've never seen your resume.

If you've got a mission to carry out, Brett, it would work a lot better if you did it positively. Gee - people accuse the free software folks of whining!

Bruce

More Pretzel Logic

[Bruce Perens]

Maybe there's a "camp" (as you call it) because there's a consensus that it's so.

You, Tom, and Bob Metcalfe.

Translation: They really don't want to be competitive. Rather, they want to force their competitors to reveal the technology which might keep them ahead of the game. This is anti-competitive, not competitive.

Wrong. Their competitors are not forced to do anything unless those competitors choose to be partners on the same piece of code. Don't use the code, you don't have to obey the license. Together, the collaborators on an open source project are a bigger and better competitor.

But then, the GPL is, too. The GPL is intended to undermine programmers and hurt their livelihoods by giving away equivalents of their products for free and denying them access to that code so they can't add value.

Actually, that's not true. You can add value to your own GPL code and put it out with a commercial license. It's only other people's code where you have to obey the sharing rules. Didn't I explain this one to you last time?

So, they will put their own code behind whatever license I come up with.

Hmmm. Since when did you become "King Bruce?" It seems that, by attempting to dictate terms and restrict access to code, you are in fact becoming the very thing that open source was originally designed to avoid.

Who said dictate? They asked me to make a license for them.

No, the GPL specifically targets commercial software developers by attempting to force them to give away the fruits of their labor. Uh, excuse me, but there aren't many poor GPL Linux programmers these days. Is this a particular personal income problem?

True sharing means sharing with everyone.

You've almost got that right. When I give you GPL code, you must share that with me and everyone else equally. When I give you code under the X11 license, for example, you need not share it equally with anyone. You are allowed by the X11 license to just take, take, take. That doesn't sound like sharing to me.

If you are ethical

Here you go accusing me of being unethical again, Brett. It doesn't shed a good light on you.

Bruce

Optimization

[Bruce Perens]

Agreed.

What happened to Bob Metcalfe?

[Bruce Perens]

Bob Metcalfe used to be a good guy. When I first started doing business with 3com back in the VAX 780 days, Bob was returning my calls to their customer service phone line. Bob's made Billions on 3com and the Ethernet and should not have reason to feel bitter. But he keeps putting out this bitter, hurtful, and poorly informed stuff. There's little to criticize on a technical level in this piece because it is so obviously sour grapes, and for Mr. Billionare to pick on little Linus Torvalds with his old car and rented home just doesn't seem right. For Bob to pose as a "journalist" is sort of silly if he's never going to take the trouble to learn much of journalism, and instead settle for being just another crumudgeon.

What we want from Transmeta is a microinstruction set description. Period. We don't ask hardware manufacturers for their chip masks - what we ask them for is full documentation of how to use the hardware from our operating system. Frankly, we can do our own "code morphing" given an instruction set description and a patent license that goes with the chip. And it's not even clear that we'd need to do code morphing from something like the x86 - I think we'd be much more interested in designing our own instruction sets to work optimally for Linux.

Thanks

Bruce

Re:What happened to Bob Metcalfe?

[ToLu the Happy Furby]

Frankly, we can do our own "code morphing" given an instruction set description and a patent license that goes with the chip. And it's not even clear that we'd need to do code morphing from something like the x86 - I think we'd be much more interested in designing our own instruction sets to work optimally for Linux.

Much as I respect you Bruce, and as impressed as I am with the OSS movement, this is precisely the reason why Transmeta doesn't want to open source their code or their native instruction set description. One of the main benefits of using code-morphing is that Transmeta can change the native VLIW instruction set on their chip very easily, without breaking compatability with any existing programs.

Indeed, the 3120 and the 5400 have completely different native instruction sets, IIRC. Any future Transmeta processors are likely to have their own native instruction sets as well. Even so, you'd better believe that if Transmeta went ahead and released complete specs for their chips, there'd be a bunch of open-source hackers who, desperate for that extra 20% performance fix, would code their own little Linux-TM3120 (which, considering the fact that it would probably require writing a version of gcc optimised for the 3120's VLIW instruction set, would be pretty darn nontrivial). Another bunch would have to do the same amount of work to come out with Linux-TM5400. And meanwhile, by the time either of these projects was finished, Transmeta would be shipping a bunch of new processors with completely different cores.

The point is, Transmeta forsees itself as finally solving the pesky problem of backwards-compatibility--but at the price of a small performance penalty. The important thing here is, they have to act a little bit like benevolent dictators to do this: in order to give us all something we want--freedom from compatibility issues--they have to take away some of our freedom elsewhere, by not releasing their native instruction sets.

The reason why this is so is because of that fundamental bane of OSS--people write OSS to scratch an itch. Now, usually their itch is an itch shared by many others--that's why OSS can have its phenomenal successes. However, in some cases, an one person's itch-scratching may be in conflict with the general software-using community as a whole.

And this is such an example. That is, if I just bought a TM5400-based laptop, and I want to run Linux on it, then I suddenly have an itch--the version of Linux I'm running on it isn't as optimized as it could be, because it's running in emulation through the code-morphing layer. The problem is that by scratching that itch--that is, writing Linux-TM5400--ruins a lot of other people's day. Specifically, it removes the benefit of backwards-compatibility for all.

All in all it's a pretty interesting issue, and maybe in a few years if Transmeta's way of doing things becomes more entrenched I might change my mind, but I figure that for the time being Transmeta's way of doing things is the right one, and I'd bet that Linus supports it all the way. After all, Linus has always said that he picked the GPL as the license for Linux not out of ideology but out of pragmatism--not because he thinks that all software ought to be open-sourced, but because he realized that Linux might be useful to others and become mildly successful if it was. While neither you nor I know exactly what the consequences would be if Transmeta open-sourced all of their (relevent) intellectual property, a much stronger case can be made that the result would be detrimental to the success of the Transmeta chip (I mean in terms of adoption, not Transmeta's financial success) rather than beneficial.

Yes: open-source Code Morphing to save Transmeta

[Morgaine]

The writer made some really silly points, like the Transmeta hardware being open-sourced and given away for free. It's silly because you can't replicate hardware for nothing, at least not until nanotech reaches maturity. But a few of his conclusions were sensible, despite poor reasoning in arriving at them.

The Code Morphing Software *can* be replicated at zero cost, so the argument then hinges on whether you understand the positive effects of open sourcing or are tied to the traditional views of proprietary developers.

I don't have any doubt that Linus believes in free software, and if he does then he will have fought long and hard to get Code Morphing source released. It is very likely that he simply failed to convince the suits.

That's tragic for Transmeta, in my view, because the reasons that they gave for keeping it closed seem totally flawed: they would not lose the ability to change the underlying hardware arbitrarily, because they would need to integrate only those changes they see fit into their internal version, and on releasing a new chip then the onus would be on everyone else to catch up. No, that's just an excuse put out by the suits.

Open sourcing the Code Morphing Software would be extremely good for Transmeta: the quality of their code would improve for the usual OSS reasons, it would have numerous new features added by the world's gurus, the external development would cost them nothing, it would get ported to dozens of non-x86 instruction sets very rapidly, and perhaps most importantly of all, thousands of developers would buy into the CMS idea straight away. That's a priceless package.

And of course, meanwhile the only people to supply the underlying hardware would be Transmeta. They would become collosal, the next Intel but bigger, since their hardware can in principle subsume that of all other computer manufacturers.

Instead they're going to die in a few years' time without trace, just because their suits don't understand OSS. A pity.

Re:Yes, /.ers, Bob Metcalf Does Make Sense

[kevin lyda]

your last point is incorrect: intel did demo nt. not at linux world, but at their initial announcement.

why he writes this stuff...

[kevin lyda]

it's not well known, but bob metcalfe runs the mail server performance under load benchmarks in conjunction with slashdot. it's a simple formula:

let p be the number of posts in 4 hours on this story
let P be the number of posts in 4 hours on all stories
let t be the average time for mail delivery during that period
let T be the average response time normally.

let r be p/P*10 - the rating of the story
let d be t/T - the amount of degredation

so for a /. story magnitude r, the mail server responds in d times the normal amount of time.

Re:Linus & Transmeta: A match made in heaven

[stevew]

Crusoe isn't revolutionary, but rather evolutionary. It isn't the first VLIW machine by a ways. Rather, it is their decendant. It DOES fix what has been the achilles heel of VLIW - that you had to recompile for every addition of a new functional unit. It recompiles the code - on-the-fly!

As for the article - he's right up there with Algore, i.e. the inventer of the internet as far as his current prognosications....at least Metcalfe can make SOME claim to being involved in networking ;-)

Re:Bob Metcalfe joins the tabloid press

[otis wildflower]

Do you think his advertisers appreciated out visit?

His advertisers didn't know I was even there ....

(have you junkbusted lately? Damn it feels good...)

Your Working Boy,

Re:Mac OS on x86??

[Uruk]

He didn't say "MacOS" he said "MacOS 8". I think the current incarnation in 7.something, which would lead me to believe that the whole Darwin business is going to be MacOS 8. And that would make sense, since they'd be starting with a codebase that runs fine on x86 though.

But if the original poster wasn't talking about Darwin, then I'm with you on this one. WTF? :)

Semiconductor manufacturing 101

[Booker]

Chips have to be manufactured -- with white coats, ovens, and stuff

I work for the world's largest manufacturer of semiconductor equipment. Bob's analysis of the semiconductor manufacturing process is laughable. "Ovens?!"

I guess that fits in pretty well with his overly simplistic view of source code, CPU architecture, and everything else he wrote about in this article...
----

Not a new phrase

[tilly]

When the second release of BO was GPLed, the Cult of the Dead Cow presentation involved someone running from the room yelling, "Open sores!"

So no, Bob didn't invent that phrase.

Cheers,
Ben

PS OTOH I expect in time to see tyop show up in a dictionary....

.5 whoo-hoos

[Signal 11]

Rather than blow of our esteemed mr. metcalfe I'd just like to point out a couple things that actually are relevant to this... because even though the article is of poor quality, the proprietary-in-an-open-source-company question does need to be addressed.

Hypothetical situation: You discover a way to make a processor 1500x faster than existing processors, zero heat output, and runs in a few milliwatts. Do you:

A) go public and tell everyone else how to do it.
B) Sell it to Intel, who buries it in a landfill never to be seen again.
C) Patent it, then go public.
D) Don't tell anyone - Moore's Law Must Be Upheld. (Only Intel may circle this box).

You see, if Transmeta released it's internal chip specifications now they'd be hosed because other manufacturers would then be able to produce clones... given Intel's vast resources this was Transmeta's only option. It was either that, or put up with an Intel chip that did code morphing but got floating point wrong 99.999938471% of the time. This is also Transmeta's way of protecting revenue - software is run by a different set of rules... try to keep this in mind when judging them.

Uh, where has Bob been?

[Accipiter]

Did he even watch the big release from Transmeta?

Has he even heard of Mobile Linux? Apparently so, because he mentions it in his column. What about the Quanta Web Pad? That doesn't run Windows. Wake up, Bob. This is an x86 processor. (Duh, what's that mean?) Meaning, you can run anything your spiffy Pentium processor runs, INCLUDING Linux, and INCLUDING (But not limited to) Windows, DOS, and every other x86 OS/Program out there.

For a start, why aren't the Crusoe chip's mask sources published for modification and manufacture by anyone?

Would you want everyone to have the code that runs the hardware of a ton of computers? How about crackers who would love to find a way to cripple your processor, all because they have the source code. It's not necessarily a matter of Open Source, it's a matter of security. You can draw the comparison by saying "Linux is open source, and Linux is secure." Sure, but Linux is software -- easily fixed. Are you going to get a new processor every time an issue comes up? Sure you can flash the ROM, but what if an issue comes up that makes your processor useless? How are you going to flash it THEN?

Worse, Crusoe is touted for running Intel X86 software, and in particular, Microsoft Windows. Doesn't the open-source community say Windows is beneath contempt?

Does that erase the fact that Windows is used by most end users who own a computer? Yes, Windows is a piece of crap but it's a POPULAR piece of crap. Does it make sense to build an x86 processor, and remove Windows compatibility because you think it's crappy? Sure, let's remove an entire market from our product because of personal opinion. Get real. Oh, and there is no "particular" about it. It's a processor for God's sake. It doesn't run one piece of software any more "particular" than any other software.

Where is the outrage?

I'm guessing this dude WANTS outrage, and if he doesn't see it where he wants it, he writes articles like this to generate it.

So just to keep Torvalds honest, I'm thinking that Crusoe chips, which are mostly software, should be open source and basically free.

Since when is Linus in charge of Transmeta operations? He works there, and has a hand in development. He's not the company's marketing department, CEO, or anything else like that. He works there. So this affects his honesty how? (I'm thinking Mr. Metcalfe isn't too honest either, be it intentional or not I don't know.)

I must ask this guy to grab a clue. He shouldn't be writing articles... (But in this screwed up world of ours, he is. And in this screwed up world of ours, people listen to him.) It's a shame really. Maybe Bob should go to work for Microsoft. Their marketing department is expanding.

-- Give him Head? Be a Beacon?

Re:Closed hardware specs. and security.

[Accipiter]

I never said closed source was more secure, my point is open source is easier to find problems with. (Which is why Linux is fixed so fast when bugs arise.)

-- Give him Head? Be a Beacon?

Read "The Magic Cauldron" by ESR.

[Goonie]

It contains a obvious, but accurate, observation of the reason why Transmeta keeps its code morphing software secret:

When the rent from secret bits is higher than the return from open source, it makes economic sense to be closed-source. When the return from open source is higher than the rent from secret bits, it makes sense to go open source.

The code-morphing software is the key to the Transmeta chips' superior features, so I'd say the rent from keeping the bits secret is pretty high.

RMS would probably argue that Transmeta has a moral obligation to free its software, but the pragmatic net payoff of a lot of the open-sourcing we've seen just isn't there in this case, and it's not realistic to expect a company to do so without it.

Re:Bob Metcalfe joins the tabloid press

[elflord]

I disagree that open source is necessarily communist.

Linux embraces many parts of the Marxist ideology, but it's not really "communist". "Communism" is a word that has been tainted by abuses and wrongs committed in it's name.

At the same time, Linux offers the anarchic freedom one associates with Liberterianism, though Linux certainly isn't capitalist. In a way, Linux offers the best of both worlds.

Check out this excerpt from Homesteading the Noosphere: there definitely is a strong sense of ownership within the community.

Hate to say it but I don't buy this. Open source is inherently anti-property in that it weakens what one may do with their intellectual assets, and in particular, it makes it hard to manipulate those assets for profit. The "rogue patch" thing has no legal ground. There's no legal ground for one patch to be called "rogue" and another "official". A succesful fork will turn the tables here. Eric's just having a hard time explaining why Open Source is all about his own political beliefs when in fact it is not inherently rightist or leftist.

Re: Not really communist or libertarian

[elflord]

Communism has been tainted by the wrongs committed in its name. And many definitions of communism have it defined as a big government thing, which Linux certainly isn't. However, Linux ( esp the GPL ) does embrace certain aspects of Marxism. "Each gives according to his ability and takes according to his needs". The very idea of sharing code instead of manipulating it for profit is very Marxist.

However, Linux offers something that sample implementations of communism have failed to do -- freedom. This is why Linux appeals to those with strong libertarian sentinments. My feeling is that Linux really brings in the best of both worlds.

Re:Bob Metcalfe joins the tabloid press

[elflord]

You are assuming that intellectual property is a valid form of property. Not every one accepts this,

Not everyone even accepts that there exists a valid form of property. Those who believe exclusively in weaker notions of "property" are still anti-property, but less so than those who don't acknowledge any concept of property.

Intellectual property is corollary of capitalism. Take away intellectual property and you remove financial incentive to produce intellectual works, including software. It is awfully hard to honestly say "I'm a capitalist", then say in the same breath "I don't believe in intellectual property".

The concept that OSS even apporximates Marxism is rediculous.

I didn't say that it did "approximate Marxism". You failed to understand my post. A better way of putting it ( as some other guy did ) is this -- "Open source is voluntary". It "approximates" volunteerism

Re:Bob Metcalfe joins the tabloid press

[elflord]

Capitialism existed very well without the concept of intellectual property until the 1800's.

Well IMO, capitalism really didn't "exist very well" prior to the 1800's. Most capitalist societies of those days were barbaric by todays standards. There was no notion of a "publically held" company ( they were just moving towards that around 1800 ).

Moreover, there were many differences between the 1800s and nowadays. Piracy is much easier now than it was then. It was much less of a problem in the 1800s, since duplicating media was not very easy until quite recently ( try burning an LP on your hi fi system some time ... )

This has little to do with beliefs

[SurfsUp]

It's all very good to want a return on investment, especially in the hideously expensive task of designing microprocessors, but if you?re going to hold a belief, then you need to live that belief, regardless of the consequences.

I'm of two minds about this article. First, I agree with its premise: that Transmeta should open-source its code morphing software. Not to the extent that competing companies would be able to use it for free, but to the extend that the worldwide hacker net can go to work on it. Second, I think it's a stupid article, and it's written in a pretty stupid way. Give away the processors for free. Yeah right.

All this has very little to do with beliefs. OK, the gnu organization started because of beliefs, but the open-source movement continues because it produces more and better software. That translates in $$$ by the simple rule that if you give customers what is best for them, i.e., cheap and good, they won't be slow to pay you for it.

Open sourcing the code morphing software is something that Transmeta should do because it makes business sense. Again, very little to do with beliefs, unless you count getting rich as a belief. Transmeta wants to sell processors - whatever they can do to sell more processors makes good business sense. Especially when it doesn't cost any money. Opening the code morphing source will sell more processors. It will make the software better. And it will help establish a loyal, perhaps fanatically loyal, core user base.

So why hasn't it happened? Well, I can only speculate that somebody in the executive suite doesn't get it. Yet. It's just a phase that will pass. With Torvalds in there to make the case I'll give this, hmm, six months, and I guess said suit will see the light. In the meantime, the pressure is going to increase. So far Transmeta is on a honeymoon with us - they can do no wrong at the moment and a little thing like the code morphing source code isn't going to dampen our enthusiasm for this technology. Much. But that's going to change - you'll see more frequent and more emphatic calls for a policy reversal on the open-source issues.

After a while we'll get what we want and Transmeta will be better off for it.

Re:Bob Metcalfe joins the tabloid press

[Fyndo]

I think you may find two distinct camps in the open source world: The socialist-like view that we should all help each other and give our source away for free. The other camp are the capitalist pigs that see good business sense in open source. The first camp I don't understand, the second, which I subscribe to, has a lot of profit potential.

I think, despite my liking of capitalism, you're going to place me in the socialist camp. Really, I don't think I'm in either, nor do I really think the two exist, or are as irreconsilable as you seem to think.

So first, I'll make my defense of Capitalism from a more socialist viewpoint. Capitalism has the advantage of being (at least in the ideal, and to a closer extent than any other system in preactice) what is termed by economists, Paredo-Optimal. This means that the allocation of labor and resources is such that no two people have bundes of goods/services that they would gladly swap. i.e. cases where I have 8 chickens, and would rather have a cow, and you have a cow, and would rather have 8 chickens will be eliminated. (well, the transaction may go through middlemen, and may get lossy, but you get the point).

Capitalism is not, however, anything goes, it does involve rules (i.e. the transaction "you have 8 chickens, and I have a big stick, give me the chickens and I won't hit you with the stick" is verboten) chosen to make sure things keep working this way.

Enough Econ 101, what about Free Software/Open Source? Well, basically, things with software are no longer paredo optimal. I have a copy of MS word, you have a copy of MS excel. I have the labor/materials required to copy MS word, and would gladly trade those for a copy of Excel. You have the labor/materials to make one, and would for a copy of Word. Copyright laws don't let us do that.

Copyright laws (presumably) exist because without them, if spent lots of money/effort writing software, and then everyone just copied it without paying me, I'd lose out, and not being a dunce, wouldn't do it in the first place. So copyright laws prevent me and you from trading MS's stuff, to give MS an incentive to make software.

However, what happens, it's easier to make software in the first place, if you can copy bits from other peoples software... So in software, as the free software/open source software movement demonstrates, software can be created without these incentives, since removing them (allowing copying) makes software cheaper.

So the economist would say "Copyright restrictions on software create an unnecessary inefficiency in the software market", the capitalist would say "Open Source Software allows us to make software better and for less money, gotta be a buck in that somewhere" and the err... "socialist" would say "Free Software lets me share software with my neighbor, so we all benefit"

Know what? They're all right :)

Uhhh... No?

[Fyndo]

A whole bunch of things I disagree with, but I'll just address two:

In other words, what you are encouraging is collusion among some vendors to destroy others. This is anti-competitive behavior, Bruce, and is highly illegal as well as unethical.

Collusion is only illegal if it is done in restraint of trade. The GPL is non-exclusionary. How are you going to drive the other vendors out of business by giving them better software?

How's this for an example... Take two companies, we'll call one..... Bruce. We'll call the other.... Eric. Bruce and Eric both have patents on processes used in their industry. They realise they could both make much more money if they had the right to use the other's patent. They agree to cross-license their patents. Another company comes along, we'll call it.... Brett. Brett comes up with a process that is better than Bruce and Eric's processes, but dependant on both. Brett wishes to use this new process, so approaches Bruce and Eric asking for permission to use their processes. They reply "sure, if you let us use yours".

Any restraint of trade here? No more than the patent system already imposes. If Brett wishes to use Bruce and Eric's Processes, they're free to license them. If they don't like the terms under which they're offered, they're free to refuse Bruce and Eric the rights to the improved process.

This sort of patent cross-licensing goes on every day...

The problem is that, since the initial code was licensed under the GPL, the code which is contributed back is licensed exclusively under the GPL because that's the license under which the contributor got it. Thus, if the original author accepts a single contribution, his whole work is irrevocably licensed ONLY under the GPL and his ability to legally dual-license goes away.

No, the code he returns to you is licensed back under whatever terms he wishes, and can legally offer. If you, as the copyright holder, are willing to allow him to release his derivative work (contribution) under some other terms, he may legally do so. Those other terms may be that in return for getting paid, he gives you the right to distribute his work as you see fit.

If he refuses to license it to you under any other terms than the GPL, then no, you can't distribute it under any other terms than the GPL. Of course, that sentence is still true removing both instances of the phrase "under any other terms than the GPL".

Just ignore this guy.

[finkployd]

He's sad really. He seems to hate open source with a passion, but is powerless to do anything about it. He doesn't understand most of what he writes about (given the glaring technical inaccuracies that often pop up in his writing), but he posses strong opinions anyway.

Don't flame him and don't write him explaining x86. Just let him sink into obscurity along with everyone else in this industry who cannot actually accomplish anything anymore, so they are forced to write about others accomplishments.

Finkployd

open source == capitalism

[jetson123]

Of course, Metcalf got it completely wrong likening open source to communism.

Microsoft is making huge profits. If anybody makes huge profits in a capitalist system, that's an indication of market inefficiencies. Open source eliminates those inefficiencies: it's an efficient way by which consumers, who currently pay unreasonably high prices for Windows, force prices down.

Open source is a completely natural, capitalist result of the cost structure of creating and distributing software.

Is Crusoe optimized for Windows?

[stx23]

Metcalfe:-Worse, Crusoe is touted for running Intel X86 software, and in particular, Microsoft Windows.
smeng58:-"Sad when a computer pundit appears not understand what x86 code is."
I think he has a fair understanding of what the code is, he seems to be infering that Crusoe will run Windows specific code optimally. Is it unrealistic that code-morphing could listen out for specific Windows calls and to translate them into a more efficient single call to the processor?
Say, for instance, a chip subroutine exists which can instantly trap a BSOD without the crash getting to the kernel, to allow you to save the machine state for recovery? What use would that be in Linux?

Re:Is Crusoe optimized for Windows?

[bssea]

The 5400 is the only processor that has been stated for use with Windows. That's because the 5400 has 16-bit code support in the software, whereas the other doesn't. That was stated at the Webcast!

The 5400 will run Linux just as effectively. The only difference is Linux won't use the 16-bit code support.

And about the fragment that Transmeta will do... The "Mobile Linux" extension WILL be made open source, this was also stated AT THE WEBCAST! I see no freagmenting in Linux coming from Transmeta.

Metcalfe really seems to have NO clue what the hell he's talking about. He apparently doesn't understand the OSS movement or anything he himself didn't help in inventing.

Re:Well...

[jawad]

It's amazing the double standard that most Slashbots have. If this was another company, they'd be flamed. But because Linus works there, it's all good, right? Err...

Re:Well...

[jawad]

The "double standard" I was referring to was referring to ravigan's comment, "Uhhh....cause they worked pretty darned hard on it, and they want to make a profit? jeez, we do live in a capitalist society...". I wasn't referring to Metcalfe's article, which is filled with the B.S. you refer to.

The double standard is that of Slashbots wanting to get the code to every bit of software. Every story with a company keeping their source code (besides this) is peppered with comments saying "well, they should just open the source, and everything will be alright" etc. Now that Linus is involved, everything is alright? I think not.

Re:Bob Metcalfe joins the tabloid press

[Roundeye]

It has been stated before in this forum and others that the last coherent thought Bob Metcalfe had was when he helped design ethernet (a technology which would have come about in much the same way regardless of who announced it first). He has been riding that wave to spew incoherent babble for years, and is basically an annoyance to the Thinking Reader. Yet another gem in Metcalfe's crown of idiocy.

Proprietary Inventions and GNU--remember RT-Linux!

[TeknoDragon]

Hey, hey... but as you might realize linux isn't anything new! AT&T did it first...

I don't think it's so much about hardware vs. software, but REVOLUTIONARY technology vs. MUNDANE technology.

You see the linux core for all intents and purposes of the cutting edge isn't really revolutionary, and it doesn't NEED to be! Now every other OS has a journaling file system, ok let's build one for linux. Firewalling, ip chains, etc etc... I'm not a hard core linux developer so I'm party speaking from ignorance, but does ANYONE know of a significant technological advance that linux has spearheaded??

So ReiserFS is pretty kewl, and xscreensaver kicks ass. Bash is a masterpiece and mozilla is on it's way in. There are hundreds or maybe thousands of extremely high quality pieces of ART that the free software community has produced. But there's nothign revolutionary about it.

Now we've got the brains behind it... and we're making steady progress... so what happens when one of us makes up something EXTAORDINARY?! Do we give it away for free and be so do we take Metcalfe's stereotype of the Open Source community and say: "for the greater good i give up all my worldly posessions so that someone can exploit me"???

No! we do what anyone convinced that they have produced a genuine invention would do: develop it in secret, patent it once it works, sell it to the highest bidder (or market it yourself), all the while integrating support for Linux and maybe even giving them a free lisence for it!

In 8 to 12 years when you've made your fortune and your utillity patent is up for renewel, you let it expire. And hey, don't forget that patents aren't incompatable with the GPL! Hey you could GPL your patented invention and have a legal double-whammy for anyone who violates your patent!!!

Hey, if RT-Linux is revolutionary, or if code morphing is really revolutionary (sounds just a bit like a beefy emulator/finer grained scheduler on a better CPU, but I can't honestly know if it was that original or not), or if some little project i'm workin on happens to light a fire under whatever market it's viable in and we patent our inventions... this does not mean that we're sellouts!

if Metcalfe posted on slashdot:

Re:Crusoe Code Morphing (Score:-1, Troll)
by Bob Metcalfe (metcalfe@idg.net) on 1:01 Friday 11 February 2000 PDT (#0)
(User Info) http://www.infoworld.com/op inions/morefromtheether.html

Nah, nah, OPEN SORES!!! Linus Torvalds Naked and Petrified eating goat cheeze and being a hyppocrite!!!

[ Reply to This | Parent ]

Yes, /.ers, Bob Metcalf Does Make Sense

[shambler snack]

"InfoWorld's Bob Metcalfe asks why, if Linus Torvalds is truly a believer in Open Source, Transmeta Corp. has seen fit to make Crusoe, or at least its VLIW "code morphing", proprietary. The column goes on to say that, since the processor will run Windows code, there must be some thing wrong with Linux. Sad when a computer pundit appears not understand what x86 code is."

I think that Bob Metcalfe has a strong grasp of x86 code. Code morphing, in and of itself, is nothing new and is a part of both Intel's and AMD's advanced processors. If I'm not mistaken I believe AMD refers to the product of converting x86 opcodes as RISCops, in which the internal machine is very RISCy. This code morphing is hardware based, and is limited to one instruction set (x86) being morphed to the internal RISC opcodes. Transmeta's value added is that they exposed and optimized the hardware layer hidden in the Intel and AMD processors in such a way that it become possible to emulate nearly any current processor instruction set.

No, I support Metcalfe's pointing out the strong smell of hypocrisy surrounding certain actions by some in the Open Source community. Let's go over some of Metcalf's points:

• "So what I want to know is, if open-source software is so cool, and if Torvalds "gets it," why isn't Crusoe open source? For a start, why aren't the Crusoe chip's mask sources published for modification and manufacture by anyone?"
  
  SUN Microsystem's own website (http://www.sun.com/microelectronics/communitysour ce) contains links to open Community Source Licensing to its picoJava and SPARC V8 processor cores. If SUN can do it "imperfectly" under SCSL, then why can't Transmeta show how it should be done with a hardware-centric version of the GPL?
  
  
• "And yes, Mobile Linux is open source, but not the "code morphing" software Torvalds helped write. Transmeta has taken the phrase Code Morphing as its proprietary trademark. And what the code does, according to Transmeta, has been ... patented."
  
  Yep, there's that ugly serpent in the Open Source Garden, software patents. How many articles have been published, and how many flames delivered, to the "clueless" individuals who would dare to tie up innovation and hamstring the inevitable march to victory of the Open Source Movement?
  
  
• "Worse, Crusoe is touted for running Intel X86 software, and in particular, Microsoft Windows. Doesn't the open-source community say Windows is beneath contempt?"
  
  What more can I say? Intel comes out and shows Linux ported and running, first on Itanium emulation, and later on Itanium silicon. Further, the Trillium group, to great fanfare, releases the kernel source. Notice that Windows NT wasn't anywhere around, because Windows NT, being "beneath contempt", wasn't worth the effort to port and show running at the various Itanium showings.
  

Linux Torvalds may not have intended this, but he has taken situational ethics to new ground with his employment at, and tacit support of, Transmeta. It's all very good to want a return on investment, especially in the hideously expensive task of designing microprocessors, but if you're going to hold a belief, then you need to live that belief, regardless of the consequences. If the Open Source movement and philosophy are strong enough that companies such as Red Hat, SuSe, TurboLinux, VA Linux, and others are willing to build a business around it, then Transmeta, with Open Source's icon as an employee, should be out in front of everbody else. Instead, they cynically use Linus to garner interest and at the same time to shield themselves from criticisms of the company's behaviors.

Re:Bob Metcalfe joins the tabloid press

[Surak]

Errmm..not that it matters at this point, but:

GPL code is owned by the copyright holder. In the case of GNU software, this is the FSF. But in the case of Linux, for example, there are many copyright holders, but the principle copyright holder is Linus Torvald's, so he said to "own" the code.

Re:Bob Metcalfe joins the tabloid press

[Surak]

I disagree that open source is necessarily communist. Many others including myself and most notably Eric Raymond, have asserted that Open Source is really more Libertarian than communist.

One of the key issues in this difference is ownership. Take the WordNet definition:

1. a form of socialism that abolishes private ownership

On the other hand, the Open Source community does have the concept of ownership. Check out this excerpt from Homesteading the Noosphere: there definitely is a strong sense of ownership within the community.

Plus, in general, the Open Source community is tolerant of capitalism. Companies like Red Hat, VA, and Transmeta are cheered on, despite their obvious capitalist nature. Yes, it is expected that these companies will give back to the Open Source community, but the idea of making money on Open Source software is encouraged rather than discouraged, as it would be in a a truly communist society.

Mac OS on x86??

[Evro]

"X86 = processor platform. Solaris, Mac OS 8, Linux, BSD, Windows, BeOS, DOS and others run on x86. "

Mac OS runs on x86? Since when?
___________________

Current MacOS is 9.0

[Evro]

The current MacOS is 9.0. Darwin is going to be part of MacOS X (ten). If he had said "Mac OS X" he still wouldn't be right because there is nothing other than rumors to suggest Apple is porting MacOS X to any other platforms. Supposedly these rumors have been substantiated somehow, but I'll believe it (Mac OS on anything other than Motorola/PPC) when I see it.
___________________

Re:Bob Metcalfe joins the tabloid press

[crush]

I know I'm probably going to get an automatic negative karma for daring to disagree with libertarianism, but I'd like you to consider this objection to the following statements:

I disagree that open source is necessarily communist. Many others including myself and most notably Eric Raymond, have asserted that Open Source is really more Libertarian than communist. One of the key issues in this difference is ownership. Take the WordNet definition: 1. a form of socialism that abolishes private ownership On the other hand, the Open Source community does have the concept of ownership.
Do you exclude GPL'ed Free software (the kernel and most of the utilities and a large number of applications) from the set of code that is Open Source?
No? OK, then - who owns the GPL'ed code sitting running your machine at the moment?.
You may well be correct that the Open Source people are more keen on allowing private ownership, in fact that's what many suspect them of (else why change it from Free if not to shift the emphasis?), however as it would appear that somehow or other Free Software is included in this you can't get away from the fact that Free software is communally shared - there is no single, hoarding, controlling owner. This is much closer to the idea of communism that libertarianism. Under libertarianism you would be negotiating with several other thousands of coders for the rights to use their contributions and would be tied up in insanely complex individual arrangements. Libertarianism is taking the valuable idea of individualism to a ludicrous and non-workable extreme.

Re: Not really communist or libertarian

[crush]

Communism is really nothing more than the principle that resources are to be considered common property. There is no morally acceptable rationale for someone being able to sequester resources. The USSR, China, Vietnam, N.Korea etc etc are all implementations of state capitalism which is a specific attempt to implement the ideal of communism through this mechanism - or so they claim. To claim that the idea of

Each gives according to his ability and takes according to his needs" is specifically Marxist is not correct. That idea is much older, and is more properly termed communism. So, the order of inclusive sets would be:

• Communism
  There are very old traditions of the principle of holding things in common widely spread across different ethnographic groups
• Socialism
  There were plenty of theorists on the scene before Marx cropped up who considered the obligations of indivduals to each other that forms a society
• Marxism
  Anarchism is an alternative at this level of the typological hierarchy. The most vociferous critic of Marx was Bakunin who believed that Marx was too authoritarian and that in paying too much attention to the end goal of communism Marx was falling into the trap of using methods which would ultimately make this non-acheivable. This lead to the split of the First Working Men's International
• Bolshevism
  Bolshevists were the majority of those that adhered to the principles of Marxism. They believed in immediate revolution, the use of force and the dictatorship of the proletariat. The alternative was Menshevism which disagreed on these points and others to varying degrees. Other alternatives included (I think, I'm not sure how much of Marxist ideology they accepted) the ideas of the Social Revolutionaries (I get the impression that they may have been essentially social-democrats but I'm really ignorant on their policies)
• Marxist-Leninism
  Alternatives at this level are Democratic Socialism and Trotskyism. The former asserted that there was no necessity for a revolution and that change could be acheived through the mechanism that had been set up by the capitalists to control society - representative democracy and the law. Trotskyists are hard for me to pigeon-hole as they adhere to most tenets of ML that I can see yet appear to have been even more authoritarian
• Maoism, Castroism and Stalinism
• We all know the effects of these, but they differ in many respects from a strict interpretation of ML. However they are more like it than not in the sense that they are
• command economies in which there is nominal (only!) ownership of the means of production by the workers. The nice government holds things in trust for the workers who have yet to reach "political maturity" and develop a "political consciousnee" at which point everything gets turned over to them (!). It is interesting that this "stewardship" of the resources by the Party was implemented by the Bolsheviks very early on even though there were fully democratic Soviets where wokers had organized themselves and were running factories without any need for outside interference. The Bolsheviks didn't trust them and persuaded them to relinquish their power.

  Anyway, my short point is that the GPL results in a situation like the simplest, original meaning of communism. It's definitely not Marxism and it's not the bogeyman that the US has been fighting for years.

Just fine.. but..

[mindstrm]

1) Linus is only an employee of Transmeta.
2) Linus has contributed greatly to OSS, before we even had this OSS evangelist movement.
3) X86 != windows. Linux will run on these things just fine.
4) The code-morphing is brand-new stuff they have invented. What I want to know is.. have their patents been granted? I remember they applied...
if and when the patents are secure, I would imagine they may release more information on the actual chip itself.

Mecalfe, you take the short bus to work, don't you

[mindstrm]

1) Linus doesn't control Transmeta. He is an employee.
2) Linux runs on Transmeta processors just fine.
3) Linus is not 'now in control' of Linux. He has final say on kernel builds, anyone who wants to could maintain their own branch of kernel development, but the fact is, the core kernel developers *like* having linus maintain the tree. In other words linus 'controls' it because nobody else wants the job, which in the OSS world, translates to nobody thinks they can do it better.

4) comparing this to animal farm really stinks. If what you say made any sense, I would say it's insightful.. but your article is just full of inaccuracies. VA bought slashdot? Yes.. and taco & hemos made *SURE* that they had *complete* editorial control over slashdot. THEY decide what happens to it, not VA, VA just gets to say that they own it. It adds percieved value, which is very important for a publicly traded company.

*LOTS* of OSS developers have real jobs, for real companies who do some proprietary software work.. why don't you go tear a strip out of them?
As for windows being replaced..you are exactly right. Anyone is free to develop something better. And where does our anger come from, Bob? Those of us who are *very* informed, find that Linux (or indeed, any unix) is MUCH better, in many circumstances, as a server platform than Windows. Our anger stems from the fact that we have this fucking WALL that MS built that stops us from using it.. our managers question us, make us look like zealouts, and have managed to turn the whole NT -vs- Unix game into a political one, not based in reality at all.
So get back on your bus and go back to the institute.

What Metcalf is Overlooking ...

[Crutcher]

Yeah, He's a fool, we know that, but the solid argument against his craziness is:

1) Open Source development is cool (read: cost effective and resource efficient) when it allows a previously existing pool of inhouse support staff accross industry to become the primary development source for a peice of software, breaking the cost up to tinny bits everyone can swallow.

2) Chip design is NOT such scenario, their are no developers which could aid Transmetta in this field that do not work for their direct competetors.

Long term, I see 'code morphing' as becoming a basic part of software development, but it would be insane for Transmeta to dump that much cash into a development project and then just 'give' it to Intel.

The important thing is that they have proved this software 'CAN' exist, and work well. So lets get the CS theorists to figure out the general case, and then you can have your open source code morphing.

No, and Not in this Case

[Crutcher]

>> Long term, I see 'code morphing' as a basic part of software development,
>> but it would be insane for Transmeta to dump that much cash into a
>> development project and then just 'give' it to Intel.
>
>Isn't this the founding theory behind Open Source? Write software and release it without any cost to the consumer in the hope that they will
>A) Use it
>B) Take the source and improve upon it
>
>The amount of time and development that goes into some open source projects has been staggering. It has been proved that a company can survive, not off of the revenues of its software sales, but off of other services such as support or advertising.

The founding theory behind 'Open Source' development is NOT commercial at all, and I can't stand the idea that it might be. The idea is:
A) I want your code so I can play with it.
B) You give it to me because you EXPECT to get something out of it, not because "Code should be free", though that might be what you expect out of it.

In this case the only people who can 'Play' with it are (currently) Chip makers. Who are in direct competition with Transmeta.

Or, more to the point, Open Source assumes their are no 'Consumers' as such, but merely other users. Transmeta's product IS a product, the software infrastructure does NOT currently include this sort of thing, and so Transmeta selling 'suport' to 'users' who would all be competing chimp makers is silly. Admit it.

I support 'Open Source' software because where it works, it works beautifuly. But it should not be a religion, thats just stupid.

Re:Bob Metcalfe joins the tabloid press

[Plasmic]

I especially enjoyed this quip from the article,

" I'm thinking that Crusoe chips, which are mostly software, should be open source and basically free,"

particularly because it points out the cluelessness of Bobby.

Oh, did anyone else notice that every single one of us was likened to an ignorant pig by Bobby? He was actually asserting that we were so dumb that we couldn't tell what was "really happening" in the open source world.

Re:Cost of Duplication or Manufacturing, Bob

[smart2000]

but it seems to me that it is in a different category because it is "tied" directly to hardware,

All software is tied to hardware in some way. If transmeta did open the source to the code morphing, then you or I, and huge bankroll could design our own chips that used the technology, or we might be able to use it at some other level.

Just because it's low level doesn't make the rules different. Metcalfe just doesn't understand the rules.

Re:Upmod!

[Wah]

no it isn't.

Don't you idiots realize that what you are getting is more free content? How many of you clicked on banner ads? That's what /. does, link to other content, and you post talking about, they serve ads, /, serves ads, it's how things work for ad supported sites. They (outside sites) don't care 'cuase they get hits, and /. gets the posts. More content for all. NO big deal.

BTW, Metcalfe is a fool, or got hit/bought on the head.

Re:Bob Metcalfe joins the tabloid press

[garver]

I think you may find two distinct camps in the open source world: The socialist-like view that we should all help each other and give our source away for free. The other camp are the capitalist pigs that see good business sense in open source.

The first camp I don't understand, the second, which I subscribe to, has a lot of profit potential. In short, I see the US, definitely, and the world, maybe (I live in the states, not sure about you foreigners. :-) ) as having more and more of a service based economy. Open source fits quite nicely in that type of economy. Software is viewed simply as a means to provide the services. Service based companies do not want to develop software to sell, they just want to use it. Therefore, if they need something that isn't available, whether open source or commercial, they can start development and release it as open source. Others that are in the same boat can improve upon it, fix bugs, etc. The new features and bug fixes are available to the original company. In short, they put forth a small effort and got larger rewards.

Again, it all hinges on whether you are in the software business. If you aren't and don't want to be, open source makes a lot of sense. If you are it still might make sense. You open source the framework and add commercial features. The framework will be more solid and will also serve to expand your user base, making it more likely that you will be able to sell your add-ons.

Finally, if software is your only business, such as Microsoft, Corel, and Transmeta (yeah they make chips, but that part is easy, the software on the chip is the cool part), then open sourcing would not be a smart move.

Ulterior Motive?

[thales]

Try Clickbait. It's related to Flamebait, but instead of replies he's after hits for IDG. The formula, Trash something, wait for links on news sites that people actually read, and watch the hitmeter go crazy. Then tell the advertisers "Metcalfe's coloum got XXXXX hits last week, Want an ad for next week?"

Linus a believer, not a crusader...

[SpinyNorman]

Linus is obviously a believer in open source, but that's not the same as him being an open source crusader. Everything I've ever read about Linus points to him being 100% practical and down to earth rather than motivated by free software idealism (for that try RMS or ESR). He didn't start Linux to create a free OS for others, or with any intent to further open source (a term which didn't even exist at the time), but simply to fill his own needs. His intent was never to start a distributed development project - he was in fact surprised when people started sending him patches.

If Transmeta even asked Linus for advice on if they should "open source" any aspect of the Crusoe technology (why would they?), I'm sure that his answer would have been a pragmatic one based on Transmeta's own interests and his own as a Transmeta employee and presumably option holder. I think it's his pragmatism that is probably the reason for Linux's success as a distributed OS project.

Bob Metcalf... I need a kill file for this guy

[ajs]

Another winner from Bob....

1. Transmeta is not Linus' company. He just works there and (I hope) has a pile of stock options.
   
2. Transmetat has not yet shipped a product. Perhaps it's a little to early to be making assumptions about what their final business model is.
   
3. Bob needs a clue really badly. Can someone please suggest a vendor to him?
   

Worse than I'd thought

[ajs]

At first, I had just skimmed the article. Now I've read the whole thing, and WOW! I have to say, Slashdot has done a disservice to the community by even linking to this man. His rant is so innane, I can only imagine that he hoped that by being this inflamitory, he would get top-billing on Slashdot. Sad when Infoworld is hurting so badly for hits that they have to go rattle Slashdot.

As for content.... Where to beign. He's just wrong on so many points. As I pointed out before, he keeps trying to imply that Linus is senior management at Transmeta. He knows this is wrong, doesn't he? For another point, what precident is there for Open Source firmware? Certainly no where near as much as in the software world. I could see Transmeta treading those waters in order to capitolize on the good-will of the Linux community, but not before getting the product out the door! Bob seems to not even know there is a distinction.

He also drops the ball on what it means to be X86 compatible. That means you can run Windows, SCO, Solaris x86, Netware, Linux, *BSD and any number of other OSes. Transmeta is not making a WinCPU(tm), they're making an Intel x86 compatible chip. This is neatly glossed over in the article.

Hemos, next time do yourself a favor and just ignore anything Bob Metcalf has to say. This guy is just a troll. I've now seen the ultimate argument for article moderation....

Re:Worse than I'd thought

[ajs]

Linus is not senior management, but [...]

There really is no but. He's touting Linux and Open Source on one side, and working for and touting Transmeta on the other. The two worlds really only touch in that Transmeta wants him to make Linux work on their chip. He's not in charge of Transmeta, and you cannot call the company hypocritical for saying one thing and doing another when Linus' comments are not the company's comments. They are completely consistant in their approach. They simply hired Linus because he was good (remember he wasn't a superstar then) and they wanted to control the OS.

Firmware is software

Firmware is a type of software. A very specific type. And, in this case, I think that the Open Source dynamic has not been proven. I'm not saying it's a bad idea, just that it has not been proven. Thus, I can understand these guys being careful. They want to get the chip out the door.

Even still. Let's say they get the chip out the door and never release the source to the firmware. Is that a bad thing? Perhaps the Open Source community will think so, but why is it worse than Phonix not releasing BIOS source? Because Linus is an employee? I think not. Asking them to release source is one thing, but expecting it is quite another....

... the Crusoe is optimized for Windows ...

Yes, it is.

My point was not that Crusoe was not optimized for Windows, but that it was not a Windows-specific chip. You could run Linux on it, or Solaris, or whatever just like you can on the Athalon.

Basically Metcalf's article was trying to imply that you have this Linux company and they have to stoop to Windows. I was trying to point out that Transmeta is not a Linux company, though they do employ Linus. This is where Metcalf doesn't get Open Source. He's still thinking in terms were employing the leader of a project gives you control of that project, and makes it an integral part of your business.

In reality Linus could leave Transmeta tomorrow, go work for their competition, and they could go right on using his OS. *This* is the power of Open Source. Metcalf has not yet seen this.

And Metcalf a troll? Uh huh, he didn't exactly post it here on /. did he?

You're right, he didn't. On the other hand, he knows damn well (now) that any inflamitory article that he publishes will end up being a Slashdot headline and generate tons of hits for his site.

I hypothisize that he wrote this article specifically to troll the Slashdot effect. Thus, he is a troll. If I'm wrong, I appologize, but the fact that he mentions Slashdot by name, certainly implies that he has it in mind....

Unless everyone who dares say anything bad about Linus or Linux (even off /.) is now considered a troll.

Oh, no not at all! Watch, I'll prove it:

Linux has really poor structure to it's on-line documentation, and finding information about any given file or program can be like pulling teeth. Witness the difficulty in trying to figure out if documentation is in man, info, ps, text, sgml, html, tex or some other format. And where? /usr/doc? /usr/man? /usr/info? On the web? Quick: where's the documentation for the "ip" command, and how do I read it?

I do not consider that a troll. It is, however a negative comment about Linux (the distributions of Linux, not the kernel specifically). I even through in a hook at the end to get people to respond, but that's now what trolling is. Trolling is trying to make people mad enough to shout back at you. It's pushing a point of view, not because it's right or wrong, but because it will get you paid attention to. In the comming war for page views, you will see more and more Trolling of this sort. Trolling not for replies or moderation (some people get off on getting negative moderation on Slashdot, it makes them feel important), but for hits.

Re:Worse than I'd thought

[TummyX]

Linus is not senior management, but it's the 'big' open source boy. And if transmeta is doing something that he likens to prostitution (software is like sex it's better when it's free) maybe he should consider working with alan at redhat. Either that or changing his views.

Firmware is software. Would be if i said, who the hell needs source to the kernel! noone needs that.

And the crusoe is optimized for windows (at least the more expensive one is).

And Metcalf a troll? Uh huh, he didn't exactly post it here on /. did he? It's his column and he can say and do whatever he wants. Unless everyone who dares say anything bad about Linus or Linux (even off /.) is now considered a troll.

Re:MILO

[ajs]

As I pointed out. There are very few examples of open source firmware development. You have pointed out one of the few. There will be more, I'm sure. This is a good thing. Transmeta may be on the bandwagon with that effort at some point (and PR-wise, I think a later announcement on that front would make more sense than an earlier one). Either way, they've done much for Linus, and by proxy (and maybe even directly) for Linux. I for one am anxious to get my hands on one of those bad boys....

Do you really think that Bob wanted Transmeta to open the source to their firmware, though? No, he's just trying to get a bunch of geeks on Slashdot in a lather about this so that he can sell hits. Guess it worked.

Re:Leaving aside the ad hominem attacks...

[ajs]

Please check out the definition of ad hominem fallacies. I think you will find that constructing a set of arguments with evidence based in fact about the matter at hand and then using those conclusions to draw inference as to the motivations of the debater in question and the value in further debate with that individual does not constitute a logical fallacy of the type you suggest.

On the other hand, the body of your post made a point, and if you re-read my posting you will find that you and I are in agreement. I was simply saying that before shipping a hardware product is not the time to go testing new business models in firmware distribution. Let them get the darn chip out the door and then chat them up about releasing the source. Of course it will be harmless for them to do so (they have patents on the hardware to take advantage of that source) and Linus will probably be quite happy that way, as he can start blurring the line between Linux and Crusoe by having the kernel modify the instruction set to suit its needs.... Could be fun. Obviously such an effort would require that the Linux kernel developers have a clear idea of how the Crusoe firmware works, and given the distributed nature of Linux development, releasing the source would be the best way to bootstrap this.

You see, I actually did think about what I was posting.

OB off-topic thought: can someone write an HCF instruction for this sucker? ;-)

Back on /. again???

[thimo]

After the last time I saw Metcalfe on /. I thought and hoped he would never again be able to make it back onto "our" frontpage, but he did it again. He really is just trying to get the Slashdot Effect (TM) on his page and I really, really doubt he believes what he preaches.

And off course I'm just pissed off he got me again, he got my hit, I read the article and wasted another 5 minutes on him. Hemos, please don't post these kind of articles. Please!!! :)

Thimo
--

Metcalfe has a beef, and a point

[Hobbex]

I think it is more or less clear by now that Metcalf has a beef with the whole open source concept and ideals. It seems strange for someone who is, or at least was, a brilliant inventor, but when you consider his personal history it does make some sort of sense. Metcalfe made a brilliant invention, started a company around it, was forced out of the company by the resident corporate baffoons, and has since been marginalized to a nothing collumist in some god forsaken magazine where he can only get attention by trolling away.

Given that, one can understand why he carries a grudge against the open source ideals of giving away you inventions rather than trying to capitalize on them, and people like Linus in particular. Linus (and the likes of Tim Bereners Lee) is a living reminder that he could have chosen another route for his invention, one where he would not have been forced out by a bunch of idiot beaurocrats, and could still be a respected senior hacker rather than a senile Anonymous Coward wannabee.

That said, I do think he has a point about this. He seems to forget that Torvalds works at Transmeta but doesn't own it (just like the rest of the press, really), and his statement that Transmeta shouldn't hype Windows performance is ludicrous. They have to sell the damn chip, and they obviously prefer Linux to Windows or they would have worked with MS to get CE on Crusoe portals (oops, I mean make them Windows Powered). But, I would still like to hear Linus' opinion on the fact that Transmeta, cool product or not, obviously don't endorse openness at all. They are keeping the code to their chip completely closed, and they have even patented the software.

Personally, I would think long and hard before working for a company like that if I could pick and choose like Linus can. The fact that it doesn't seem to bother Linus at all _is_ a thorn in the side of the honesty and clarity of his motives, as much as I hate to say it. I would never raise myself to the level of judging Linus, there are few people alive today for whom I have as much respect, but I have to admit that I can't help but wonder what makes the two legged house dwellers at Transmeta different from those and Sun and Microsoft.

-
We cannot reason ourselves out of our basic irrationality. All we can do is learn the art of being irrational in a reasonable way.

Re:If Open Source == Good, then all else == Evil??

[Sun Tzu]

Is that your logic? ;)

I think Linux is great and I run it on all of my personal machines. As the operating system on my machines, it is the single most important piece of software on my systems -- and the one that benefits the most from being open source software.

I have no qualms about non-open source software. However, I have no patience with companies that can and do abuse their proprietary position by taking advantage of their customers. Meanwhile, I have no objection to programmers making their living by selling their wares. Do you?

Re:Physical world vs. Digital

[w3woody]

Simple: if hardware is not proprietary, someone with an existing chip factory will make it cheaper than you can and sell their value-added package (motherboard) for less and put you out of business; hardware and true physical-domain engineering (EE, MechE, CivE, EnvE) and invention need to be proprietary so the inventors can be compensated for their work.

Um, actually the same thing can be said about software. The cost of most proprietary software packages is *not* the cost to reproduce, but to pay for the software engineering and software invention--that is, so that the software programmers can be compensated for their work.

In fact, your argument (about paying for the R&D rather than giving the R&D away for free so that a fab company without R&D overhead can reproduce it and sell it for less than you) is better suited for software, where the R&D overhead is substantially larger than the cost of reproduction. Frankly what you have described (about companies taking R&D and adding value and undercutting the market) exactly describes RedHat's business model.

Further, there are a number of companies who are experimenting with the open source model for hardware. http://cera2.com/micr/opensrc.htm The reason why it hasn't taken off quite as quickly is because there is more overhead in reproduction. However, if you read through a copy of the EE Times, Open Source Intellectual Property is becomming a fairly big deal in the chip design and fabrication arenas. And Open Source Intellectual Property has always been in the hardware arena, at least in the form of "reference hardware designs" for using various chip sets.

I'm not saying Open Source is a bad thing--to the contrary, it benefits the both the hardware and software community to learn from each other and to have more eyes looking at the IP to verify it's correctness. But Open Source is not just for software.

To those who say it can't be done, it's being done

[w3woody]

It's being done by Sun with the picoJava and SPARC cores.

eg3.com has a list of Open Source hardware links.

Tom Coonan has donated a free 8-bit microprocessor core to the Open Source IP community. One interesting aspect of this is that you can "build it yourself" using an FGPA booted from an EEPROM you can burn yourself.

And speaking of FGPAs, Xilinx has a whole page of IP for downloading and burning into their FGPAs here. What makes this super-spiffy is that you can write logic for these things and program them yourself--they download the gate configuration logic from an external ROM (or EPROM) or other source. In fact, many people are using these things by downloading the gate configuration from other sources, such as a data file.

The Open Source Hardware community apparently is thinking along the lines of using FGPAs to experiement with creating a various open source microprocessor cores in order to get the bugs out. Once the bugs are out, you can then create an ASIC core from the same data files and burn chips for production. What makes this strategy interesting is that probably for around $500 (or less) in hardware, you can build your own test bed. In fact, I could see building an FGPA "loader" which is basically a 6502 and a UART chip connected to your serial port which contains all the logic necessary to boot and download logic into an FGPA from your desktop as a sort of "in-circuit" emulator.

But my hardware days are behind me, at least for now...

Re:intellectual property

[w3woody]

As others pointed out, Hardware open source is a completely different beast than that of software. I can't just build the crusoe in my spare time as a recreation. I have to devote serious resources.

Same has been said about operating systems. So don't expect an open source operating system any time soon...

Re:intellectual property

[w3woody]

Xilinx makes a field programmable gate array which allows you to wire the thing on the fly. That is, they'll sell you a chip which contains anywhere from 100,000 gates to 1,000,000 gates which can be dynamically wired to provide all sorts of functionality, from microprocessor cores to UARTs to RAM cells. You don't need millions in fab equipment. Just one of these chips and an EEPROM programmer and some freeware software (links here) will do the trick.

It's not the same as editing masks using a VLSI design tool, but it does the trick.

Further, most people who design chips don't have or need millions in fab equipment. When I was at Caltech about a dozen years ago, I took a class on VLSI design where we simulated the results, and for the final, sent our design to a fab house which specializes in one-off fabrication for testing. One-off fabrication costs a few hundred to a few thousand per chip, but gives you a way to test your designs in hardware once your prototype checks on the simulator software.

Beyond that, you don't really even need to do this if you simply want to translate your FPGA design into an ASIC core for mass production. There are several fab houses who will take your FPGA data and turn it into an ASIC core for you by automatically laying out the chip-level logic from the FPGA data.

So no, you don't need "tens of millions of dollars worth of fab equipment." Far from it; just a couple of FPGA samples from Xilinx, and some software, and some descrete components for building prototype circuitry that uses your FPGA circuit from a company such as Electronix Express will do the trick.

And hell, just poke around the Free IP site; they've got two processor cores available for download, including one of which simulates the 6502 very well on several FPGA vendor's products.

Limits of free software

[divec]

> Will open source reduce the corporate demand for computer programmers?

With a growing pool of free software tools and libraries to draw on, programmers can become more and more productive. It becomes practical to tailor existing software to the needs of a particular corporation, thus creating a whole range of jobs in the bespoke software business. Nevertheless, this is a hypocritical question, coming from the industry which has charged ahead into destroying more traditional jobs than any other industry (though at the same time creating many more service jobs).

> What limits are there on the kind of products that can be created? Obviously games are out of the picture since 1 company would develop the engine and the others would leech off their work.

The Quake engine is already GPL'd. If a few free engines get very good, it may become cheaper to tailor these engines to your game's needs than to build a new engine from scratch. It makes more sense for game companies to concentrate the "content" rather than the engine; this is what happens in the non-electronic world. Monopoly is a popular game because of how it works, not because of the wood pulp it is printed on.

Privacy

[divec]

I can't comment on the validity of your claims, so I shall presume they're accurate. But FSF accounts have passwords, and the current license terms for Emacs don't prevent anyone from using it who agrees to the license terms, including MIT, Microsoft and Saddam Hussein. The GNU project has developed the GNU Privacy Guard, which assists untappable communication. So whatever the background was to the situation you describe, RMS and the FSF clearly uphold the right privacy today.

Your comment about books and music is just plain wrong. RMS has said he finds it acceptable for authors/composers to prohibit large-scale redistribution of their novels/music, provided people are free to give copies to a few friends. His view on software manuals is that they should be free, including the freedom to commercially redistribute.

Re:Privacy

[Brett Glass]

The founding of the FSF was bankrolled by John Gilmore, who was (and is!) seriously in favor of strong crypto. This may have something to do with the FSF's tolerance of passwords, etc.

--Brett Glass

Free Software advocate != left winger

[divec]

I believe the government should not grant everlasting monopolies to software companies. I don't believe that the state should have a monopoly on the creation and distribution of wealth, or that the state should have a monopoly on rights at the expense of the individual, or that income should be equally distributed.

You shouldn't relate the two concepts just because they both seem alien to you; they are almost entirely unconnected.

APSL

[divec]

One of RMS's main objections to the APSL was that it didn't let you keep your modifications private.

Which link shows that the FSF opposes the right to privacy?

Re:Please....

[divec]

I certainly wouldn't store stuff on my work PC which was meant to be secret. (Confidential files, e.g. bank details, on the central server are a different matter)

In 1983?

[divec]

Brett's anecdote must date back to at least 1983. Wage data probably wasn't on the MIT computer in those days. Do you have evidence that RMS or the FSF believe, or have ever believed, that people's bank info should be publically available?

Re:Licenses and firstborn children

[divec]

Cool - we agree on something. Let's hope UCITA fails. (And this damn encryption thing we have looming in the UK...)

Lack of warranty

[divec]

I think that's fine for the download or Cheapbytes version. However I think a company should be broadly responsible for what they sell. If there are errors in Red Hat which cause major problems to a customer, I think the customer has a right to a refund, at least.

It's different for a $2 CD. Neither the vendor nor the authors should be responsible for any errors in this.

Re:What if....

[divec]

Legally, the only person who can be responsible for a fault in something you bought is the vendor. You don't have a contract with anybody else. This is a good thing. If there is a kernel problem, Red Hat should have warned their customers thaht this is a possibility. Otherwise they should accept the consequences for misleading you into thinking Red Hat is reliable.

This "flaw" is the same flaw if I buy some poisoned Coke from the newsagents. Was it the newsagent's fault? Or Coke's?

Legally, I can only sue the newsagent. But then the newsagent can sue Coke, so it kind of works out.

Stallman believes in the right to privacy of code

[divec]

RMS believes people should be able to keep their code private. (And also other knowledge)

He, and the FSF, just believe that you shouldn't be able to prevent others from sharing the code that you give them, or to not provide someone with source code if you expect them to be able to run your binaries.

RMS must be the most misquoted hacker of all time. His, and the FSF's, position are made quite clear in about 3 pages of text here, yet people seem to continually misunderstand.

Re:Ugh, this is even more Ill-Informed

[Hard_Code]

All Transmeta really needs to do is give away some SDK that allows people to "plug in" their own ISA, or "flash" the rom or something. People /don't/ need to know the native destination ISA, they just need a mechanism for exploiting the chip's abilities by supplying their own ISAs. I think this would be great for Transmeta, because without breaking compatibility, they would gain all the ISA's that anybody wants to provide. The only issue is a change to the code morphing software spec/implementation /itself/. If this changes, then the SDK has to be re-released, and ISA's will have to be regenerated for the new software, etc.

I think they just want the ball in their park for a while, which is understandable.

Jazilla.org - the Java Mozilla

Re:The FSF and "stewardship"

[Hard_Code]

And can't I, as an author who has licensed previous code under the GPL, relicense a new version of that code (only my code!) under any other license, say, some proprietary license?

Technically I should be able to license a file under the GPL, add some whitespace, then license the new version for $10K a pop.

Right? The power is in the copyright holder's, i.e. my, hands.

Jazilla.org - the Java Mozilla

Re:Bob Metcalfe joins the tabloid press

[mochaone]

Metcalfe seems to think people take him seriously. He is very wrong. Metcalfe is the same buffoon who thinks Linux is flawed because it is "based on 30 year old technology". He of course was referring to Unix. If we were to apply Metcalfe's line of reasoning to other areas, we should not be using the telephone, automobile, radio, tv, etc. because they are all based on ancient technologies as well.

Metcalfe seems to have an ulterior motive. I don't know what it is yet but I don't think even he believes the crap that he so frequently spews. At any rate, since Metcalfe is 30 years removed from doing anything significant, I propose that he is flawed and as such should be put out to pasture.

Um...

[twit]

That's the point of a SEC form 10-Q - to disclose all the possible downsides of investing in this company. That it sounds bad is unsurprising, because it's supposed to sound bad; it's supposed to sound very, very bad indeed.

But is it proof of anything? Nothing at all: all it says is that you can lose money investing in the equity market, which is hardly news. At least not since the early nineteenth century.

--

Something wrong with Linux?

[quonsar]

The column goes on to say that, since the processor will run Windows code, there must be some thing wrong with Linux.

Well, the article may be total bullshit, but I read it 4 times and I'm still looking for the statement quoted above. It isn't there. The only mention of x86 code is, and I quote, "Worse, Crusoe is touted for running Intel X86 software, and in particular, Microsoft Windows. Doesn't the open-source community say Windows is beneath contempt?".

Is someone at Slashdot gunning for an editorial position over at ZDNN, or what?

======
"Rex unto my cleeb, and thou shalt have everlasting blort." - Zorp 3:16

Honesty and motives

[fpepin]

In most of the interviews that Linus gives, he says that he's an engineer first of all. That if you really want an open source advocate, pick Stallman instead, he's more of an ideologist. Look at his motive and you'll see that he's very honest about it. He started Linux because it is a cool and interesting project. It still is, the same way he finds the job he does at Transmeta to be very interesting.

The way I understand it, he sees the open source movement as being a way to reach a mean, but not necessarily the mean by itself. It's an interesting idea, but that doesn't mean that other ideas aren't interesting.

If it means that Linus has some little selfishness in him by working for a close-sourced project that he likes, and doesn't focus his whole life on open source projects, does that make him bad. No, he has to eat like everyone else and he probably wants to try different things now and then. He never said he was like that, so please don't say he's not being honest and start questioning his motives because he works for Transmeta.

The GPL doesn't undermine programmers livelihood.

[Convergence]

Do you think you have a right to take the fruits of my labor, without restriction, even if it is unpaid and volunteer, then resell it for money to further your livelihood?

Just because I didn't earn money doesn't mean that that labor doesn't have value, the GPL/LGPL codebase (linux, emacs, gcc, glibc, bash, ....) has an estimated value in the multi-billion dollar range. Why should we, the group who created that value, share that wealth with you, without restriction, unless you're willing share any additions to that wealth BACK so the rest of us can enjoy the new pool? Regardless of your answer, its OUR choice how it is to be shared, and tough luck, you have to respect that.

Yes, we require that you share back any additions to that pool of wealth, but is that so high of a price to pay for the billions of dollars of code sitting in it? There are even exceptions to that requirement; you don't have to share back your additions if they're for your own use. You can even use any of the software in it completely freely.

Few individuals get free(monetary) access to billions of dollars of code. Fewer get access and the ability to alter and redistribute it how they wish. Why do you not thank your lucky stars that you and we are so fortunate and lucky to?

Money

[Convergence]

Your world may revolve around money and jealousy over not getting any. I like to think that other things are more important... Like a system that's stable, flexible, and powerful and does what I want. (Which is a value without price, I might add.)

Re:Money

[Brett Glass]

Then why begrudge others the chance to make money? Especially since anything they make, if they start with your publicly available code, will be the result of their improvements. (After all, they can get the original for free, so if people pay money for the improved version the programmer must have added plenty of value.)

--Brett

Re:Bob Metcalfe joins the tabloid press

[barleyguy]

Actually, Linus has mentioned his views on fragmentation in his last two public interviews. One of them is here: http://slashdot.org/article.pl?sid=00/02/09/042824 0
So, please don't accuse someone of making up something which is actually a current event.

I do agree with you on Metcalfe, though. Every article I've read of his in the last year or so has been equally as clueless.

As far as non-proprietary software, there are more and more companies doing it. I think what many companies are learning, is that programs that are mainly written for internal benefit - i.e. performing a task inside their company - can be released as open source, with more good effects thatn bad ones. I am working on a project right now that is mainly for my employer, but is also open source. Unless a company's main income was built on proprietary software, it makes sense to open source internal projects. However, sometimes the "value added" part of a product depends on the software, such is the case with Transmeta, and maintaining a closed product can have a sink-or-swim effect on the company. I don't blame them, in this case, for keeping the lid on their software, at least in the short term.

Cost of Duplication or Manufacturing, Bob

[fwr]

I think what he doesn't get is one of RMS' concepts which I believe is that software should be free because of the zero cost of duplication or manufacturing. Hardware is in a different "realm" when it comes to manufacturing. Chip companies have to pay billions of dollars to make a chip plant capable of producing the latest chips. Even though Transmeta may not be making the chips themselves, and PAY IBM to make them in their chip plants, they still have to PAY to have them made.

I can't speak for RMS on whether the Transmeta morphing code should be free or not, but it seems to me that it is in a different category because it is "tied" directly to hardware, which should not be free (due to the cost of manufacturing). AND the "tying" is not on purpose to make a profit, it's because it's an intergal part of the way the hardware works.

I personally believe that companies should have "rights" to sole-source HARDWARE for a limited amount of time in order to be able to make a profit on the investment in time needed to develop the hardware, built the processing plants, etc. Kind of like a patent, but severely reduced at least for the computer industry. I'd guess the MAXIMUM timeframe for retaining exclusive rights would be 5 years. This would be enough time for companies to recover their development costs and make a healthy profit, but force them to innovate in order to stay in business. Could you imagine if Intel no longer had the rights to the Pentium core, and possibly the Pentium II core (I don't know when the Pentium II was released)?

I also personally think that we can't, or shouldn't if we could, FORCE people to release their software as free. If Microsoft or some other company wants to keep their source closed, then let them. I believe that in the long run free software will show better quality and capabilities than closed software, and the market will reveal this soon enough.

Re:Cost of Duplication or Manufacturing, Bob

[alleria]

An interesting distinction between software and hardware, of course, but isn't the line blurred in the case of the Crusoe? Like many articles have pointed out (esp. the one on ArsTechnica), the Crusoe implements in software much of what other chips implement in _hardware_. Should that mean that the software which is replacing the hardware should be open? Or that since it indeed replaces the equivalent of hardware in other chips, that it should be closed? IMO, this type of software is really hardware. But on the other hand, Linux programmers seem to have no problems distributing their software modem drivers under the open source model, so it _is_ hard to justify why code morphing software, which complements hardware also, should be exempt from open source ...

If it were true, so what?

[Duxup]

*duck flames*

Now granted it would seem Mr. Metcaclf has of "some" of his facts wrong here. However I wonder if he was right fundamentally if it would be that big of a deal? It would be nice if everyone who contributed to the open source movement would release software in that manner, but if you support OSS must ALL of your software be so?
First of all Linus works for this company and doesn't run it (that I'm aware of), I have several friends who work on OSS software in their spare time but work for a commercial company who (like most) don't release their code. So should they quit, find another job, perhaps have to move their families and such? I'd say no.
Secondly lets say that he did own the company. Would this be the end of the world? Mr. T releases some code that's not OSS. So what? It's sad but often it seems that some more vocal advocates of the OSS movement has become somewhat more "harsh and angry" and implies that all software must be OSS, ignoring the fact that people still are free in many of the countries of the world. I wonder if that helps or hurts as the movement moves into the general public's eyes in the future. Now granted many of the OSS movement are perfectly nice people, but some seem to have taken on an angry feel and that worries me.
However if this were true, and Mr. T had control if the software is OSS or not, and he didn't release it as such, is that so bad? I'd still think he was an alright guy, I wouldn't like it, but it's his choice. If he wanted to make some $ working for a company that interests him and that doesn't release it's software, no biggie to me.

Amen Brother Man

[Duxup]

LOL
Amen.

I like that they put the comments from the person who submitted the article up there. However they should at least ask the person who submitted it if they can edit TOTALLY inaccurate information that's in it. I've seen this happen several times before, and sadly somewhat more frequently as of late, I don't mind if it could possibly be a judgment call or posed as the submitters opinion (even if it is way off). However outright stating that something is in the article that is not in the article just is disappointing coming from /.

The GPL only shares with those who share alike

[JordanH]

• When you add one line of other people's GPLed code to you work, you must give away all of your own code in that product. Not share proportionately or fairly with the author who wrote that one other line, but give it all away, to the entire world, forever.

Yet you advocate the MIT X and BSD licenses as being "true" sharing licenses. These licenses allow you to take and take and take from the work of others and "not share proportionately or fairly" or at all with the original authors.

So, it could be said that these other "Open" licenses discriminate against programmers who wish their works to be extended only by those who are willing to share those extensions back to the original author.

Yes, you can say that the GPL discriminates against programmers who wish to profit from the works of others and give nothing back. It does so in favor of users who need real software reuse and the accretive benefits of a large body of code.

Every license is targeted to a certain advantage, a certain group. That's the reason for licenses. Even the absence of a license (Public Domain) is advantageous to some while detrimental to others.

I think it's about time that we, as an industry, put the emphasis on the poor user who just needs to get work done. God knows they've been ignored long enough.

This boogeyman about how the GPL is going to put programmers out of work is just fantasy. Somehow, with the explosive growth in the use of GPLd software in the last few years, the acute deficit in programming personnel grows worse every day. Most programmers do maintenance work or custom contract work. These programmers will be helped by a large body of code from which to draw.

Other programmers who wish to sell only proprietary works can continue to do so. These programmers may find that they are at a competitive disadvantage against the body of GPLd work, but maybe not. If they can't compete, then it clearly shows that the user has benefitted. The user will be able to get products at a low cost that can always be self-maintained, or contract with programmers to maintain them for them, that is of high enough quality that commercially licensed software can't compete. Finally, it is possible for the user to benefit from true software reuse. I guess there's no cloud that doesn't have a silver lining for someone, eh?

-Jordan Henderson

Re:Bob Metcalfe *IS* tabloid press

[G27 Radio]

So, in 1996, CD-ROMs through Federal Express will emerge as the information superhighway. Instead of an Internet brimming with Web pages under construction, too few of us will haunt ghost pages.

I used to think the crap that Metcalfe has said in past stemmed from him being a grumpy old man that wanted to rain on the parade. Eventually I realized that he likes being contrary to get attention. This is what we call a troll.

One explanation about why the VLIW set isn't released is because it changes from chip revision to chip revision. The software layer is absolutely necessary to maintain compatability. On the other hand, the details of the code-morphing software are a trade secret that I'm sure competitors would love to get their hands on. This combined with the fact that developers don't need it gives Transmeta very little incentive to realease the source code. How long before Intel or AMD or someone else figures out how to do the same thing on their own? I'm sure Transmeta needs to keep their lead here as long as they can if they want to stay competitive in the processor market.

I also remember a couple years ago when he said that the Internet would suffer massive breakdowns and outages which would render it unusable. He then said he would eat his words if it didn't happen within the year. So in the end he held a press conference where he put his article in a blender then ate it or drank it or something. More publicity for him.

This is business as usual these days for Metcalfe. It's a shame to see him reduced to trolling after all he's accomplished.

numb

Re:Bob Metcalfe joins the tabloid press

[jedrek]

Lemme pull up a quote I found from the esteemed Bob Metaclfe:

So, in 1996, CD-ROMs through Federal Express will emerge as the information superhighway. Instead of an Internet brimming with Web pages under construction, too few of us will haunt ghost pages.

Does he have a clue about technology? I guess co-inventing Ethernet is pretty important in our current computing world. Does he have a clue about trends and how people and technology interact? Your call.

Jay

-- polish ccs mirror

Re:Bob Metcalfe joins the tabloid press

[TummyX]

Now lets be facing it, Bob Metcalfe is not a stupid man. He knows as well as you and I that Linus does not own or control Transmeta, the very suggestion that a commercial company should give away it's prized possession for free goes against every principle of the free market. It's also completely silly to suggest that should be done just because one Linus Torvalds is currently employed there.

Point is that Linus has hardly been a quiet pundant. How did he put it? "Software is like sex, it's better when it's free". Ok, I want "free" code morphong please. Point is, Linus shouldn't be bashing MS (he does it all the time in his speeches) to gain beenie points. It certainly makes him look 'arrogant' to me.
And Metcalkfe is totally right about Linus' changing views. Now commercialsm is GREAT, and fragmentation can be GOOD...what's next? This man is just trying to start an argument. Linus has done a vast amount of good for the open source community, should we really be attacking him for being involved in a closed source project for his current employer?

Not for contributing heaps to open source projects. Attacking him because he's been such a propronent for open source with a "holier than thou" attidude to non open source projects/companies. This project is hardly a small project for transmeta, it's huge, and it's hugely software - so Linux is now working for a proprietry 'software/hardware' company. What a pity it's only stylish to bash the other software company.

Open-source code morphing

[Tim Behrendsen]

If I could extract one of the (few) valid points in the guy's article...

I'm not going to cast blame on Torvalds, since I don't think he necessarily controls anything, but if one believes in Open Source, shouldn't the Code Morphing code be released?

It would be interesting to know where RMS comes down on this, and firmware in general, such as the downloadable Microcode in Intel chips.

--

I prefer to give the benefit of doubt

[JamesSharman]

Well, for those who were never forced to study "Animal Farm" at school, it is a parody of the Russian revolution set on a farm where the pigs take the role of the communist leaders.

As you correctly pointed out, this article could have been read, as Bob implying that open source was akin to Communism. From the tone of the article however I prefer to read it as him likening the open source movement to a revolution that is turning full circle. Maybe I'm just a nice guy but I prefer to think of Bob as a self justified fool, not as a vindictive bigot that would slap the communist label on something he doesn't like in order to start a witch-hunt.

Bob Metcalfe joins the tabloid press

[JamesSharman]

I really have to admit I was expecting an article along these lines to come out sooner or later. For those who haven't read the article it likens Linus going along with a closed source windows targeting system to the pigs in the Orwell's 'Animal Farm' slowly becoming what the revolution supposedly toppled.

Linus (like most of us) is a true supporter of open source and everything it stands for, he (also like most of us) is a realist who realizes the world isn't what we would like it to be. Linus makes very little out of linux and contrary to what the article states does not really 'control' it. The article makes stupid statements like the following:

"So just to keep Torvalds honest, I'm thinking that Crusoe chips, which are mostly software, should be open source and basically free."

Now lets be facing it, Bob Metcalfe is not a stupid man. He knows as well as you and I that Linus does not own or control Transmeta, the very suggestion that a commercial company should give away it's prized possession for free goes against every principle of the free market. It's also completely silly to suggest that should be done just because one Linus Torvalds is currently employed there.

If I had any choice in the matter, every thing I write would be free and open source, my mortgage would magically pay itself and the world would be free of hunger and poverty. However since reality requires me to get a job I chose to get a job that interested me. For the last 5 years I've been working as a 3D programmer in the games industry, now would Bob suggest that SCI (my previous employer) should open source Carmageddon2 just because an employee (me) who is an open source advocate played a small role in it's development. I expect Linus took a job with Transmeta for the same reasons, because it looked like a fun and interesting thing to do. Transmeta employed linus because he is a damned good programmer, of course the fact he is the man behind linux was never going to hurt.

The article goes to the depths of saying

"But with Torvalds saying some animals are more equal than others, why is the sanctimonious open-source press still cheering him on? Are the likes of Slashdot.org, just gobbled by VA Linux, also porking out in Orwell's farmhouse?"

This man is just trying to start an argument. Linus has done a vast amount of good for the open source community, should we really be attacking him for being involved in a closed source project for his current employer?

All in all this is Bob Metcalfe doing the 'tabloid thang' and trying to attack the open source community in the lowest possible way

No, Bruce; YOU have it backwards.

[Brett Glass]

Oops, here comes another representative of the bitter and hurtful camp!

Maybe there's a "camp" (as you call it) because there's a consensus that it's so.

You got it backwards, Brett. Can you guess why I am working on the problem of modifying a license (not necessarily the GPL) for the situation of ASPs? Because some ASPs asked me to. Why did they ask me? To protect their work and their competetiveness. They want to put out Open Source software and benefit from the collaboration of other ASPs and the community. They want to create a commons in which no one party to has an unfair advantage over the other in this collaboration.

Translation: They really don't want to be competitive. Rather, they want to force their competitors to reveal the technology which might keep them ahead of the game. This is anti-competitive, not competitive.

But then, the GPL is, too. The GPL is intended to undermine programmers and hurt their livelihoods by giving away equivalents of their products for free and denying them access to that code so they can't add value.

So, they will put their own code behind whatever license I come up with.

Hmmm. Since when did you become "King Bruce?" It seems that, by attempting to dictate terms and restrict access to code, you are in fact becoming the very thing that open source was originally designed to avoid. The parallels to "Animal Farm" are obvious.

The GPL treats commercial software developers the same way as everyone else. In fact, only paragraph 3(c) of the GPL even has a reference to commercial vs. non-commercial distribution, and not in a way that would impair commercial distribution.

No, the GPL specifically targets commercial software developers by attempting to force them to give away the fruits of their labor. Distributors might be able to make a living (if they're lucky!) by pressing disks. But the programmers themselves -- whom Richard Stallman wished to reduce to the level of starving graduate students -- get nothing.

You can't understand this sharing thing, can you?

I understand it better than you do, Bruce. The GPL is not "sharing;" it's a variation on the old-fashioned, cruel playground game of "keep-away." True sharing means sharing with everyone.

Want to use my GPL code? Fine! Share it with this excellent set of sharing rules we've cooked up called the GPL. Return value equal to the value I put in, and treat me as I treat you.

No, this is not what the GPL says -- nor what it does. It says, "Some people can share the code for free, while others have to pay a price that's so steep that it will destroy their livelihoods." Again, "Some users are more equal than others."

But you're saying no share! gimmie!

No, that's what you're saying -- to one specific group. Programmers. This discrimination violates the very "Open Source Definition" that you wrote, since it discriminates against a field of endeavor.

There are indeed licenses that truly share. They're called the MIT X and BSD licenses. They don't discriminate against a field of endeavor. The GPL does.

Go over to freshmeat.net and look at all the GPLs on new code. Every one is offering you partnership in that code, if you can just learn to share.

What this shows, Bruce, is that the GPL has successfully pulled the wool over the eyes of many people. If you are ethical, you will fix this. If you merely want to ride the GPL to wealth, and are as greedy as those you revile, you will show this by continuing on your current course.

And it happens that yes, I am on some corporate boards. I'd rather have me there than someone who doesn't believe in Open Source,

From the above, Bruce, it appears that you yourself do not believe in open source. You believe in source that's open to some people, but not to others. And you want to get to do the choosing, according to who strokes you, pays you, or otherwise happens to get on your good side. You're a sellout, Bruce, and this shows it.

--Brett Glass

Exploitation? Yes, of the programmer.

[Brett Glass]

The GPL doesn't just require the programmer to "return" value to another programmer. It requires him or her to give away his or her work to the entire world for free. It's a Faustian bargain which asks too much and therefore should not be accepted.

--Brett Glass

Thank you for your offer to share your pretzel ;-)

[Brett Glass]

Maybe there's a "camp" (as you call it) because there's a consensus that it's so.

You, Tom, and Bob Metcalfe.

And many more already. There will be more still, as people begin to recognize the true origins and motivations of the GPL.

The GPL was born of spite and malice. It was written by a man who harbored such a grudge -- such bitter anger -- against commercial software developers that he succumbed to carpal tunnel syndrome in an obsessive effort to sabotage them. This is all well-documented.

Wrong. Their competitors are not forced to do anything unless those competitors choose to be partners on the same piece of code. Don't use the code, you don't have to obey the license. Together, the collaborators on an open source project are a bigger and better competitor.

In other words, what you are encouraging is collusion among some vendors to destroy others. This is anti-competitive behavior, Bruce, and is highly illegal as well as unethical.

Actually, that's not true. You can add value to your own GPL code and put it out with a commercial license.

This doesn't work, Bruce, and I happen to have a tape of you saying so yourself at the Fall 1999 LinuxWorld.

The promise of being able to "dual license" is, alas, a trap which the GPL sets for unwary developers. It seduces developers into stamping the GPL onto their code.... But then, once they do so, they discover that it is virtually impossible to license their work for money. Here's why.

First, when the GPLed code is released, the author will doubtless receive suggested changes and improvements -- often in the form of code. The problem is that, since the initial code was licensed under the GPL, the code which is contributed back is licensed exclusively under the GPL because that's the license under which the contributor got it.

Thus, if the original author accepts a single contribution, his whole work is irrevocably licensed ONLY under the GPL and his ability to legally dual-license goes away.

What if the author refuses to accept the changes to avoid this? In this case, a second mechanism kicks in. The contributor -- or anyone else -- can fork the project to create a GPL-only work that competes directly with the author's and drives him out of business.

Finally, dual licensing does not work because only a very un-savvy businessman would license code for money when there's a GPLed version available. This is true for two reasons. First, the existence of the GPLed version effectively reduces the market value of its functionality to zero; anyone can get that functionality for free! Thus, if one pays money to license GPLed code, one is paying for something which has no market value to end users. This puts the commercial developer "in the hole" from the start. Second, the GPL often allows the author's potential licensees to use the code without licensing it. (For example, some vendors of print drivers for UNIX invoke GNU Ghostscript but then post-process the output through their own software after that. They don't change GhostScript itself.) So, in many cases, they have no need to license the GPLed code, and the author loses.

You've almost got that right. When I give you GPL code, you must share that with me and everyone else equally.

Translation: I must give everything away to the whole wide world. Which means that I must give away the farm. In short, what the GPL really obliges the programmer to do is not to pay back other developers; it's to forfeit any opportunity to make money from his work.

He may well choose to give some things to the world, and in practice that happens frequently. That's what the BSD and MIT X license are about. But he should not be forced to do that. Forced "giving" isn't giving; it's confiscation.

Here you go accusing me of being unethical again, Brett. It doesn't shed a good light on you.

The first rule of any code of ethics is, and must be, "Do no harm." The GPL is intended to do harm (Stallman explicitly said so in his "GNU Manifesto") and as such is per se unethical.

I think that to recognize this, and reject the GPL, sheds a very good light on anyone. To use, promote, or further the ends of the GPL is to intentionally hurt others by perpetuating an agenda of spite and malice. That's not ethical, Bruce.

--Brett

The GPL's forced "sharing" isn't sharing

[Brett Glass]

It's only other people's code where you have to obey the sharing rules.

Not true, Bruce, and it's disingenuous for you to say so. When you add one line of other people's GPLed code to you work, you must give away all of your own code in that product. Not share proportionately or fairly with the author who wrote that one other line, but give it all away, to the entire world, forever.

It's the GPL that "takes, takes, takes." And that's its purpose.

--Brett Glass

Stallman does not believe in privacy or IP.

[Brett Glass]

RMS believes people should be able to keep their code private. (And also other knowledge)

This is not so. At the MIT AI Lab, Stallman -- who served as a system administrator and system programmer -- even refused to let users put passwords on their accounts!

Yes, 'tis true, and it's documented. People who were at MIT at the time report that Stallman decrypted password files and sent messages to users saying, "I see your password is X. I suggest that you switch to the password [carriage return]. It's much easier to type, and also stands up to the principle that there should be no passwords."

He also modified the system code so that it would echo users' passwords to a public system console and system log as they logged in -- perhaps the first documented case of "password sniffing."

Steven Levy's book "Hackers" describes other ways in which Stallman fought security measures. Levy writes:

Stallman kept fighting, trying, he said, "to delay the fascist advances with every method I could." Though his official systems programming duties were equally divided between the computer science department and the AI Lab, he went "on strike" against the Lab for Computer Science because of their security policy. When he came out with a new version of his EMACS editor, he refused to let the computer science lab use it. He realizes that, in a sense, he was punishing users of that machine rather than the people who made policy. "But what could I do?" he later said. "People who used that machine went along with the policy. They weren't fighting. A lot of people were angry with me, saying I was trying to hold them hostage or blackmail them, which in a sense I was. I was engaging in violence against them because I thought that they were engaging in violence to [sic] everyone at large."

And all because the system -- which was on the rapidly expanding Net and therefore subject to cracking -- required passwords! To him, these simple provisions to prevent tampering and to keep information private were "fascism."

Stallman has also come out strongly against the notion of artists' rights. He believes that all intelletual property should be abolished and that authors of books and music not be able to hold copyrights.

--Brett Glass

Licenses and firstborn children

[Brett Glass]

Yes, I do advocate a legal system which invalidates certain contract terms. (In legal parlance, these terms are called "unenforceable" or "unconscionable.") A license which required someone to turn over his or her firstborn children would be a good (if a bit extreme) example.

"Copyleft" is legally suspect for a number of reasons, and I frankly believe that it is unenforceable.

--Brett Glass

Re:Licenses and firstborn children

[Brett Glass]

I'm interested ... what do you think of typical proprietory [sic] EULAs?

I think that companies should stand behind their products. Unfortunately, a company that does stand behind its products will not do as well as one that doesn't. Therefore, consumer laws which already require merchantability and fitness for use should be extended to software, so that the playing field is level and the standards are higher. Unfortunately, UCITA -- unless it is rejected in all 50 states -- will move things in the wrong direction.

--Brett Glass

Faustian bargain

[Brett Glass]

The GPL isn't a fair contract; it's a Faustian bargain. Its "preamble" is intended to mislead about its true intentions, and its terms are designed to come back to haunt the programmer who uses it.

What's more, it discriminates -- specifically against programmers. Anyone can use the code in the way that benefits him or her most, except the programmer who seeks to make a reasonable living by licensing his or her code. Shades of Animal Farm: "Some users are more equal than others."

Why do you insist on holding GNU software to such vastly higher standards of selflessness than commercial software?

Because it claims to have "such vastly high standards of unselfishness."

--Brett

Red Hat lost $1.5M per employee this quarter.

[Brett Glass]

Brett, your whole argument relies on a a completely unsubstantiated premise, namely that you can't make money with GPLed code. Unfortunately - for your argument - Red Hat is making a fortune

Not so. Red Hat has never made a dime and has in fact lost millions of dollars per employee. What's more, it does not own Linux -- the product it sells -- so it also has virtually no assets other than computers, desks, chairs, etc. No profits? No assets? Doesn't sound like a strong business proposition to me.

Red Hat's own Form 10-Q, filed with the SEC, states:

RISKS RELATED TO OUR LINUX-BASED OPEN SOURCE BUSINESS MODEL

OUR BUSINESS MAY NOT SUCCEED BECAUSE OPEN SOURCE SOFTWARE BUSINESS MODELS ARE UNPROVEN

We have not demonstrated the success of our open source business model, which gives our customers the right freely to copy and distribute our software. No other company has built a successful open source business. Few open source software products have gained widespread commercial acceptance partly due to the lack of viable open source industry participants to offer adequate service and support on a long term basis. In addition, open source vendors are not able to provide industry standard warranties and indemnities for their products, since these products have been developed largely by independent parties over whom open source vendors exercise no control or supervision. If open source software should fail to gain widespread commercial acceptance, we would not be able to sustain our revenue growth and our business could fail.

--Brett Glass

Re:The FSF and "stewardship"

[Brett Glass]

The FSF cannot withdraw the GPL from a program.

It can, in a sense. When a new version comes out, it can license it only under the latest version of the GPL and not any older one(s). Who would want to run an older one with known bugs? The FSF can thus tighten or otherwise change the licensing essentially at will.

--Brett Glass

I support open source. The GPL is NOT open source.

[Brett Glass]

And if you believe the conspiracy theories above, I have some GPLed code for you. Go ahead, use it in your programs. It won't do you any harm. All you must do is give up all the income you would have made as a programmer -- forever.

It just so happens that I support open source vigorously. However, the GPL is not open source, because it is not open to all. Anyone can use GPLed software in the way which benefits him or her the most -- except commercial software developers. This discrimination was explicitly intended, by Richard Stallman, to reduce programmers' economic prospects to those of starving graduate students. He says so in his "GNU Manifesto."

The GPL is not a valid open source license, even according to the "official" Open Source Definition posted at http://www.opensource.org. Why? Because it violates a key principle listed there: an open source license must not discriminate against a field of endeavor. The GPL discriminates against the well-established field of commercial software development and is thus not a legitimate open source license. It is unethical to use the GPL for this reason.

The MIT X License and the BSD license, by contrast, do qualify because they do not discriminate.

Re:The GPL forces forfeiture of IP, income

[Brett Glass]

Every single line of code that has been licensed under the GPL has been licensed as such because the AUTHOR OF THE CODE WANTED IT THAT WAY.

Not true. Stallman himself says that the goal of the GPL is to force programmers and their employers to give away their work. The GPL is intended as a lever with which to pry it away from them, no matter how much they are entitled to it.

However if I wanted to use a line or module from a GPL'd app and distribute it under something else, I'm fully free to CONTACT the author and ASK his permission; and if what I'm doing doesn't violate what HE SPECIFICALLY AS A HUMAN INDIVIDUAL wants, then he'll allow me to do it.

Actually, the GPL is designed to prevent this. Because it aggregates the works of many authors, and any can veto the commercial use of the code, it is generally impossible to locate all of the authors. And the odds of securing an agreement from every one of them are vanishingly small, because authors of GPLed code often subscribe to the GPL's malicious agenda.

Brett, your points simply have no merit,

Proof by assertion: you say it, therefore it must be so. Sure.

and stem from a need to hate things which are successful. Hatred is just too easy; try something else.

An unwarranted ad hominem attack. The GPL is motivated by spite and anger (and this is well-documented; read Stallman's own words or any of several published accounts of how it came to be. My postings are motivated by concern for programmers, the future of the industry, and ethics. It is unethical and wrong to attack and hurt people out of spite, and that is what the GPL does.

--Brett Glass

Appropriation of others' work

[Brett Glass]

IP exists as either copyrights or patents. The GPL does nothing to appropriate someone's copyrights or patents.

Stallman advocates its use for precisely that purpose. In his essay, "What is Copyoleft?" Stallman advocated the surreptitious use of GPLed code by programmers who wish to create GPLed software on their employers' time and then force the employers to give away the result:

Copyleft provides another benefit as well. People who write improvements in free software often work for companies or universities that would do almost anything to get money. A programmer may want to contribute her changes to the community, but her employer may ``see green'' and insist on turning the changes into a commercial product.

When we explain to the employer that it is illegal to distribute the improved version except as free software, the employer usually decides to release it as free software rather than throw it away.

It's the programmers who are not allowed to "confiscate" the IP of the GPLd work when a copyrighted work is covered under the GPL.

Not so. GPLed software has zero market value, since it is available for free to all users. Therefore, programmers should be able to use it completely freely in their own work, whatever that work may be, at no charge and with no obligation. Why does the GPL begrudge them this? Because by doing so it undermines their livelihoods.

--Brett Glass