Submitted by Silicon Graphics Inc. has filed for Chapter 11 bankruptcy. The press release sugarcoats: “Silicon Graphics today announced that it has reached an agreement with all of its Senior Secured bank lenders and with holders of a significant amount of its Senior Secured debt on the terms of a reorganization plan that will reduce its debt by approximately $250 million, greatly simplifying its capital structure.” El Reg, The Inq, and the WSJ have more.
And mourn for this once great company? Anyone with any good commentary on this?
SGI have been screwed since about 1998 or thereabouts, it’s amazing that they have lasted so long. It was a major mistake trying to go into the Windows NT workstation market, a major waste of money which should have been put into developing Irix and their MIPS systems. The demise of the R18000 chip back in 2003 was really the final nail in the coffin for their IRIX kit that’s for sure. Great systems and all but who can justify spending 18k on a graphics workstation when the graphics system hasn’t been updated since 2002 (Tezro with Vpro12).
Likewise they didn’t continue development of their heavy metal visualisation systems after IR4 (Infinte Reality 4 — released bout 2002) instead bringing out the Onyx 300/350/3000 series with ATI gpu’s. As we all know ATI drivers aren’t that great whatever platform you tend to be on.
If they had stuck with what they were good with instead of going intel they’d probably be in better state today.
It’s a pity, because IRIX is my favourite Unix out there, awh well c’est la vie.
I hope that OpenGL lives on for a long long time to come. Brilliant technology and we should all go out and buy a copy of Mark Kilgards book on OpenGL. SGI had some great days but missed an opportunity to rethink and reinvent in the past 5 years.
Sadly .. the first thing that they do is steal from their own investors and only the “secured” banks get anything to pick from this sinking boat. So, dump your stock right away because its worthless today and has no provision for recovery in the near future.
“the Company believes that SGI’s currently outstanding common stock and unsecured subordinated debentures have no value.”
That really says it all .. almost.
This is the bitter pill to swallow :
“All of SGI’s existing common stock and the unsecured subordinated debentures will be cancelled upon confirmation of the plan by the court and receive no recovery.”
wow .. so the message is run .. run for the hills.
And mourn for this once great company?
Yes, of course. Every time an innovative computer company gets into trouble, we all lose. The reason is that it becomes harder and harder to change the accepted paradigm. Our habits and ways of thinking are thus monopolized by the prevailing powers. Fortunately for us all, SGI is not dead yet. There is still hope that they may hang in there and go back to their former innovative and creative selves. My advice to SGI is the same one I give to Sun Microsystems:
Don’t try to beat either Intel, Linux or Microsoft at their games. You will lose. I suggest instead that you do something that will take the rest of the industry completely by surprise. Invest your remaining resources and passion into the next big thing, the one thing that will solve the nastiest problem in the computer industry today: unreliability. Put all your money in non-algorithmic, signal-based, synchronous software. It will revolutionize both the hardware and the software industry and usher in the most dramatic change in computing since the days of Charles Babbage and Lady Lovelace. Don’t say you weren’t warned. ahahaha…
> Put all your money in non-algorithmic, signal-based,
> synchronous software.
Or better yes, don’t put your money in anything software based, since as soon as you make a breakthrough, you will find you can’t capitalize on it anyway since the open source people will reverse engineer it, figure out how you did it, copy it in such a way that makes just enough changes to avoid getting sued for copyright infringement, and then give it away free, once again, reducing the monetary value of your capital and time investment to zero.
Again, this is why the whole idea that closed source software hurts innovation is a total FUD myth spread by the open source crowd. It is actually open source leaching of ideas that hurts innovation, since companies don’t want to invest the millions of dollars and man-hours it takes to come up with innovative solutions when they know that the open source people will have a competing knock-off product in a year or two that will reduce the monetary value of their product to zero before they have even had time to reccover their expenses.
Simba, you’re right but consider this. The open source community is not really the source of the problem. The patent system is. We are not doing it right. The idea is to compensate and reward innovators for their hard work, not to infringe on freedom. All intellectual property should be free. We should keep the patent registration process but instead of putting restrictions on use, we should find an alternative way of compensating the creators. IMO, society as a whole should pay on the basis of the idea’s importance (how many people use it and how beneficial it is to society) and the amount of work it took to bring it to fruition. Some compensation formula can be perfected and, if changes are needed as we gain experience with it, retroactively applied.
Having said that, and taking current reality into account, I agree that SGI should not be in the business of selling software. Their forte is their hardware know-how. However, any potentially revolutionary processor must be based on a revolutionary software model. I am convinced that the sign-based synchronous model is the way of the future.
I am convinced that the sign-based synchronous model is the way of the future.
Correction. I meant to write “signal-based” and not “sign-based”.
> All intellectual property should be free.
No, it shouldn’t Especially when it comes to software, and especially not in today’s world of open source leaching of ideas.
The vast majority of time and money spent on software development is spent on the intellectual property aspect of it, not on the actual coding implementation itself. And in today’s world where companies can’t protect the implementation itself against open source leaching and cloning, they have no choice except to resort to patents and protect it at the intellectual property level.
> IMO, society as a whole should pay on the basis of the
> idea’s importance (how many people use it and how beneficial
> it is to society) and the amount of work it took to bring it to
> fruition. Some compensation formula can be perfected and,
> if changes are needed as we gain experience with it,
> retroactively applied.
Tha’ts never going to happen. And you are out of your mind if you actually want it to happen. Here’s why: What authority determines how valuable an idea is to society, and determines how much the rest of the “peasents” should be forced to pay for that idea? This kind of system is not freedom at all. It is opression. What if this governming body determines that something is of high value to society, but because I don’t follow the “main stream beliefs of the time”, I find it morally or ethically repulsive. I should NOT be forced to pay for that just because some group of so called “experts on what society needs” determined it was in my best interest to have to pay for that.
Yes, the open source community is the source of the problem. It’s not worth a company’s time to spend millions of dollars to do R&D on innovative software given how fast the open source community will clone it and reduce the monetary eturn on investment to zero.
Edited 2006-05-09 16:09
You are mistaken. I am not promoting the idea that compensation should be decided by a government body. It should be a peer (i.e., democratic) system where the people decide. And you are forgetting that a government body is already imposing its dictatorial will on the rest of us and has already decided to infringe on the people’s freedom so as to give an unfair monopolistic advantage to a minority while enslaving the majority. Your problem is greed, selfishness and enforced controls, not the leeching of ideas.
> I am not promoting the idea that compensation should
> be decided by a government body. It should be a
> peer (i.e., democratic) system where the people decide.
So now the “moral majority” gets to decide what my money has to go to, and gets to decide what a good idea is tha t benefits society? Because they know what is best for me? Sorry. No thanks.
Sorry to burst your bubble. But socialism doesn’t work. The Soviet Union proved that all to well.
No. Nobody is trying to tell you what is good for you. You’re the one who is promoting socialism for the benefit of the few by imposing artificial and unfair controls on the market. I am promoting a free market system in the purest sense of the word. As a case in point, there are many excellent windowing systems out there. Why should only a few people reap monumental rewards just because they know how to exploit the IP system better than others by creating a monopoly? Heck, they did not even invent the desktop. Of course, you want to maintain the status quo. Why? Because of greed, dishonesty and selfishness, that’s why. Sorry for being so blunt, but I always tell it like I see it.
> You’re the one who is promoting socialism for
> the benefit of the few by imposing artificial
> and unfair controls on the market.
I have not promoted any socialist ideas. If you think I have, I think you have a somewhat warped idea of socialism.
You are not promoting a free market system at all. You are promoting purely socialist ideas. The idea that everyone has to contribute to things that society in general determines is good is a purely socialist idea.
> Of course, you want to maintain the status quo.
No I want a system where people can actually get financial rewards for their hard work and fiancial investment. Not a system where people do hard work and spend a lot of money, only to be ripped off by free-loaders who clone it and give it away free.
> Because of greed, dishonesty and selfishness
You might do well to read Richard Dawkin’s The Selfish Gene. Selfishness? Yes. That’s natural human condition. And its the primary reason socialism doesn’t work.
Dishonesty? No. But please tell me how it is not dishonest to steal and clone someone else’s hard work and research, and then give it away free?
Dishonesty? No. But please tell me how it is not dishonest to steal and clone someone else’s hard work and research, and then give it away free?
It is dishonest, cruel and unfair to build a multi-billion dollar monopolistic empire starting with a lousy product like DOS. Just my opinion. In fact, I think it’s criminal, especially in a country with 4 million homeless people.
One more thing, you already live in a highly socialist country whether or not you realize it. Any system that gives 90% of private land to an elite minority and confiscates the rest has to have a socialist system in place in order to keep the people (read slaves) from revolting.
> It is dishonest, cruel and unfair to build a multi-billion
> dollar monopolistic empire starting with a lousy product like DOS
A: IBM’s loss was Microsoft’s gain. IBM failed to recognize that the money was going to be in software and not hardware. And that’s where Bill Gates capitalized.
B: Consumer created monopolies are not illegal: Consumers have a choice. They overrwhelmingly choose Windows PCs. They don’t choose Linux because Linux has usability problems for the average user. And they don’t choose Mac beccause Apple’s prices are too high for the average consumer.
> Any system that gives 90% of private land to an elite minority
> and confiscates the rest has to have a socialist system in place
> in order to keep the people (read slaves) from revolting.
Citation please? I didn’t think so. You pulled that number out of your ass and you know it. And it is pure fiction.
Edited 2006-05-10 03:17
What I mean is that the IP laws that allows an IBM or a Microsoft to monopolize a market are criminal to start with.
And I am not that far off in my estimation of land ownership. A minority does own over 90% of private property. And the government (federal and state) does own a shitload of land, 60-90% in some states. Here are some stats:
“80% of U.S. wealth is held by 20% of the families in the U.S.
The richest 5% of families own more than half of all private property.”
My source:
http://www.sage.edu/academics/schoolofprofessionalstudies/managemen…
PS. This my last post on this thread.
> What I mean is that the IP laws that allows an IBM or a Microsoft
> to monopolize a market are criminal to start with.
Except that patents are often designed to protect small companies from larger ones.
Case Study; If you are old enough to remember the DOS days, you might remember a product called Stacker, produced by a little company called Stac Electronics, which allowed you to increase your hard disk space by compressing and uncompressing data on the fly (this was back when hard disk space was really expensive and the average person had a 10 Mb to 20 Mb hard disk). When Microsoft came out with DOS 6, they included a utility called DoubleSpace which… You guessed it… Allowed users to greatly increase their hard disk space by compressing and decompressing data on the fly.
To make a long story short, Stac Electronics sued Microsoft for patent infringement… And they won.
The moral of this story? Often, patents protect the little guy. Not the big guy.
>The richest 5% of families own more than half of all private property.”
“half the property” equals 50%… A far cry from the original 90% you tried to claim don’t you think?
And besides, keep in mind that a lot of people don’t own private property because they choose not too. I make a pretty decent salary. but I choose not to buy a house because I don’t want the burden of having to sell it if I ever decide to move. I also don’t want the unpredictable expenses of maintance and that kind of thing. So I prefer to rent instead.
So that number you quoted is misleading, and basically a worthless statistic.
Edited 2006-05-10 05:21
Sad to see all non x86 platforms disappear
Actually POWER is still out there, and Genesi sell PowerPC systems. And of course there’s still SPARC, and Itanium is selling like hotcakes! ;-P
Seriously, though: Wow, guess I should have seen this one coming.
Sad to see all non x86 platforms disappear
Indeed, but SGI dissapearing has nothing much to do with it, since they dumped alpha for Itanium a long time ago. And we could only wish Itanium would dissapear along with SGI.
BTW, ARM and PPC aren’t that much far behind to X86 in volume of sales. ARM for example dominates 75% of the embedded market (PPC probably has a sizeable chunk of the rest). Might not be the workstation/desktop market, but still..
And PPC didn’t do that bad in the servermarket.
SGI never sold Alphas, they used MIPS processors.
HP’s the one that dumped the Alpha for Itanium.
You mean HP dumped both the PA-RISC (old HP) and Alpha series (old Compaq through DEC purchase) for Itanium?
Damien
Yep. To be fair, dropping PA-RISC for what became Itanium was on the cards long before HP picked up Compaq, in fact from around the time Commodore went belly-up, iirc.
SGI never sold Alphas, they used MIPS processors.
Are you sure?
I know they used MIPS, but I remember seeing adverts in the UK edition of PC Pro a long time ago with “Silicon Graphics” selling Alpha based workstations at 667Mhz – I remember thinking how fantastic it was as everyone else was selling 400Mhz Pentium II’s at the time!
I’m presuming that “Silicon Graphics” was later renamed SGI – is that correct?
Strangely enough i’ve never been able to track down any info about those systems as I recently wanted to try and purchase one before Alpha went completely terminal.
SGI bought Cray Research, which, at that time, had a DEC alpha based “supercomputer”. So yes, SGI did sell alpha based product for a while, but they’re own workstations were always MIPS based before they switched to x86.
SGI never sold Alphas, they used MIPS processors.
HP’s the one that dumped the Alpha for Itanium.
Oops, my bad. Either way: Intell sold crack to both of them
“HP’s the one that dumped the Alpha for Itanium.”
intel bought the alpha development team to help creat itanuim. and the future version of itanium that has 4 cores and is slated for release in early 08 is absolutly amazing http://www.realworldtech.com/page.cfm?NewsID=361&date=05-05-2006#36…
Chapter 11 bankruptcy does not mean that the company will cease to exist anymore. Hell, look at K-Mart, they filed chapter 11, what, 3 years ago?
You file for bankruptcy because you have more debt than you (or your company) can handle and you neggotiate with the bank to forgive some debts, make payment plans for others, and allows you to just reorganize things much more easily to a place where your company can handle the rest of the debt that you have.
SGI is not a dead company, at least not yet.
Most people probably get the impression that Ch.11 -> nonexistence because Commodore filed for the equivalent in a Bahamanian court and then, er, non-existed.
You are absolutely right, Ch. 11 is not the end, but it is not possible to compare SGI with K-Mart. The former has been in a downward spiral for the longest time, and can do very little to change the trend. K-Mart has and is in a fluctuating market where actions that increase their margins, even a little bit, can turn around the company. It is a cutthroat market there K-Mart is in, but they had many different options to increase their margins, and they picked a good one.
First of all going down the x86+windows was a huge mistake. Then changing their logo to sgi was another mistake finally going down the Itanium+Linux was the fatal blow.
SGI can still come back if they just throw out Itanic+Linux and go back to AMD64+IRIX.
I wasn’t aware that IRIX had been ported to AMD64; are you sure you don’t mean MIPS?
Either way, I think IRIX and MIPS are dead in the water at this point.
SGI can still come back if they just throw out Itanic+Linux and go back to AMD64+IRIX
just throw out Linux. SGI was doing great before they began offering linux. SGI should improve IRIX, so can compete with Solaris, BSDs and Windows. If SGI is wise they will see the decision they made to go with linux was a mistake it did not pay off.
Porting IRIX to AMD64 is a no runner pure and simple, most of the IRIX devs have been made redudant, this isn’t suprising seen there hasn’t been a major release of IRIX since 1998 (6.5), all subsequent releases have been point releases which basically are security patches/new hardware drivers/some minor improvements etc.
They did start the IRIX port to Itanium but it was canned in 1999 (or thereabouts) when they decided Linux was next Best thing ™.
*start pipe-dream*
IRIX is a “Unix95” OS, if they were to go back into the whole Unix market they be better off doing a deal with Sun, porting Opensolaris to Itanium (though Sun supposedly did this with solaris couple years ago) and adding their value Add ontop. Least that way they would have a “modern Unix2003” OS, I’ve always found xsgi to be one of best X servers that I have used, so there probably is enough in their IRIX codebase that they could add ontop of Opensolaris to differate it from Solaris.
However tbh, just because ye wish something doesn’t mean it will happen, so I don’t see them budging from supporting Linux/Itanium at this stage.
*end pipe-dream*
Lets be realistic here. Linux is what really forced SGI into bankruptcy. Many of SGI’s staple customers have been migrating to Linux running on cheap x86 clusters.
totally agree. Why would you BUY linux when Ubuntu and Fedora are practically giving it away?. SGI donated XFS to get some browny points from the “Linux Community” and tell me if there’s even a single distro shipping XFS as default.
This constant whoring to the linux community is killing once mighty companies like SGI, Novell – Sun is next but Apple isn’t on the list because Apple doesn’t do ANYTHING for linux – not even quicktime or bootcamp.
Edited 2006-05-08 18:00
Linux is what really forced SGI into bankruptcy.
You may be right. But I suspect there are companies that are maintaining marginal profitability using Linux today that would be bankrupt (due to marginal unprofitability) if they had had to buy SGI instead. It would be fun to see a comparison of the economy with and without Linux. I bet Linux adds a lot more to the economy than it takes.
> It would be fun to see a comparison of the economy with and
> without Linux. I bet Linux adds a lot more to the economy than it
> takes.
That might be true. But it would also be interesting to see how many computer science student FOSS programmers programmed themselves right out of a job when they graduate–because commercial companies they otherwise would have been able to work for have had to seriously downsize because of competition from FOSS.
You have to remember that most programmers don’t work for the big companies selling shrink-wrapped software. They work for companies designing systems to solve particular problems. The former types of companies are hurt by Linux, because they sell a competing system. The latter types of companies are helped by Linux, because Linux provides them a good, free, building block for their solution.
I’d be seriously surprised if the economic loss due to FOSS wasn’t offset by the economic gain. For every lost IIS license, there are probably a dozen cases where small businesses improve their bottom line by using Apache. For every lost *NIX license, I bet there are a dozen cases where an academic or research institution saves money by using a Linux cluster.
I bet Linux adds a lot more to the economy than it takes.
Yes, it would be interesting to see a good analysis of “with Linux” and “without Linux” to see what impact(s) it has had in the computing world.
I suspect, and this is only my opinion after a brief think – no backing facts, that Linux has been a force that ‘redistributes’ earning abilities and power. The larger (and in this case Unix-based) behemoths are falling by the wayside, unable to keep up in the long run. However, smaller local businesses are able to offer local Linux support and services. Red Hat is probably the best known example of this, since they offer their own distro of Linux.
Whether you view this as:
good: “Robin Hood taking from the rich and giving to the poor”,
or neutral: “if you are going to make an omelette, you have to break a few eggs”,
or negative “Look at the carnage! Oh, the humanity!”,
depends partly on your view of change.
Robin Hood … break a few eggs … carnage!
I don’t think any of those capture the issue. Linux creates wealth! That’s right, it is a tool that businesses have that gives them capability they would not have to do business otherwise.
Compare it to the cost of diesel fuel. If it’s cheap, you can ship stuff cheap and make a profit doing lots of things that would be impossible with a higher cost. Rising fuel or computing costs result in cost-push inflation. Falling fuel or computing costs empower more businesses to make a profit and therefore employ more computer industry people.
> Linux creates wealth!
…Unless you are one of the developers working for a commercial Unix vendor who loses their job because your company had to seriously downsize because of competition from Linux… Then it certainly doesn’t create wealth for you…
Edited 2006-05-08 19:10
The fact is that Linux is only biting the UNIX vendors on the a** because they made the mistake of charging too much for it and making their UNIXes incompatible. The Linux market is what should have happened to UNIX, and if it had we wouldn’t now be dealing with XP.
> The fact is that Linux is only biting the UNIX vendors on the
> a** because they made the mistake of charging too much for it…
That’s not really true. Because anything looks like it costs to much when it is competing against free.
Unfortunately, however, free does not pay developer’s salaries, so commercial software vendors have to downsize, layoff developers, etc.
“Linux is only free if your time costs nothing.”
For one thing, developers who develop for Linux are repaying the rest of the Linux developer community for their efforts by joining in, even if they don’t give $$.
For another, Red Hat and other Linux companies are profitable, because people who don’t have the time to cobble it all together pay them to do the job for them.
Thirdly, it’s not that going up against Linux charging high prices is killing UNIX, it’s that going up against Windows charging high prices already did kill UNIX. My edition (Third Edition?) of Running UNIX states that at the time of writing, a single-user version of AT&T System V UNIX for the PC cost $1500.
If you can afford $1500 per user for an OS, you’re probably Bill Gates, and rumour has it he has an OS already. 😉
If Linux weren’t attracting novice users who like to point and click, there wouldn’t be any incentive to develop KDE, GNOME, and YaST amongst others. It follows that if novice users had seen UNIX as an attractive proposition based on price, CDE would have become both the standard interface on desktops and a much better interface than it already was.
“….and making their UNIXes incompatible”
Oh, as if the Linux distros are the model of compatibility.
It is still ashame to see this happen to companies that helped shape computing today. I do agree with another poster how SGI really brought this on themselves.
Yes, compared to UNIX, they are. The only things that are “incompatible” between Linux distros are the installer and the package management system, with some minor variations in file structure. And most of those are shared between more than one distro anyway. That in no way compares to the headaches of supporting multiple UNIX vendors. And when Linux is old, a new OS will come along where even those problems don’t exist. But it will look a lot more like UNIX/Linux than anything else.
> Yes, compared to UNIX, they are. The only things that are
> incompatible” between Linux distros are the installer and the
> package management system, with some minor variations in
> file structure.
Not really. Having to build a package by source because of differing library versions on Linux distros is an extremely common occurance. So there are quite a few times where pre-compiled binaries don’t work across distros.
The only things that are “incompatible” between Linux distros are the installer and the package management system, with some minor variations in file structure.
You forgot system management tools. Each seems to have its own. Also, ideas of what are in the ‘default’ system and what has to be added if needed, which can be a problem in a multidistro shop when you start to make assumptions about what’s available.
Oh, and level of support for various devices.
And kernel patches that change the working feature set.
and . . .
Sounds like the same problems you’d get if you were running a mixture of Windows and Mac, or even XP and W2K.
Unless … Then it certainly doesn’t create wealth for you.
Who are all the would-be SGI customers that are using Linux instead of buying SGI? It’s certainly creating wealth for all their employees.
Consider ILM. They moved from a bunch of expensive IRIX/MIPS boxes to a bunch of Linux/x86 boxes. Their artists are more productive, because the machines are a lot faster, and their bottom line is improved, because the machines are a lot cheaper. You tell them Linux doesn’t create wealth for people.
> Their artists are more productive, because the machines are a lot
> faster, and their bottom line is improved, because the machines are
> a lot cheaper. You tell them Linux doesn’t create wealth for people.
I never said it doesn’t create wealth for some people. Just that it doesn’t create wealth for all the programmers at SGI who are going to lose their jobs now as a result of the restructuring because of the bankruptcy.
Outsourcing creates wealth too–It creates wealth for upper management. But it most certainly does not create wealth for the developers in the trenches who lose their jobs because of it, just as a point of comparision. Yeah, Linux creates wealth for some people. But for others, it costs them their jobs.
Everything creates wealth for some people and hurts others. Hell, if I do my job right every day, I’ll be screwing my competitors out of potential wealth. That’s the way of the world. The question is whether SizeOfEconomy(withLinux) > SizeOfEconomy(withoutLinux). I’d say that relationship is true.
PS) It’s the same for outsourcing by the way.
> Hell, if I do my job right every day, I’ll be screwing my competitors
> out of potential wealth.
Except if you are one of those computer science student Linux hackers, not only are you screwing commercial companies out of wealth. But you are potentially screwing yourself out of wealth as well when you can’t get a job after you graduate because you have made the products you produce into a free commodity with no monetary value.
> PS) It’s the same for outsourcing by the way.
No it isn’t. Cheap products don’t mean very much if people don’t have jobs and can’t afford to buy them anyway, and taxes skyrocket to support a strained welfare system because of increasing unemployment. Despite with the Bush adminstration would like you to believe, losing jobs overseas is not good for the American economy… Sure it’s good for the economy of the countries that are getting the business, and it is good for the pocketbooks of the big business execs who contribute a bunch of money to political candidates who will support things like oursourcing, But it’s not good for the economy here.
Edited 2006-05-08 21:17
Just how many computer science student OSS hackers do you think are out there? If we presume that there are twice as many as apply for Google’s Summer of Code (a fair estimate, I’d say), that’s about 9,000 worldwide. Do you honestly think that the contributions of 9,000 students distributed all over the entire globe is larger than the contributions of the millions (check out the Office of Employment Statistics for exact numbers) of employed developers in the US alone? I think your dislike of OSS has clouded your ability to think logically here.
First off, what product with value in the market could be out-programmed for free by a student in their spare time? Certainly nothing of any discernable size or complexity. Speaking as a student, I don’t exactly have a full work-day’s worth of free time every day. There’s no way I, or even a consortium of myself and a large subset of those 9,000 of my OSS hacker buddies, could create a piece of software in a timeframe that could effectively compete with a company who can pay developers to program all day long.
Furthermore, you seem to think that computer science students with a portfolio of code are somehow at a disadvantage in the job market. I can tell you (also from personal experience) that this is most certainly not the case. Employers in any given field *want* to hire an employee with past experience in that field. If a programmer has worked on an OSS medical claims application, they stand a much *better* chance of gaining employment with a medical claims software house than someone who has no experience in the area. It’s not only common sense, it’s reality. Live it.
Finally, you seem to be giving a lot of weight to the economic impact of the “big” *nix houses and their collapse, while downplaying *every other major sector* that stands to benefit from OSS. If you sit down and calculate the total developer employment of all the *nix vendors in the US (or the world, for that matter) and compare it to the total employment of application developers and systems developers, you’ll come up with a ratio of anywhere between 1/1,000 to 1/10,000 or more. Are you willing to say that the economic impact of that .1% – .01% outweighs the productivity gains of the other 99.9% – 99.99%? That calculation doesn’t even factor in companies that stand to gain truly vast amounts of money by effectively maintaining their application structure, but migrating their platform to a free alternative. Just about every major sector, from finance to automotive, manufacturing to research stands to save billions of dollars by migrating from proprietary Unix variants to Linux. If your argument is true, SGI’s developers are worth about 100 times their weight in platinum.
In light of real, actual logic, your arguments appear to be mainly emotionally motivated. I’d be willing to bet you’re a disgruntled SGI employee facing a possible pink slip or merely a vitriolic anti-OSS zealot. Good luck with the job hunt.
> Finally, you seem to be giving a lot of weight to the economic
> impact of the “big” *nix houses and their collapse, while
> downplaying *every other major sector* that stands to benefit
> from OSS.
We don’t have to limit it to the *nix houses:
– Borland has gone from the largest development tool vendor in the world, to barely being able to survive because OSS has comoditized development tools.
– Application server vendors have gone from being powerhouses, to being in trouble because OSS has comoditized application servers.
Those are just a few examples.
> I’d be willing to bet you’re a disgruntled SGI employee
Nope. I’m not. I actually have a decent development job. However, job security is a nice thing to have. And OSS has removed a lot of job security from development jobs. If you were a professional developer (and I suspect you are not), you would understand that. Do I ignore the greater economy? Perhaps. But that’s only natural. I’m a developer, and I care about developers. I care that developers will continue to make a decent living practicing their craft. And people who want to become professional developers will still be able too. I don’t want to see professional software development become a lost career because there is no money in it anymore because OSS has comoditized everything.
> anti-OSS zealot.
No. But I am anti “let’s sit around and see what kind of innovative software the commercial companies spend a lot of money to research and develop, and then cloan it and give it away free”. And unfortunately, that’s what a lot of OSS software is. Just clones of softwae that cost commercial companies a lot of time and money to research and develop.
Edited 2006-05-08 23:04
You *continue* to turn a blind eye to rayiner’s and my point that the vast majority of developers are *not* employed in these sectors. Most work in industries with no software output whatsoever.
I am a professional software developer (contrary to your assumption) and care deeply about my craft. I feel the value of the software I create is in its functionality, not the license it’s distributed under. You appear to have overlooked the obvious fact that most contributors to open source projects are *also* professional software developers. You need to get over this baseless fear of OSS as the great capitalism-killer. As long as there’s money in making computers run, software developers will have job security.
Job security is a wonderful thing, but there are a number of ways to attain it other than trying to suppress innovation. You seem to be one of those people who cried for the candle manufacturers when the light bulb caught on. Capitalism, my friend, is an adapt-or-die proposition. When people buy a product, they do so because of the value it offers them. You seem to think the only value in OSS is its tiny price tag. You may need to reconsider this assumption.
Furthermore, your claim that all OSS is “cloaned” (I think you meant “cloned”) from proprietary software is ridiculous on any level. You’re just the kind of person who would be downright shocked to find out how much the proprietary world has lifted from the Open Source realm (most notably the BSD networking stack). There’s plenty of innovation on both sides of the fence. The trick is pushing that tinfoil hat back from your eyes long enough to see it.
Furthermore, your claim that all OSS is “cloaned” (I think you meant “cloned”) from proprietary software is ridiculous on any level.
I don’t recall the original claim being that all OSS is ‘cloned’, but it is certainly true that many OSS projects consist mainly of replicating functionality already present in commercial systems. The Gimp, and OpenOffice, are two pretty good examples, and Gnome is trying hard to be a Windows UI clone.
>Gnome is trying hard to be a Windows UI clone.<
Gnome is much more OSX like in its behavior then Windows like. KDE would be the semi windows clone. However, neither is really a clone of anything, they are both similar in some ways to other os’s, however they are both so very very different as well.
The entire software industry is built upon taking previous ideas and building upon them to make them into something better, this is true on both ends of the spectrum (oss and proprietary)and both take ideas from eachother. Why reinvent the wheel?
True innovation in the software industry is very rare in this day and age.
Edited 2006-05-09 01:07
> Gnome is much more OSX like in its behavior then Windows like.
I would really have to disagree there. Gnome is much more Windows like than OS X like. Gnome doesn’t really resemble OS X behavior at all.
Isn’t Gnome supposed to be like Classic Mac OS?
> Job security is a wonderful thing, but there are a number of
> ways to attain it other than trying to suppress innovation.
Ah yes… The old “supress innovation” myth of OSS… Well, I hate to break it to you, but that’s what it is: A myth. OSS does NOT foster inovation. In fact, it stifles it. And here is why:
Innovation takes a lot of R&D. And R&D costs a lot of money. That’s simply a fact. Now tell me, where is the incentive for a company to fund R&D if they do not have reasonable assurance of return on investment? If they are going to spend millions to develop something, only to have the open source community clone it and then give it away free, there is no incentive to inovate. And so inovation stops.
Don’t think innovation and improvement costs money? Well, let me give you an example: Gnome usability was horrible for many years. When did Gnome really become a decent environment? Only after Sun injected huge sums of money to fund a usability study and figure what kinds of problems typical users had using Gnome, and what could be done to fix them.
On a side note, this is why software patents will become more and more common as open source becomes more common. Commercial software companies can no longer protect their R&D investment at the implementation level because of OSS cloning. So they are left with their only option being to protect it at the “idea / IP level” level.
> Furthermore, your claim that all OSS is “cloaned” (I think you
> meant “cloned”) from proprietary software is ridiculous on any level
First of all, I didn’t claim that ALL OSS is cloned. I said much of it is. And is that so ridiculous? Name me one innovative product that has come out of the modern open source movement. (I’m not talking about things like the BSD networking stack that came out of University research projects).
Edited 2006-05-09 01:04
Only *one* innovative open source product? Wow, you have pretty low standards. Or else you’re pretty damn ignorant for a supposed developer. Ok, how about the Ruby programming language. Or the Python programming language. PHP? Perl? You know what scripting languages are, right?
You’ll probably try to find some way to weasel out of these, so let’s start with some things that might be simpler to comprehend for a non-developer.
The ARPANET was developed openly using RFCs in what is widely regarded as the first collaborative open source project. That little project blossomed into what we “developers” now call the Internet.
Is the Web itself innovative enough (the first web server and browser were open source, as is the most prominent web server today)? What about the concept of Wikis? Or Wikipedia specifically? Hell, the whole concept of worldwide online collaboration was brought about by the development of the Linux kernel. Is the largest and most complex collaborative project of all time innovative enough? No?
Maybe Dasher is innovative enough for you. If you’re unaware of what that is, I encourage you to use Google’s services (which are built entirely upon the non-innovative open source projects I listed above) to research it further. Perhaps Jabber, the first truly extensible chat protocol (also adopted and improved upon by Google) is innovative enough for you.
Bittorrent has revolutionized file transfer and publishing. CVS and now Subversion are the de facto standards in distributed development, although you already knew that.
By the way, good call using a *usability study* as your one example of innovation. I in no way said there’s no room for money in open source or development. You, however, made the wild claim that open source is anti-capitalism and detrimental to the economy. You have yet to actually prove this claim, but have chosen to attack (very poorly, might I add) details and semantics in my arguments. What’s next? An argument on what exactly constitutes innovation?
The products that I’ve outline above not only meet your requirements as open source and innovative, they’ve also, for the most part, vastly *improved* the economy and *created* jobs for developers such as you and I. Or are you willing to argue that a world without the Internet, the Web, the aforementioned scripting languages and the most popular revision control systems would be *better* for developers? Would there be more jobs then?
I eagerly await your reply.
P.S. On a side note, the rise in software patents has *absolutely nothing* to do with OSS. It has everything to do with lawsuits. Against other companies. If you actually sat and thought before penning your grammatically poor response, you’d have realized that the profits a company would stand to gain from suing a disjoint international confederation of individual programmers with no consolidated center and no collective funding are essentially zero. Companies compile patent portfolios to do war against each other. Even the most dense of the anti-OSS zealotry acknowledge this.
> Ok, how about the Ruby programming language. Or the
> Python programming language. PHP? Perl? You know what
> scripting languages are, right?
What’s so innovative about interpreted languages?
> The ARPANET was developed openly using RFCs in what is
> widely regarded as the first collaborative open source project.
ARPANET was developed by the DOD using the assistance of Universities. Not exactly a traditional open source project.
> What about the concept of Wikis? Or Wikipedia specifically?
What’s so innovative about wikis? They are public web sites anyone can edit. How is that innovative? It’s nothing more than a twist on technology that already existed for a long time.
And what is so great about Wikipedia? All it is is a hotbed of blatent plagurism, uncited “facts”, and often times very innaccurate and sometimes even downright slanderous information.
> The products that I’ve outline above not only meet your
> requirements as open source and innovative,
No, they don’t. There is nothing innovative about a wiki. There is nothing innovative about yet another scripting language that just mixes ideas from Smalltalk and ideas from Java.
> On a side note, the rise in software patents has
> *absolutely nothing* to do with OSS.
You are totally kidding yourself if you think that commercial companies will not turn to software patents to protect themselves against open source leaching of their ideas. As I said, they can’t protect their software at an implentation level. So they have no choice except to protect it at the idea / IP level.
> you’d have realized that the profits a company would stand to
> gain from suing a disjoint international confederation of
> individual programmers with no consolidated center
> and no collective funding are essentially zero.
Completely and totally wrong. And if you had stopped to think for a moment, you would have realized that collecting money from a lawsuit is not the only motive for enforcing a patent. Another motive is to prevent any further loss and damage in profitability by forcing another entity who is infringing on your patents to cease and desist distributing the infringing product. Even if the software companies cannot collect damages from open source, they can force them to stop distributing the infringing product, which can prevent further damages to the company.
And the fact that you seem to have to resort to personal insults indicates you are losing this argument.
Edited 2006-05-09 04:36
Whoa, whoa, whoa there, cowboy.
Now, I know I foretold you ducking the whole interpretted programming language thing and my ability to call you on your own predictability may have insulted you, but did you even read my main point? Here’s a hint, it was above the section that began “P.S.”. I see you’ve avoided it, so I’ll spell it out for you again: the Internet, the Web and every single networked collaborative tool have an open source pedigree. Read again over my last post. There’s some pretty innovative stuff in there. Certainly at least one product, no?
Now, to my original question: would you argue that developers today would be better off if there were no Internet, no Web, no PHP, no Python, no Perl, no Ruby, no CVS, no Subversion and none of the other of the slew of open source tools that have revolutionized the technology industry? Please answer this. Ignore the rest if you absolutely must. At all costs, resist the urge to debate trivialities and cherry-pick.
The fact that you seem to have to resort to arguing semantics instead of addressing the main point indicates *you* are losing this argument.
P.S. Wikis realize Tim-Berners Lee’s original concept of a truly collaborative web. And I fail to see how plagiarism makes Wikipedia any less credible a resource when used as a first stopping-off point for information on a topic. The Web itself is full of plagiarism, innacuracies and slander. Do you disavow its utility as well?
> Now, to my original question: would you argue that developers
> today would be better off if there were no Internet, no Web, no PHP,
> no Python, no Perl, no Ruby, no CVS, no Subversion and none of
> the other of the slew of open source tools that have revolutionized
> the technology industry?
Developers would be better off without open source trying to hijack virtually every commercial software product anyone comes out with today. Which is basically what it has been doing lately. (And yes, developers would be better off without PHP and Perl. A lot of the security holes so common in Web apps would not exist if people didn’t use those two inherantly insecure languages for writing Web apps. Both are very poorly designed languages from a maintainability standpoint and a security standpoint… You did ask…)
> The fact that you seem to have to resort to arguing
> semantics.
I have addressed the main point. You on the other hand, have tried to throw up red herrings by pointing out “pseudo-open source” projects that in reality were actually government funded research at Universities.
> And I fail to see how plagiarism makes Wikipedia any less
> credible a resource when used as a first stopping-off
> point for information on a topic.
Vandalism (which is a constant problem at Wikipedia) is one thing that makes it less credible. The fact that virtually nothing is cited makes it less credible, etc.
But I give Wikipedia two years before it is shut down anyway. It’s only a matter of time before it is hit with a massive class-action copyright infringement lawsuit, similar to the one Google is being hit with.
Edited 2006-05-09 05:20
What’s so innovative about interpreted languages?
Python and Perl are fairly innovative in some of the details of their syntax and semantics, though on the whole they are indebted to previous designs. However, the point remains that in the space of programming language design, the innovation is squarely in the OSS camp. Java and C# are systematically ripping features off of Lisp and ML. The truely interesting stuff (provable languages, advanced type systems), are being developed as open source projects at universities.
ARPANET was developed by the DOD using the assistance of Universities. Not exactly a traditional open source project.
Aie, so we’re distinguishing between “traditional” and “non-traditional” open source now? Open source at the academic level *is* the traditional open source project.
How is that innovative? It’s nothing more than a twist on technology that already existed for a long time.
Most “innovation” today is a twist on technology that has already existed. It takes a substantial leap of logic to go and allow random people to edit your website.
As for the whole patent thing, I don’t know if companies really want to go down that route. Most new ideas in CS are developed in academia, which is in the Open Source camp. Amazon could’ve patented one-click shopping, but Stanford could’ve patented garbage collection. A patent free-for-all has the potential of becoming uncomfortable for everyone involved.
Just how many computer science student OSS hackers do you think are out there? If we presume that there are twice as many as apply for Google’s Summer of Code (a fair estimate, I’d say), that’s about 9,000 worldwide.
Thing is…I don’t think that’s a fair estimate at all. The number of people that apply for GSoC is a tiny percentage.
My point (which still stands) is that the number is tiny in comparison with the number of developers currently employed. I appreciate your baseless contention, however. My estimate, at least, had some basis in reality, and was not pure conjecture.
I challenge you, however, to go out and troll the contributor lists for all the open source projects on, say, sourceforge and then narrow the lists down to individuals currently still enrolled in school. The number, you’ll find, is actually pretty low. Most OSS developers on sustained projects are actually employed (usually by a company that uses or distributes OSS).
Again, I don’t have the time to do this, so I used a number that was known (Google’s SoC applications) and, although it represented a subset of worldwide college developers, I only statistically applied it to US employment numbers.
This number is still probably high, however, as there are only about 48,000 computer science students in the United States. My calculations assume an already high OSS contribution rate of 1/5 CS students. What numbers do you think are sane? Would you propose that more than 20% of the CS students in the US are actively contributing to OSS projects? You’ve clearly never been in a CS program if you think so. That’s unbelieveably high, considering Linux’s current desktop market is lower than this and Microsoft actively funds most CS departments in the US in some form or another (subsidized software, free training, etc). I, personally, believe the number is far lower and my point still stands. I encourage you, however, to prove me wrong.
Well, actually, I have been in a CS program. And obviously computer science students are not the only ones contributing to open source. But you are hung up on that and avoiding the real issue, which is that OSS can, and does put developers out of work.
And, no, Microsoft does NOT fund most CS departments in the US. In fact, the University where went to school did not even use Microsoft software in their curiculum because they didn’t want to use platform specific tools.
But you are hung up on that and avoiding the real issue, which is that OSS can, and does put developers out of work.
If you are so intent of believing this garbarge then prove it. I find your comments especially funny considering OSS developers like Linus, Ben Goodger, and Sean Egan were all hired because of their Open Source projects, not in spite of them.
> If you are so intent of believing this garbarge then prove it. I find
> your comments especially funny considering OSS developers like
> Linus, Ben Goodger, and Sean Egan were all hired because of their
> Open Source projects, not in spite of them.
And meanwhile, virtually every commercial UNIX vendor has layed off developers, as has virtually every commercial development tool maker. Some of those development tool makers don’t even exist anymore–all because their flagship products were turned into commidities with no monetary value by open source.
And meanwhile, virtually every commercial UNIX vendor has layed off developers, as has virtually every commercial development tool maker. Some of those development tool makers don’t even exist anymore–all because their flagship products were turned into commidities with no monetary value by open source.
OSS had nothing to do with the demise of Unix and its commercial development. The cost of competing with Microsoft was the usual reason for companies getting out of the Unix business. This had more to do with the way the computer industry operated when Unix was first deployed and the plethora of incompatible Unix variants.
The upshot was that a lot of Unix vendors each spent about the same money developing equivalent but competing features. They then each sold their Unix variant to a small market, meaning that their cost per customer was high, compared to Microsoft, which sold to a much larger customer base.
Had the Unix vendors paid attention to the handwriting on the wall sufficiently early they might have gotten behind a unified Unix, shared development costs, and gotten to a competitive cost per customer, but OSF and the Sun/AT&T collaboration came far too late in the process. (Not to mention that OSF produced nothing but crap.)
What does that tell you, other than that there is no longer a market for commercial UNIXes and development tools? Hell, Microsoft gives their devleopment tools away for free — I don’t see you hurling any vitriol in their direction…
As for the UNIX vendors — their market share was on the decline before Linux ever showed up. The fact that they’re losing money has less to do with open source than with the fact that nobody wants to buy the products they’re selling.
> As for the UNIX vendors — their market share was
> on the decline before Linux ever showed up.
Um… Actually, Sun’s prime was in the mid to late 90s…about the same time Linux was really starting to make itself known.
> The fact that they’re losing money has less to do
> with open source than with the fact that nobody
> wants to buy the products they’re selling.
And quite simply, the reason no one wants to buy them, is because they can get open source clones for free. So yes, free open source very much has to do with the reason why no one wants to buy the products they are selling.
Edited 2006-05-09 20:22
And meanwhile, virtually every commercial UNIX vendor has layed off developers, as has virtually every commercial development tool maker. Some of those development tool makers don’t even exist anymore–all because their flagship products were turned into commidities with no monetary value by open source.
Numbers, numbers, numbers. You cannot prove anything you say without numbers. How do you know that those developers didn’t get jobs elsewhere? Perhaps even working on OSS projects. Companies that cannot compete in a free market don’t deserve to exist. If you want to assign fault somewhere then assign it to the free market.
It’s amazing that you cannot even name one actual company or one actual product that lost out because of Linux. The reason is because your argument is bunk. The biggest consumers of software are corporations and those corporations aren’t using unsupported software. Even if they did switch to OSS solutions they have support contracts with Redhat or Novell.
> How do you know that those developers didn’t get jobs
> elsewhere? Perhaps even working on OSS projects
Some of them probably did. But most probably did not, expecially not with OSS projects. Due to the fact that everything in OSS is available free, the only way fo OSS projects to make money is off of support contracts, which less than 10% of customers actually buy. so OSS projects operate on very limited budgets, and can’t afford to hire very many full time developers. (And most open source project constantly operate at a loss as well).
> Companies that cannot compete in a free market don’t deserve to exist.
A free market does not equal free software.
> It’s amazing that you cannot even name one actual company
> or one actual product that lost out because of Linux.
Apparntly you haven’t been paying attention, because I have named several companies: Sun, Borland, SCO, etc.
> Even if they did switch to OSS solutions they have support
> contracts with Redhat or Novell.
Wrong. As I said, less than 10% of customers who use OSS software actually buy support contracts. So lets do a little math here:
If you sell a support contract for $1,000 a year, you would need to sell 1,000 support licenses to bring in $1 million a year. and that would mean you would have to have approximately 10,000 companies using your software. And $1 million a year doesn’t go very far when running a real business. How many support personel or developers can you pay on $1 million a year? After you take out operating expenses? Not a whole lot really.
And as far as Redhat and Novell, keep in mind that Red Hat is barely profitable, and the vast majority of profit it does have comes from investment income. NOT operational revenue.
Novell’s Linux operation is operating at a loss. Novell has never been able to make their Linux operation profitable, and I would very surprised if they don’t sell it off, resulting in yet more layoffs from Novell.
Apparntly you haven’t been paying attention, because I have named several companies: Sun, Borland, SCO, etc.
Borland maybe. Sun and SCO…are you serious? Do you honestly think either one of them lost business to unsupported software? Again you produce no numbers because If you could you would realize you are completely wrong.
My point (which still stands) is that the number is tiny in comparison with the number of developers currently employed. I appreciate your baseless contention, however. My estimate, at least, had some basis in reality, and was not pure conjecture.
How has your contention that the number of FOSS coders is twice the number of GSoC-coders, any more basis in reality than my contention that that is bloody unlikely?
Note that I didn’t disagree with the rest of your post, but assuming approximately one half of an entire population participates in a program that has a strictly limited amount of enrollment places and a strict set of requirements (both formal and contextual), is, if you can forgive me for saying so, ludicrous.
Except if you are one of those computer science student Linux hackers, not only are you screwing commercial companies out of wealth. But you are potentially screwing yourself out of wealth as well when you can’t get a job after you graduate because you have made the products you produce into a free commodity with no monetary value.
OSS doesn’t make software into a free commodity with no monetary value. It takes software that is already commoditized and makes it free. This was something that happened long before OSS came into vogue. Think about web-browsers. Long before Mozilla, nobody was willing to pay for web browsers, because they’d become commodity software. Who pays for multimedia playing software, or instant messaging software, or e-mail software? Almost nobody — these things are commoditized. Yes, this commoditization causes the loss of jobs, but its the commoditization that does that, not the open-sourcing.
Beyond that, you’re ignoring the bigger statistic I presented. Most programmers do not write software to sell to others. They write software to solve specific problems. Such software is inherently not commoditizable. Also, you’re ignoring the economic benefit other people gain from commoditized software. There are more users of commodity software than there are developers of it. The economic loss from the loss of development jobs can thus be outweighed by the economic benefit from the lower-cost software. Think about something like GCC. GCC put a lot of compiler writers out of business. However, its probably been a net economic gain, because it has enabled a huge number of people, especially in the embedded world, to save money by not ponying up for expensive compilers.
Fundementally, what OSS does is chance the makeup of the software market. It makes building commodity software less profitable, but at the same time, it makes building domain-specific software more profitable. The only programmers who should worry about their jobs are those stuck in the “shrink-wrap software” model. Those programmers with skill in solving domain-specific problems will always have a job, as long as people continue to have domain-specific problems!
No it isn’t. Cheap products don’t mean very much if people don’t have jobs and can’t afford to buy them anyway, and taxes skyrocket to support a strained welfare system because of increasing unemployment.
I’m not going to have this argument with you. You clearly are not familiar with the economic theories around outsourcing. Don’t make ad-hoc arguments about a subject that can be analyzed more precisely.
Edited 2006-05-08 22:35
> OSS doesn’t make software into a free commodity with no
> monetary value. It takes software that is already commoditized
> and makes it free.
I disagree. And here is why:
* Operating systems were not free until OSS
* Web servers were not free until OSS
* Databases were not free until OSS
* Office suites were not free until OSS
* Compilers were not free until OSS
* IDEs were not free until OSS
That’s just a small list.
> Long before Mozilla, nobody was willing to pay for web
> browsers, because they’d become commodity software.
If the fact that no one wants to pay for it is what determines whether something should be free or not, than money should be outlawed and everything should be free.
> I’m not going to have this argument with you. You clearly are
> not familiar with the economic theories around outsourcing.
I am very familar with the economic theories. But the fact that you claim I am not and refuse to have the argument suggests you probably know it is one you will lose.. Studies have show the first person who accuses the other one of “clearly not understanding” is usually doing it out of defense to cover up their own lack of understanding.
And I am not ignoring your “bigger statistic” The problem is that the jobs being produced by those openings are entry level “code monkey” jobs. Like database programming and such.
This is more correct regarding world economy and less regarding US economy. Largest computer system producers are US companies (including Microsoft). There are very few outside of US, so having linux as most adopted system benfits the world, but for USA it’s questionable.
Rest of the world doesn’t exist for 80% American citizens, though.
SGI was a one-trick pony, and long before Linux became an issue, they ran out of tricks. SGI was about Graphics. Nvidia had more to do with SGI going bankrupt than Linux.
SGI has been having financial problems for a while now but I never though it was going to end with a Chapter 11. Also at this moment I really don’t see how they are supposed to come out of the hole as there really isn’t any perspective for them. While Kmart is valuable because of their stores and the sheer number of locations that they have this hardly applies to a thechnology company. In the tech world filing for Chapter 11 pretty much means closing doors. I don’t see any company that would actually benefit from taking over SGI and this is really the only way that they can come out of this mess. The only one that might be interested is Sun and right now I don’t think that Sun can handle a purchase like this one.
It’s really sad … first Transmeta stopped making hardware and now SGI files for Chapter 11 … are we really drowning in conventionalism …
Incorrect..
chapter 11 is the safe way to file for bankruptcy… It is far from closing doors. Just gives the company the opportunity to go into a period in where they can write off a bunch of debt, screw people over, and restructure themselves.. then issue new stock. Like MCI/Worldcom.
> chapter 11 is the safe way to file for bankruptcy… It is far from
> closing doors.
That’s true. But at the same time, realistcally, most companies that do file for chapter 11 never successfully emerge from it and ultimately end up gone anyway. After all, their credit is shot now. They didn’t make good on hundreds of millions of dollars in debt. None of their suppliers will want to do business with them anymore, no banks will want to lend them capital, etc.
But at the same time, realistcally, most companies that do file for chapter 11 never successfully emerge from it and ultimately end up gone anyway.
http://www.usdoj.gov/ust/eo/public_affairs/articles/docs/abi98febnu…
suggests other wise. It looks like 2/3 of companies that enter chapter 11 come out of it successfully.
SGI has sold everything off that could have made them successful in my honest opinion.
It looks like they switched to IA-64 because they knew no one would want to buy MIPS/IRIX but a company like HP may want to acquire their big IA-64 machines.
This seems to me like this has been apart of their plan since they started offering IA-64. It would be logical, sadly for investors they filed for Chp 11. Bonds in SGI will probably become worthless if a judge approves conversion from bonds to stock like they did for MCI.
Now lets see if they can make themselves attractive enough to buy.
No, SGI never sold a single Alpha based workstation or desktop system. If your magazine claimed so, then they are a worthless waste of paper. SGI at some point sold Alpha based system in the form of a line of supercomputers, namely the T3, when they adquired CRAY in the late 90s.
And no, SGI was not doing great before opting for Linux. There is a reason why they decided to move to Linux, and that was because they were pretty much dead in the water with Irix/MIPS. So let’s not rewrite history here. Linux is the least of SGI’s blunders. There are even more important screwups that led to the dimisal of this company. 1) Buying CRAY, which had overlapping product lines, the T3 was a direct competitor to their Origin line, and their vector supercomputers represented a negligible revenue stream. That was the first nail in the coffin, years later SGI had to get rid of CRAY at a substantial loss. Then 2) Mr. Belluzzo’s reign of stupidity, SGI spend even more valuable resources trying to come up with an X86 NT workstation, which was a) Incompatible with the rest of the PeeCees out there (from the propietary memory slots, to the 3.3V PCI slots which no vendor supported, to the special version of NT that they required) which meant that SGI had a system that was 2x as expensive as most NT systems out there that offered the same specs. So SGI totally misunderstood the PeeCee market, then Mr. Belluzzo thought it would be great to pretty much give the crown jewels for free to Microsoft (Anyone remembers Farhenheit?), that whole experiment alone left SGI in such a bad shape tha they had to go to 3) down a route in which they bet their future on a non-existan part from their competitor (Merced by Intel). It is a rule of common sense: never base your future on something that does not exist and that comes from one of your rivals.
All of those events happened well before SGI decided to bet on Linux. Had SGI and its management had had the slightest spec of common sense they would have stayed clear of CRAY, they would have entered the PeeCee market as a gfx board maker, they would have kept their crown jewels safe away from Microsoft, and they should have developed H1/H2 which were coming along and which had a higher probability of entering the market earlier than Merced at a substantial performance advantage. But I guess one has to leave their brains at the door when joining an MBA program….
I agree with the poster saying they should’ve entered the PC market as a gfx board maker.
Wishful thinking here, but this is how I would love things to go from here:
They start making gfx boards that centralize on exellent OpenGL performance.
Sure, they should support DirectX, but OpenGL should be the priority.
They keep their drivers open source, and make sure their cards work beautifully on OSS systems, making them an OSS enthusiasts wet dream.
IMO, this could really help both the OSS systems and SGI.
OSS systems would (best case scenario) get a push into the gaming market, OpenGL would stop beeing threatened by DirectX and SGI would rise from the ashes.
But it’s just a dream, it’ll never happen, not even close. So sad..
Edited 2006-05-08 23:46
http://youtube.com/watch?v=nZMushCnp6o
Proprietary systems are dead. If you arent Open Source or perscribing to open source you future is very bleak.
What do you see as the difference between “being Open Source” and “[prescribing] to open source”?
They were always named SGI… Silicon Graphics, Inc. And last I checked no they never sold any Alpha-based systems, especially considering there is to my knowledge no variant of IRIX that will run on such boxes.
There was an article not too long ago:
http://abcnews.go.com/Business/SiliconInsider/story?id=508399&page=…
Basically compared SGI to Microsoft. Ouch.
Ovious this was gonna happen – everybody I think subconsciously expected this when the last SGI article here on osnews was around .
What is clear again & again is that SGI has a huge base of SGI lovers as there are now 50 comments & whenever SGI is mentioned there is lots of feedback – awsum hardware
All the best to SGI & hopefully a recovery as well .
Good luck.
P.S.: Funny how emotional one can get about materalistic things when they are really good .
SGI have been screwed since about 1998 or thereabouts, it’s amazing that they have lasted so long. It was a major mistake trying to go into the Windows NT workstation market, a major waste of money which should have been put into developing Irix and their MIPS systems. The demise of the R18000 chip back in 2003 was really the final nail in the coffin for their IRIX kit that’s for sure. Great systems and all but who can justify spending 18k on a graphics workstation when the graphics system hasn’t been updated since 2002 (Tezro with Vpro12).
I’d say the issue goes further than that; like the spinning off of MIPS as a seperate company.
If they really did need to seperate it and pull in more expertise in respects to running a processor design company, they should have split the company into the two, the parent company being SGI, with the two respect companies being SGI Semiconductor and SGI Systems – each focusing on their respective strengths.
The alternative today is looking pretty bad; one could be to work with AMD, work with the Opteron and see if it can fit into the FlexNUMA achitecture, possibly ditch Linux in favour of using OpenSolaris, and rebrand it as SGI Irix 8 – buy out Trolltech and OpenSound, couple these two with KDE, rebranding it as the “Irix Desktop Environment” and start pushing out workstations and servers.
They also need to move beyond their niche and start pushing the machines into areas such as database, webservers, number crunching etc. etc.
Ultimately, however, nothing will stop SGI’s demise, because like all idiots who hold the strings – institution investors, they’ll place a old university chum in charge of the business, they’ll cut, cut, cut and cut until there is nothing left to cut, the business goes tits up, and the CEO and the likes walk away with a golden handshake with the former employees (if any remain) scatter around looking for the shatter pieces of a company which they dedicated themselves to.
It depends on whether you see the size of the economy, at a national, local or international level.
If a job is created in Bangalore for programming, who then purchases a good produced in the US, which requires two people employed, is it really so bad that the one programmers job as be lost in favour of two production jobs?
The problem is, people here think about things at the national level – screw the national level; I think Number 2 off Austin Powers put it best, ‘there is no world, there are corporations!” – those who keep parading their flag of nationalism need to head back to the 19th century, they’re calling, and they want their isolationist, nationalistic jingoisms back.
> If a job is created in Bangalore for programming, who then
> purchases a good produced in the US, which requires two
> people employed, is it really so bad that the one programmers
> job as be lost in favour of two production jobs?
Except it doesn’t work that way. “production” jobs, ie: blue collar labor, was one of the first forms of work in the US to become a victim of outsourcing. High tech jobs then followed.
define inovative then
> define inovative then
Sure. Innovative is something that is technical breakthrough. Example, Sun’s dtrace, or dynamically patchable kernels.
Innovative is NOT doing trivial things with existing technology. Example, a wiki is about the most trivial web application you can write. There is nothing remotely innovative about it. Trying to say a wiki is an innovative idea is almost as bad as Amazon’s claim that one click ordering was an innovative idea.
Edited 2006-05-09 05:06
Jesus are you cherry picking. What is so innovative about dtrace? It just seems like a mix of trace and gdb — just a twist on existing technology, as it were. As for dynamically patchable code, that’s innovative, but Sun didn’t do it. Dyanamically-patchable code has been a part of Lisp forever and a day.
You would’ve been better off mentioning Sun’s Self project as an example of innovation. Little more innovative than that has come out of Sun in years.
> What is so innovative about dtrace? It just seems
> like a mix of trace and gdb
You haven’t used dtrace have you?
> As for dynamically patchable code, that’s
> innovative, but Sun didn’t do it.
There is a *HUGE* difference between dynamically patchable code running in an interpreter, and a dynamically patchable kernel that doesn’t need a reboot. One os fairly trivial to implement. The other is not.
maybe it’s just me but it seems to me most people tend to define an innovation as a new unique idea or way of doing things, often changing (sometimes radically) the way we think about or do things.
> maybe it’s just me but it seems to me most people tend to define
> an innovation as a new unique idea or way of doing things,
> often changing (sometimes radically) the way we think about or
> do things.
My idea of innovation involves two key ideas though: “non-trivial” and “non-obvious”. Wikis fail both the non-trivial and non-obvious tests. PHP is basically a “me too” language that ripped off ASP, which was a bad idea in the first place, etc.
Any sane person would agree with you, axel.
I think if we could harness the power generated by Simba’s backpedalling, we could cut our dependence on Middle East oil
I mean, the dude’s now claiming that the Web isn’t the biggest boon to his industry since the invention of the computer. I wouldn’t take him too seriously.
> I mean, the dude’s now claiming that the Web isn’t the biggest boon
> to his industry since the invention of the computer.
I claimed no such thing. Now you are putting words in my mouth. Nor have I backpeddled at all off my original position.
Again, you are resorting to ad-hominen attacks, which basicaly proves you can’t attack my real issues, so you attack me instead.
Wait, I’m confused.
You have made two major assertions thus far:
A. Open source software is not innovative.
B. Open source software is economically detrimental and hurts devolpers.
I then demonstrated that not only is the idea of open source software responsible for the Internet and the Web (along with a huge number of applications pervasive in the software industry), but that these are largely responsible for the vast majority of development jobs today.
What, pray tell, is your position? Certainly you can’t accept my arguments and yours. They’re orthogonal and contradictory.
> I then demonstrated that not only is the idea of open source
> software responsible for the Internet and the Web
And I stated that these projects are “pseudo-open source” and red herrings thrown up to support your position. The Internet, was in fact, a colaboration of Universities working under the directive of, and funded by the DOD. That’s not exactly a traditional open source project. (In fact, the public didn’t even have access to the Internet for many years. It was restricted to use only by the government and academic institutions. That’s not exactly very “open”.)
You did not demonstrate that the Internet was a a product of open source. You claimed it was, and have completely ignored my objections to the idea that the Internet was developed under the modern notiion of open source. It simply wasn’t.
Edited 2006-05-09 06:20
What a wiggle artist!
Sure, the Internet was developed by sharing source code amongst a large number of disparate researchers, contractors and government departments, all collaborating collectively and with no joint financial motive. The source code was freely available to anybody on the network at the time. That they didn’t mail the source code on punch cards to the general public doesn’t make it closed source. However, for the sake of argument, I’ll concede the point, if only to show you how it’s done
I couldn’t help but notice how artfully you dodged the topic of the Web and its role so you could attack a tangent. The Web qualifies as both innovative and open source (both in architecture and initial implementation). How do your arguments jive with this?
Also, I appreciate how you’ve carefully shifted your position. Whereas you previously claimed (as I outlined in my last post), that open source stifles innovation and hurts developers economically, you’re now saying your position is that in some cases open source makes sense. How convenient. Once again, the two don’t work together. You can’t claim a universal rule and in the next breath add a caveat without breaking your rule.
I’m not here to argue about fanatics or politics. There are just as many zealots on the proprietary side of the debate as there are on the FOSS side (this is the part where you pick up a mirror and recognize yourself in that statement). I will agree whole-heartedly that there is a time and place for open source, just as there is for closed source development. I will agree that there are those who will boycott any proprietary software maker just as there are those who believe all FOSS developers are script kiddies and German hackers.
I disagree, however, that the cooperative FOSS community is dead and has been replaced with individuals pushing a militant agenda.
More to the point, I will continue to disagree with you that open source software is not innovative and is detrimental to the economy or developers at large. This, your original position, is just as invalid now as it was when this debate began several hours ago. No amount of wriggling is going to undo that now.
=======================================================
Edit: Stealth edit alert! Way to remove a large chunk of your response, Simba. The section in question, for posterity’s sake, is attached below:
=======================================================
> What, pray tell, is your position?
My position is that in *some* cases, open source makes sense. But that it has gotten way out of hand to the point of being loaded with fanatics. It’s practically a religion today, with some of the biggest mouthpieces of open source promoting the idea that all software should be free and open source, and that people should boycott any company that dares to not play but that rule.
It’s gotten way out of hand. It’s no longer a “community of people trying to help each other”. Instead, it has turned into “a community of people pushing a militant agenda and trying to force commercial software companies to either follow their religious ideals, or close up shop”.
Edited 2006-05-09 06:41
Sure, the Internet was developed by sharing source code amongst a large number of disparate researchers, contractors and government departments, all collaborating collectively and with no joint financial motive. The source code was freely available to anybody on the network at the time. That they didn’t mail the source code on punch cards to the general public doesn’t make it closed source. However, for the sake of argument, I’ll concede the point, if only to show you how it’s done
Um, no. The internet protocol suites, especially the BSD software that really made the “internet” as we know it now, were definitely not freely available. To get a copy of BSD, you had to have an AT&T Unix source license, which, at one point, was going for around $50,000 dollars. (although at other times, it went for $250 US if you were a university.)
The development was almost entirely funded by the Department of Defense. (We funded some of it at NASA, and various DOE related entities picked up a bit of the tab,) until the research part was done.
The protocol suites didn’t become freely available until after Berkeley and AT&T signed their consent decree that made it possible for BSD code to be released without an AT&T source license. This was long after the development was done and paid for.
“The internet” is the longest running distributed collaborative software development project, but it’s not even the first such creature, and it didn’t really succeed at what it was designed for. (If you must fail at your intended goal, cover it up by doing something else even cooler, instead.)
The first, and probably most innovative ever, open source collaboration wasn’t the internet, and wasn’t a distributed collaboration, but it set the basis for the industry for its first 10-15 years, and that would be the development of the original Fortran compiler.
GCC is the nth in a very long line of freely available language implementations, including such famous predecasors as forth’s original interpreter and the original portable Pascal compiler.
The idea of freely available source has been around since the 1950s (see, for instance, SHARE,) the only thing particularly new between 1980 and today is the GPL, and that was formulated nearly 20 years ago.
Thanks Cloudy. I learned something today
> contractors and government departments, all collaborating
> collectively and with no joint financial motive.
There was no financial motive? I guess the huge sums of research dollars they were getting from the DOD was not a financial motive? And their University paychecks were not a financial motive? Give it up. You have clearly lost this one. The Internet was NOT an open source project. It was not even available at all (source or otherwise) to people outside the government and outside of University research departments for the first several years of its life. That’s not exactly open source.
> The Web qualifies as both innovative and open source
> (both in architecture and initial implementation). How do
> your arguments jive with this?
Actually, the Web was mostly developed internally at a University as well before being “open sourced”.
> Also, I appreciate how you’ve carefully shifted your position.
> Whereas you previously claimed (as I outlined in my last post),
> that open source stifles innovation and hurts
> developers economically, you’re now saying your position is
> that in some cases open source makes sense. How convenient.
> Once again, the two don’t work together.
But they do work together, and there is no conflict here. Basically, open source makes sense for “workhorse” utilities that are largely in just a maintenance mode anyway with no real innovation being done on them–software that no company wants the financial burden of maintaining anyway. A perfect example would probable be the various mail transfer agents out there. These are programs that have been around forever, are basially not changing much except for undergoing bug fixes and security hole patches, etc. For those, it makes sense for the community to maintain them.
> More to the point, I will continue to disagree with you that
> open source software is not innovative and is detrimental to
> the economy or developers at large. This, your original position
> is just as invalid now as it was when this debate began several
> hours ago
Again, I didn’t claim that open source was detrimental to the economy in general (I claimed that outsourcing was). What I did claim, was that open source was detrimental to the future of high quality jobs in software development. And I stand by that statement. We will just have to agree to disagree on this.
And the reason for the stealth edit was simple. After I posted it, I decided I didn’t want to turn this into a “zealot flame war”, so I removed the stuff about militants and all that.
Except it doesn’t work that way. “production” jobs, ie: blue collar labor, was one of the first forms of work in the US to become a victim of outsourcing. High tech jobs then followed.
How so? I look in Australia and at home, manufacturing jobs are increasing, only difference, they’re moving higher up the value scale – rather than trying to create the $10 t-shirt to compete with China, these companies are making the $50 t-shirt which uses local creative talent to differentiate – Kia Kaha is an example of this in New Zealand with the success of Michael Campbell in golf.
The problem with the US; stupid employment laws requiring the employer to pay everything from healthcare to retirement to god knows what – that isn’t the roll of the business! doesn’t anyone in the US realise that! I mean, if you are IBM, and you have to assemble a server, are you going to assemble it in New Zealand or Malaysia (GDP per capita are around the same) where all they have to pay is the employees wage, or are they going to setup shop in the US where they’re expected to not only pay the employee, but pay for their health, retirement plus the possibility of getting sued into the ground because no one has the guts on capital hill to pass the relevant legislation to get rid of suing in the workplace.
Edited 2006-05-09 05:44
> The problem with the US; stupid employment laws requiring
> the employer to pay everything from healthcare to retirement to
> god knows what – that isn’t the roll of the business!
Um… Those those laws are in place to prevent companies from practicing worker exploitation–something that is way too common in some of the places where manufacturing jobs are being outsourced to. Sure we could get rid of those regulations And then clothing companies could make their clothes here for dirt cheap… After all, we could allow companies to exploit child labor, and ignore safety requirements, and make workers work 12 hour shifts 7 days a week, and not pay any benefits, and not have to pay up when the fact that they ignored safety requirements results in workers being injured, or worse, and unable to provide for their families, and that kind of thing.
Edited 2006-05-09 05:50
Healthcare and such should be the government’s concern.
Even when it’s the government’s concern, as it is in the UK, the government still takes part (actually most) of the healthcare contribution from the employer, paid out of the gross amount you get paid.
> Healthcare and such should be the government’s concern.
Ok. Multiple question quiz time.
Government health care for 300 million people will:
A: Bankrupt the federal budget
B: Raise taxes to 70% or 80% of a person’s income.
C: Result in lower quality health care and long wait times to be seen
D: All of the above
E) None of the above.
1) Americans spend $5300 per capita on healthcare, more than any other country on earth.
2) Our key healthcare indicators are mediocre at best, and in our inner cities, cease to be competitive with those of other developed nations entirely.
3) Countries like the UK run decent nationalized health care systems, without bankrupting the budget. The UK’s 2005 budget deficit was ~3.2% of GDP. Our 2005 budget deficit was 20% of GDP. They also accomplish this without rising tax rates to “70% to 80%” of a person’s income. The middle tax rate in the UK (on up to ~$60,000 of income) is 25%, 3% higher than our middle tax rate (on top to $70,000 of income). Their top tax rate is 40%, about 5% higher than our top tax rate. Lot’s of countries manage to have national healthcare without France-like tax rates!
Right now, Americans are spending more and getting less for their healthcare dollars. Meanwhile, the Europeans have at least shown workable nationalized alternatives that don’t bust the budget while offering better healthcare. Yet, Americans are for some reason deathly afraid of even talking about a nationalized healthcare service! We sure as hell couldn’t do a lot worse than we’re doing now!
> 1) Americans spend $5300 per capita on healthcare,
> more than any other country on earth.
And guess where the vast majority of new presscription drugs come from for fighting disease? Hint: It’s not the UK. It’s something called “American drug companies” And guess what? That research costs money.
The problem with the US; stupid employment laws requiring the employer to pay everything from healthcare to retirement to god knows what
This is a common misunderstanding. US employers are not required by law to pay for healthcare or retirement benefits. Rather, the whole idea of “benefits” came out of an attempt during WW-II by US employers to solve a chronic labor problem. They weren’t allowed to raise wages, as those were controlled by the government as part of the war effort. But there was nothing in the rules to prevent them from providing ‘benefits’ instead.
Meanwhile, Hennry Kaiser, a California ship builder, discovered that his plants had less absenteeism if he spent some money helping keeping his employees fit, so he invented what became known as the Health Maintenance Organization (HMO) and provided it as a benefit.
At first, HMOs paid for themselves in terms of reduced absenteeism and pension benefits got around the wage freezes.
Then the war ended but the customs stayed in place.
This was further exacerbated by the US refusing to admit it was a socialist country and not putting decent safety nets in place, so many people came to rely on the pension and health benefit systems — and the unions came to demand increases whenever they were in good negotiating positions.
Um… Those those laws are in place to prevent companies from practicing worker exploitation–something that is way too common in some of the places where manufacturing jobs are being outsourced to. Sure we could get rid of those regulations And then clothing companies could make their clothes here for dirt cheap… After all, we could allow companies to exploit child labor, and ignore safety requirements, and make workers work 12 hour shifts 7 days a week, and not pay any benefits, and not have to pay up when the fact that they ignored safety requirements results in workers being injured, or worse, and unable to provide for their families, and that kind of thing.
Same protection here, but the difference, we have ACC, there is no suing, and we don’t have the same level of costs which businesses would inccur setting up in the US.
I’ll tell you what the average manufacturing worker gets here, $10 per hour, the average work week (when I worked in a production plant) of 50 hours, 6:00am to 4:30pm, with 1/2 lunch break. I certainly don’t call that ‘exploitation’.
“I’ll tell you what the average manufacturing worker gets here, $10 per hour, the average work week (when I worked in a production plant) of 50 hours, 6:00am to 4:30pm, with 1/2 lunch break. I certainly don’t call that ‘exploitation’.”
Of course not, problem is… you have no idea what you are talking about. In the US workers have to pony up part of their health care costs, in the form of private or HMO plans, the companies pony up their part. In your beloved land down under, like other places in the world. Workers don’t have to pay for a private health plan, since they pay a higher tax rate. And thus it is the government which offers health care, at the end of the day workers have to pretty much pay up the same amount of money, let it be in the form of a private health provider or the higher taxation in a country with universal health care. Now this is the kicker, and obviously another proof that you have no much experience with the real world besides whatever it is that you are exposed to inside your parent’s basement. In the US the company has to contribute to the worker’s health care fund and retirement fund, many times in the form of managed investment plans (which means the company gets to keep even more of the money they actually owe their workers). In other places companies have to pay/contribute to the social healthcare and retirement structure in the form of again taxation and social security funds. In the end, it is actually the US which has laxer laws regarding what companies get to do with their money. Iti is just unmitigated greed that fuels the outsourcing of jobs. And believe me, those jobs are not going anywhere in Australia nor New Zealand, which have had their share of manufacturing stagnation.