Yes, I left Pixar after 12 years and now work for myself. It seems to me that grown-ups do not work for other people, but perhaps I am being too arrogant.
Yes, I'm happy and I like myself. There is a lot of stuff in the world to be unhappy about, and I get to spend time fighting it, too. I pretty much get to do what I want now, since I now own my company. And I get to do things that matter. This is how I want to be.
Look at the people involved in Open Source development projects when you are hiring. You will find just the sort of person that you are looking for at work. You will be able to read how they interact with others, at least using electronic communication. You will get an idea of the quality of their code. You will probably be able to choose what project to look at by the expertise you are looking for.
In 1981, myself and half a dozen other people got jobs as computer disk operators in what was then the worlds foremost laboratory developing computer graphic animation. It was the predecessor of Pixar - indeed, the Pixar founders had just left there to go to work for George Lucas.
We made $2.15 per hour. Operators would wait for someone to call to have a disk changed, and we would mechanically change the disk and run the predecessors of fsck (icheck, ncheck, dcheck, etc.). The systems ran Version 6 Unix, and there were two VAXes which ran 4BSD.
There were several ways to entertain yourself while waiting for a call to do something. You could do your homework. You could try to learn more about Unix and C, which they didn't teach in the college because they had little practical application at the time in the eyes of the CS department. Or you could play rogue, an interactive computer game that ran with really primitive graphics on the VT100 terminal.
Most of my colleauges chose to play rogue. I read the Unix and C documentation, which was only one book and about a foot of papers at the time. I had some computer programming experience, including assembler, but no formal classes in programming, as I was a communication arts student. But once I had read all of the available literature on Unix and C, I was able to get a job as an assistant systems programmer and start moving up in the lab. That eventually got me to Pixar.
The folks who played rogue? They did OK, I guess. But I think they would have done much better if they'd taken the opportunity as seriously as I did.
Look around. There is probably something to do that would be much more important, and eventually more fun, than the game-playing. You only get one life. Start living it.
It's useful in situations where - for example - you've a net cafe and you don't want people to be able to overwrite or read the machine.
That's not what DRM is for, and DRM is not usable for this purpose. You are thinking about the privilege mechanism, which has been in Linux from the earliest days.
consider a DRM where it allowed everything but logged a history of the file.
It's not possible to both allow everything and enforce logging.
All of the lossless compression algorithms assume that the data has large amounts of exact duplication. Runcoding assumes a large number of the same pixel. LZW assumes a large number of the same sequence of pixels, rather than the same pixel.
High-quality photographic data has neither. And any attempt to make its spectrum look less like noise will discard real data. So, lossy compression is really the only form of compression that can be applied to it.
This is so different from Earthviewer. Given the reflections and shadows in a high-quality scene, you don't necessarily get to discard any detail from the render.
Bruce
Distributed rendering can be compute-bound
on
Rendering Shrek@Home?
·
· Score: 4, Informative
At Pixar, distributed rendering, even within the same building, was sometimes I/O bound rather than compute-bound. The problem is that high-quality rendering requires a lot of textures, some of which are algorithmic (a function in shading language) and some photographic (a big image file, probably not compressed because compression would add artifacts). So, you have to ship a lot of files around to render a scene. Photographic textures may be large because of the scale problem. They may have to look right through a 100:1 zoom, as objects move and the camera viewpoint changes.
This is not just tracing all of the rays in the scene.
The only people I have caused to be fired were at Known Safe and Linux Capital Group. Known Safe's CEO had a severe anger-management problem and poisoned his own company. Linux Capital Group was a bubble casualty, and the ex-employees and cash stockholders have just recently gotten a nice present from the founders.
If you read my recent press (use news.google.com) you'll notice that I am doing anything but sell indemnity. The Robert Macmillan article especially makes that point.
Regarding the FUD messages, we have gone very far out of our way not to amplify FUD and I seriously doubt that hearing of OSRM is turning any customer away.
But there is something you can tell your customers. If we don't do more about software patents, especially patents in standards, a few years from now Linux won't be Free Software any longer. You'll have to buy a patent-licensed version at a steep mark-up from one of the commercial distributions. And the commercial distributions have been doing hardly anything to help is with the patent fight.
Unfortunately, donating to FSF doesn't do much about legal defense. FSF has one legal counsel, who works pro-bono and happens to be off writing a book and other stuff this year so his availability to FSF is extremely limited. They have never built a legal team. And they have been entirely innefective regarding software patents for years, They called their anti-patent effort LPF - League for Programming Freedom, and it has been unstaffed for most of a decade.
And unfortunately, the software companies that we compete with do view us as a consortium of large businesses out to kill them. They and their legal teams will treat us that way.
I think you aren't giving credit for a number of things that are out there for you to see.
First, go to the OSRM news page and read all of those SCO articles. You will notice the common theme is that we say that SCO doesn't have a case. If we wanted to capitalize on FUD, we would say "maybe", not "no way" about their case.
Second, you really should give me some credit for the years that I have spent attempting to reform just the laws you are talking about, an effort that I continue with OSRM's support. And notice that OSRM is funding work by PJ of Groklaw as well, who certainly is carrying on the fight you are calling for as well.
In my CNET editorial, which will run sometime this week, I hope, I call it "Meteor insurance" rather than "Tiger Insurance", but I make the same point as you, and then explain who really is at risk and why.
Frivolous and unfounded lawsuits are not ABNORMAL, unfortunately. Any large business has learned that, and now it's our turn. SCO will not be the last we face.
Don't you think it's a good idea for us to be prepared? I think a permanent legal team to help defend us, and a revneue stream to support it, are no small thing.
Why am I liable for the misdeeds of another person?
Well, you aren't. But proving that could be expensive. The problem is that the patent statute says that you can be prosecuted for various forms of infringement, including use. If you got to court, your first action would probably be to attempt to sever yourself from the case, for just the reasons you state. And it might work. But we need clearer law here, or at least good case law.
I know that there are some cases I could win as a defendant, but I'd have to spend all I have to get there, and wouldn't get it back. Is that really winning?
I have been thinking about non-Open-Source risk management as you suggest. But there are many entities that can claim expertise on "Software". We can offer a specialized expertise. Also, having focus is a very good thing for a start-up company. If you try to do everything, you fail at doing anything.
To capitalize on SCO's claims, someone would have to believe them. If you look at any survey of the IT industry, you know that's not the case. And you should also know that I have worked very hard to knock down SCO and continue to do so. But it is the nature of Free Software that we do gain something from our enemies. If Microsoft hadn't been saying all of those bad things about us ("unamerican", forsooth), people in business wouldn't have been curious about what was scaring Microsoft, and would not have looked as closely at what we had to offer as they did.
The best thing you can do when thinking about OSRM is to assume that SCO is gone (soon enough that'll be true) and consider what our role is when facing patent claims. There will be enough of them.
but instead some legal assistance in the case there is an IP problem with open source software. Is this correct? If so, why go that route? It seems to offer little tangible benefit other than the "FUD Protection" angle.
It lets us establish a permanent legal team who work on a number of similar claims against Open Source. They'll be up to date on their research, etc. That sort of efficiency will save everyone money.
Why not just pro-actively sell the legal assistance -- for example if a company wants to use OSS Project X, you would perform research and certify that the IP in Project X is "clean" and the project is freely redistributable.
We're doing that too, but if you didn't know: you can never finish a patent search definitively. Regarding the copyright issue you point out, that is easier to deal with.
Currently we have to work with insurance companies. We can't offer it ourselves yet.
What if there was a company that said it was using BeOS, and really wasn't, but just wanted people to think so until they were ready to release their product?
Just pretend that SCO doesn't exist, because there isn't a chance that they will prevail. Then consider what can happen with software patents.
Self-promotion? This interview was arranged by the company's PR firm, so sure. Macmillan calls me every week for something else, and this is the first time he's had a PR firm ask me to call him since publicity for my books at LinuxWorld NY. But I'd not get on the phone if I didn't think I had something interesting to say.
SCO is going down quickly and there is no risk that they will win in a way that would cause OSRM to have to pay someone. But SCO's not really OSRM's business, anyway. Think about software patents and the future.
Does your particular religion really posit that the entire book is directly authored by God? And not just (for example) the 10 commandments and some other pieces? And what is your stance on translations?
homosexuality was not involved , I think you're getting your bible stories mixed up
The story of Onan is about Levirate marriage (look it up) but has been mis-interpreted as if it is an objection to any release of semen in a context that would not make a woman pregnant. And that's why it's been used against homosexuals.
Slashdot Mirror
Yes, I left Pixar after 12 years and now work for myself. It seems to me that grown-ups do not work for other people, but perhaps I am being too arrogant.
Bruce
Thanks
Bruce
Bruce
We made $2.15 per hour. Operators would wait for someone to call to have a disk changed, and we would mechanically change the disk and run the predecessors of fsck (icheck, ncheck, dcheck, etc.). The systems ran Version 6 Unix, and there were two VAXes which ran 4BSD.
There were several ways to entertain yourself while waiting for a call to do something. You could do your homework. You could try to learn more about Unix and C, which they didn't teach in the college because they had little practical application at the time in the eyes of the CS department. Or you could play rogue, an interactive computer game that ran with really primitive graphics on the VT100 terminal.
Most of my colleauges chose to play rogue. I read the Unix and C documentation, which was only one book and about a foot of papers at the time. I had some computer programming experience, including assembler, but no formal classes in programming, as I was a communication arts student. But once I had read all of the available literature on Unix and C, I was able to get a job as an assistant systems programmer and start moving up in the lab. That eventually got me to Pixar.
The folks who played rogue? They did OK, I guess. But I think they would have done much better if they'd taken the opportunity as seriously as I did.
Look around. There is probably something to do that would be much more important, and eventually more fun, than the game-playing. You only get one life. Start living it.
Bruce
That's not what DRM is for, and DRM is not usable for this purpose. You are thinking about the privilege mechanism, which has been in Linux from the earliest days.
consider a DRM where it allowed everything but logged a history of the file.
It's not possible to both allow everything and enforce logging.
Bruce
High-quality photographic data has neither. And any attempt to make its spectrum look less like noise will discard real data. So, lossy compression is really the only form of compression that can be applied to it.
Bruce
Bruce
This is not just tracing all of the rays in the scene.
Bruce
At Pixar, distributed rendering, even within the same building, was sometimes I/O bound rather than compute
Anyone else, do not assume I got them fired.
Bruce
Regarding the FUD messages, we have gone very far out of our way not to amplify FUD and I seriously doubt that hearing of OSRM is turning any customer away.
But there is something you can tell your customers. If we don't do more about software patents, especially patents in standards, a few years from now Linux won't be Free Software any longer. You'll have to buy a patent-licensed version at a steep mark-up from one of the commercial distributions. And the commercial distributions have been doing hardly anything to help is with the patent fight.
Unfortunately, donating to FSF doesn't do much about legal defense. FSF has one legal counsel, who works pro-bono and happens to be off writing a book and other stuff this year so his availability to FSF is extremely limited. They have never built a legal team. And they have been entirely innefective regarding software patents for years, They called their anti-patent effort LPF - League for Programming Freedom, and it has been unstaffed for most of a decade.
And unfortunately, the software companies that we compete with do view us as a consortium of large businesses out to kill them. They and their legal teams will treat us that way.
Bruce
First, go to the OSRM news page and read all of those SCO articles. You will notice the common theme is that we say that SCO doesn't have a case. If we wanted to capitalize on FUD, we would say "maybe", not "no way" about their case.
Second, you really should give me some credit for the years that I have spent attempting to reform just the laws you are talking about, an effort that I continue with OSRM's support. And notice that OSRM is funding work by PJ of Groklaw as well, who certainly is carrying on the fight you are calling for as well.
In my CNET editorial, which will run sometime this week, I hope, I call it "Meteor insurance" rather than "Tiger Insurance", but I make the same point as you, and then explain who really is at risk and why.
Frivolous and unfounded lawsuits are not ABNORMAL, unfortunately. Any large business has learned that, and now it's our turn. SCO will not be the last we face.
Don't you think it's a good idea for us to be prepared? I think a permanent legal team to help defend us, and a revneue stream to support it, are no small thing.
Bruce
Well, you aren't. But proving that could be expensive. The problem is that the patent statute says that you can be prosecuted for various forms of infringement, including use. If you got to court, your first action would probably be to attempt to sever yourself from the case, for just the reasons you state. And it might work. But we need clearer law here, or at least good case law.
I know that there are some cases I could win as a defendant, but I'd have to spend all I have to get there, and wouldn't get it back. Is that really winning?
I have been thinking about non-Open-Source risk management as you suggest. But there are many entities that can claim expertise on "Software". We can offer a specialized expertise. Also, having focus is a very good thing for a start-up company. If you try to do everything, you fail at doing anything.
To capitalize on SCO's claims, someone would have to believe them. If you look at any survey of the IT industry, you know that's not the case. And you should also know that I have worked very hard to knock down SCO and continue to do so. But it is the nature of Free Software that we do gain something from our enemies. If Microsoft hadn't been saying all of those bad things about us ("unamerican", forsooth), people in business wouldn't have been curious about what was scaring Microsoft, and would not have looked as closely at what we had to offer as they did.
The best thing you can do when thinking about OSRM is to assume that SCO is gone (soon enough that'll be true) and consider what our role is when facing patent claims. There will be enough of them.
Bruce
It lets us establish a permanent legal team who work on a number of similar claims against Open Source. They'll be up to date on their research, etc. That sort of efficiency will save everyone money.
Why not just pro-actively sell the legal assistance -- for example if a company wants to use OSS Project X, you would perform research and certify that the IP in Project X is "clean" and the project is freely redistributable.
We're doing that too, but if you didn't know: you can never finish a patent search definitively. Regarding the copyright issue you point out, that is easier to deal with.
Currently we have to work with insurance companies. We can't offer it ourselves yet.
Bruce
Bruce
Bruce
What if there was a company that said it was using BeOS, and really wasn't, but just wanted people to think so until they were ready to release their product?
Bruce
This is not about SCO.
Just pretend that SCO doesn't exist, because there isn't a chance that they will prevail. Then consider what can happen with software patents.
Self-promotion? This interview was arranged by the company's PR firm, so sure. Macmillan calls me every week for something else, and this is the first time he's had a PR firm ask me to call him since publicity for my books at LinuxWorld NY. But I'd not get on the phone if I didn't think I had something interesting to say.
Bruce
Bruce
And I'd answer, if anyone actually asked a question.
Who are you, anyway? I probably didn't like you either.
You goofed. The numbers should start with 0.
Bruce
Thanks
Bruce
The story of Onan is about Levirate marriage (look it up) but has been mis-interpreted as if it is an objection to any release of semen in a context that would not make a woman pregnant. And that's why it's been used against homosexuals.
Bruce