Linuxs initial target was desktops and smaller systems. Conventional wisdom is that Intel procs before the Pentium Pro could not do SMP, but that is an over simplification. They were the first that could do SMP easily. At least as far back as 1981 people were making "Supercomputers" out of Intel CPUs. http://www.computerhistory.org/VirtualVisibleStora ge/artifact_main.php?tax_id=03.04.07.00
Linus thinks working on small, embedded, systems is perhaps more important then massive scaleability with SMP. That said: SGI has been producing massivly SMPd machines that run Linux for a long time. Prob with MIPS first, definitly with IA32 now, and with a heavily hacked kernel, but still Linux.
I think Suns strategy of "ultraslows, but many" is much older then their "big iron" history. At least as far back as 32bit Sparcs, desktops, not to mention "workgroup" servers, were SMP capable. I dont know if it is realy a strategy so much as a necessity given RISC style processors. Of course, extending that argument, IA32/64 systems, compared to 370 arch IBM Mainframes are RISC too. And to scale up IA systems, you go SMP. Comparing Sun and Intel micro computers to a mainframe might be compleatly useless though:)
Imagine how much Sun will charge for the same IDE hard drive the corner store charges $99 for, once blesed by Sun, and put in 'format's database. I perdict it will list at $749.
Yes, MS today is very much like IBM used to be. Except for one major difference: IBM stuff works, and did work back then. Why do you think we have the phrase "Mainframe level stability/uptime/reliability"?
Air Canada and US Airways are both members of the Star Alliance. But American is not. I think some level of scheduling coordination is the primary reason for Star to exist, but that doesnt explain why American was smashed as well.
"UNIX" means different things to different people. One definition would be something that contains ATT UNIX code. Another would be something that has a bunch of certifications. Linux has neither.
But, the BSDs, and I believe even Solaris and AIX have a Linux compatability layer, or at the very least, "the GNU toolset", GCC, glibc, etc. Of course you wont beable to run IA32 binaries on a UltraSPARC, regardless of the compatability layer, but you could run IA32/Linux stuff on IA32/*BSD, or SparcLinux stuff on a Solaris box.
I guess Im trying to say, given that lots of things can run Linux binaries, can cleanly compile Linux targeted sources, Linux + GCC + glibc may be a better standard to target then POSIX and whatnot. It is definitly more modern. Or to put things another way, UNIX is irrelevent, the question shoud be: can UNIXes ever be Linux?
Having a 512mb card will allow for higher resolutions? WTF?
Back in the day before video cards had texture memory, resolution was directly (root-icly?) related to your video memory. So a 640x480x16(bit colour) required 614400bytes of memory - a 1mb card. 1920x1200x32 would require 9216000bytes, or a 9mb card. This is the "frame buffer" - something writes to that buffer, and something else (a DAC) converts it to video. Some video cards had (have?) multiple frame buffers to allow for quickly switching, well, frames. There are other factors limiting resolution, of course - DAC quality, for one. But as most "frame buffer" only cards had upgradable memory, it is the memory that is the practical limit.
Resolution is, well, fairly down the list of important qualites of a video card. It took me a while to dig around to get that resolution size (the largest that a Quadro FX 4000 will put out). Ever since about the time that AGP came out resolutions stoped being highlighted.
I dont know how much "frame buffer" memory modern video cards have. I dont even know if it is just a segment of the main memory. Anyway, since a 256mb card has about 28 times the memory then is strictly necessary for driving a 1920x1200 screen, jumping to 512mb wont help at all.
This is not to say that when 512mb cards come out they wont have higher resolutions. But it wont be because of the memory. Other technologies would also increase. Or, more likely, existing but expensive technologies would be put on these new expensive cards.
That you replied with a formal math proof proves my point. We are not talking about formal math. We are talking about what normal people think of when you say "measurable".
You have a problem with different fields using "measurable" differently. In non-measurable sets, most non-set theorists would call that non-countable (which may or may not be a different theory. I remember "non-countable' from my discret math days, but never non-measurable. Beh). Many people would say that infinity is sufficent measurement (which is why I perfer countability to measurability).
So far as Gib is concerned, "very big" is a measurement. If you have 3 sets of infinite size, saying that they have sizes of infinity, infinity-1 and infinity-2 (if you get them in the right order) is better then nothing. Such a definition would outrage any math geek. But now you have some metric
Yes, Cingular pays wholesale prices for phones, but the wholesale price that Cingular pays, and the wholesale price that Bob Cell Phone shack in the mall pays are radicly different.
The standard post-apocalyptic story involves humanity loosing vast amounts of knowledge. In the Matrix they don't even know what year it is.
I believe that the US Patent Office stores all their records in an old mine, or otherwise excavated area. They seem to be somewhat concerned about keeping knowledge around though some sort of massive catastrophe.
True. And to add to the story, what they released as OSS diddnt even compile. The was a bunch of necesary 3rd party code in it that was removed. And they (Netscap) / we (outside developers (but not me)) spent about a year trying to work with the old code before they/we gave up.
But the question is not if the current Mozilla codebase contains any Netscape <=4.x code in it, but if the release fo the 4.x code spawned a community that is still active. And that is a resounding YES.
The thing about glibc, gcc, the Linux kernel, etc, is that you can take a course, or read a book, on libc, compiler writing, Unix internals. In theory, any upper year CS undergrad should be able to contribute to those projects. But there is no manual on writing a web browser. It is hard because it is new. The Mozilla codebase is huge. Contributing to the project thus requires a steep learning curve, and implicitly, highly motivated coders. A lot of that motivation comes from "because it is Netscape" and now "because it is Mozilla", beyond just "its an OSS web browser".
Most EULAs specificly talk about being valid for a single computer, the GPL does not. One legal interpertation of the GPL for "internal" uses would be that the entire company is using the software. Bob in accounting is not the licensee, MegaGlobalCorp is. Once Bob steps into MegaGlobalCorp property he, in this context, ceses to exist as an individual. Sally in HR is not going to demand for the source code changes that Bob has made, because within MegaGlobalCorp, they are the same thing. Neither Bob nor Sally has agreed to the GPL, and any work Bob does at work is MegaGlobalCorps work, not his.
I beleive that this paticular fighter pilot, while active duty Air National Guard, was once a Top Gun pilot. That is, a fighter pilot trained to protect his multibillion dollar carrier and screening force, before all other considerations. Shoot first, ask questions later.
This kind of incident is exactly why USMC pilots (indeed, all USMC officers save for MDs and chaplins) spend at least one year as a platoon leader.
They may be scared of it, but they do SFA when it happens.
Proof: "Friendly fire" incident in Afghanistan where Canadian soldiers were killed by the hands of a US Pilot. I put "friendly fire" in quotes because to me, it doesn't even qualify as that. Im prepared to accept combat deaths, even friendly fire ones. Anyway: the Canadians were on a known live fire range (and the Americans knew, even if that information didnt make it to the pilot and air controlers). The pilot reported fire, and was specificly told to hold fire. The fire was coming from a range well known to be used by friendly ground forces, including US Forces. He fired anyway, killing 4, wounding a dozen others. He claimed he was "under attack", but how a machine gun could threatin a F18 at 20,000ft, I dont know.
Final end result: loss of 1/2 months pay for 2 months, letter of reprimand. In other words: exactly nothing.
If the US DoD were concerned about friendly fire there would have been a minimum of 3 people, if not incarcerated, then discharged from the service. The pilot, the person (or persons) who didn't provide sufficient briefings, and the commanders for letting that happen.
Password "hashes" are just that. Special types of hashes known as cryptographic hashes, but hashes none the less. They are not encryption in the traditionl sense of the word.
It is, in a sense, possible to "break" a crypyographic hash. You simply put all possible inputs into the hash function and record their outputs. Indeed, a service that can return the input for a given crypt()ed output was recently featured on slashdot. It only worked over a limited range of possible inputs, but over the very likely range of inputs (alphanumeric, <10 characters, IIRC). Dictionaly attacks on password files are common, both by sysadmins and crackers alike. It is only a matter of determination (and storage capacity) to produce such a table for all possible inputs to crypt(), and its modern, stronger, friends. Now, you dont necessaraly have the "correct" input returned when you do such a search, but you get a correct enough input.. And for cryptographic hashes a correct enough input is correct.
Interesting question: how long would it take for Distributed Net to search the entire hashspace of MD5, SHA-1?
Re:I can't make out exactly what you are looking f
on
GPS and Portability?
·
· Score: 1
As some one else has posted, there are a number of GPSs with USB interfaces.
But RS-232 should give up and die. If you have exactly two devices that are 232 and you connect them once and leave it alone it is great. But given how lax the spec is - or perhaps how lax all the implementaions are - if you have any more then two devices, you're screwed. You need an exponental number of cables and/or a drawer full of gender changers, null modem adapters, and some goat blood to get things working.
Even if you get things connected up together so they kinda-sorta work, there is no gaurentee that they do actually work. Hardware vs. Software flow control. 7 bit. 8 bit. Stop bits. Bacon bits. Ever plug a Sun into a terminal, and then unplug the power from the terminal? Ever see a high traffic Sun server go into boot prom mode? Whose idea was it to start putting 232 over RH-45 terminated cables? Everyones. You think they would have gotten together and made the problem better. But no, every manufacturer has there own idea how they should work, many with multiple incompatable ideas from the same company! Thus there are even more possibilities with pinouts there then with D9/D25. Please make it stop!
The world of RS232 is so compleatly fucked that it is one technolgy that I am very much waiting to see die off.
Simple Mail Transfer Protocol. To subit mail you should use Message Submission (RFC 2476). MSA is more or less SMTP that is both less and more restrictive. The grandparents point is that you, a client, should not be using SMTP at all. You should not be forced to use your ISPs SMTP server... And indeed, if you point is to send outgoing mail from one of your domains, you shouldnt be using there SMTP server at all. You should MSA the mail to your server (which is SPF published as acceptable).
I dont know that that was HTMLs purpose. I keep hearing people say that, and it being "easy" may be true, but I dont know that that was its goal.
HTML is NOT easy. Writing a valid HTML documenta 20-30 lines long by hand is possible, but for any non-trivial page it is near impossible. (that there are no good HTML editors is besides the point). Any website containing more then 1 page should use some kind of automation to create HTML, be that continiously generated dynamic pages, or generating the HTML once per change. Spreading the rumor that HTML is easy to learn and that anyone can, and should, use HTML is a disservice to humanity.
If you are a non-technical person, or even a technical person whose job is not specificly HTML creation software writing, then you should not generate HTML by hand. Use some kind of CMS. Use some HTML editor. Use Docbook and "compile" HTML.
Why are Wiki's popular? Because they use a markup language that actually is easy, one that is hard to screw up.
Please, pleae, please, dont continue to spread the lie that HTML is easy enough for anyone to learn.
But having to do with names being different in different places.
This summer I got a new (uncomputer) job, and for direct deposit, I gave them my personal account information. Unfortunatly, Ive been using my computer "business" account 100% of the time for a couple of years - my personal account was closed. (hmm, would have been nice if they had told me).
My first pay stub I get has my last name spelt wrong (no supervisor, I did not spell my own name wrong on the forms you gave me. I was not the one who typed it into a computer. Perhaps that person was the moron who got it wrong.) I have some cash, so I dont check that the pay went in. Next pay I do. Invalid PIN. Talk to a person: that account was closed. Any attempts at deposits? Nothing that I can see.
Talk to work. We havent gotten a rejection back. Cant help you. (changed to my "business" account at this point)
Return to $MY_EX_BANK and bash some heads. The pass it up the chain.
Get a call: Yes, we got a deposit request on that account. Rejected it, account closed. But we sent the rejection back WITH YOUR NAME SPELT CORRECTLY. $WORK_BANK got confused and rejected the rejection. Money sitting in limbo. No klaxons going off in a machine room somewhere.
Harass work again. Yes, Im prepared to believe you that you havent gotten a rejection notice. Harass your bank. You put $100,000,000/year through them, you should be able to get some answeres. I no longer have an account with $MY_EX_BANK and they got back to me.
Work calls $MY_EX_BANK and hears that there was a rejection. Still waiting to hear back from $WORK_BANK.
ARGH!!!! Still waiting on 4 weeks of pay.
So tell me, collection of computer geeks, what kind of moron built the banking network that rejections are keyed on peoples names, not account numbers, or transaction IDs? Why did a message box not pop up on someones computer when the network got a transaction it couldnt grok?
And for fuck sake. My name is Jeff Warnica. You know how many Warnica's there are in North America? Like 1000. How many in Nova Scotia? <15. How many Jeff Warnica's? 1. Not almost one. Exactly 1. What the fuck happens to people with the name John Smith?
The FSF, though the GNU project has without a douby produced a lot of code, and lots of others have transfered their IP to them. But GPL does not necessaraly lead to GNU/FSF.
Linus himself sums it up quite nicely:
Also note that the GPL below is copyrighted by the Free Software
Foundation, but the instance of code that it refers to (the Linux
kernel) is copyrighted by me and others who actually wrote it.
The Linux kernel is no where near alone in using the GPL, but not assiginig IP to FSF. I dont want to speculate, but it is possible that Linux was the first to do so, but these days it is quite common. Of the 19 packages in my/usr/src with a license/copy file in the top level, 7 of them use the GPL. None of them are (c) FSF. Compleatly unscientific, but anyway...
To suggest that the FSF is even in the top 100, ney top 1000, owners of IP is simply foolish. I doubt they would even be in the top 25 if you limit that down to software only. Besides, how do you measure that? LoC? Market worth? Revenue? Hours invested?
Slashdot Mirror
Linus thinks working on small, embedded, systems is perhaps more important then massive scaleability with SMP. That said: SGI has been producing massivly SMPd machines that run Linux for a long time. Prob with MIPS first, definitly with IA32 now, and with a heavily hacked kernel, but still Linux.
I think Suns strategy of "ultraslows, but many" is much older then their "big iron" history. At least as far back as 32bit Sparcs, desktops, not to mention "workgroup" servers, were SMP capable. I dont know if it is realy a strategy so much as a necessity given RISC style processors. Of course, extending that argument, IA32/64 systems, compared to 370 arch IBM Mainframes are RISC too. And to scale up IA systems, you go SMP. Comparing Sun and Intel micro computers to a mainframe might be compleatly useless though :)
Imagine how much Sun will charge for the same IDE hard drive the corner store charges $99 for, once blesed by Sun, and put in 'format's database. I perdict it will list at $749.
Sun should be working on making their very expensive UltraSparc chips as fast as ultra cheep Intel and AMD processors.
Uh, this is /.. Vent here, but we dont need explaination. Explain to the NYT.
Yes, MS today is very much like IBM used to be. Except for one major difference: IBM stuff works, and did work back then. Why do you think we have the phrase "Mainframe level stability/uptime/reliability"?
Air Canada and US Airways are both members of the Star Alliance. But American is not. I think some level of scheduling coordination is the primary reason for Star to exist, but that doesnt explain why American was smashed as well.
By that logic, would CP/M not be VMS?
But, the BSDs, and I believe even Solaris and AIX have a Linux compatability layer, or at the very least, "the GNU toolset", GCC, glibc, etc. Of course you wont beable to run IA32 binaries on a UltraSPARC, regardless of the compatability layer, but you could run IA32/Linux stuff on IA32/*BSD, or SparcLinux stuff on a Solaris box.
I guess Im trying to say, given that lots of things can run Linux binaries, can cleanly compile Linux targeted sources, Linux + GCC + glibc may be a better standard to target then POSIX and whatnot. It is definitly more modern. Or to put things another way, UNIX is irrelevent, the question shoud be: can UNIXes ever be Linux?
I could be wrong, but techniclly, an Army base is not part of the state that it is in. State laws, for example, dont apply.
Back in the day before video cards had texture memory, resolution was directly (root-icly?) related to your video memory. So a 640x480x16(bit colour) required 614400bytes of memory - a 1mb card. 1920x1200x32 would require 9216000bytes, or a 9mb card. This is the "frame buffer" - something writes to that buffer, and something else (a DAC) converts it to video. Some video cards had (have?) multiple frame buffers to allow for quickly switching, well, frames. There are other factors limiting resolution, of course - DAC quality, for one. But as most "frame buffer" only cards had upgradable memory, it is the memory that is the practical limit.
Resolution is, well, fairly down the list of important qualites of a video card. It took me a while to dig around to get that resolution size (the largest that a Quadro FX 4000 will put out). Ever since about the time that AGP came out resolutions stoped being highlighted.
I dont know how much "frame buffer" memory modern video cards have. I dont even know if it is just a segment of the main memory. Anyway, since a 256mb card has about 28 times the memory then is strictly necessary for driving a 1920x1200 screen, jumping to 512mb wont help at all.
This is not to say that when 512mb cards come out they wont have higher resolutions. But it wont be because of the memory. Other technologies would also increase. Or, more likely, existing but expensive technologies would be put on these new expensive cards.
That you replied with a formal math proof proves my point. We are not talking about formal math. We are talking about what normal people think of when you say "measurable".
So far as Gib is concerned, "very big" is a measurement. If you have 3 sets of infinite size, saying that they have sizes of infinity, infinity-1 and infinity-2 (if you get them in the right order) is better then nothing. Such a definition would outrage any math geek. But now you have some metric
Yes, Cingular pays wholesale prices for phones, but the wholesale price that Cingular pays, and the wholesale price that Bob Cell Phone shack in the mall pays are radicly different.
The standard post-apocalyptic story involves humanity loosing vast amounts of knowledge. In the Matrix they don't even know what year it is. I believe that the US Patent Office stores all their records in an old mine, or otherwise excavated area. They seem to be somewhat concerned about keeping knowledge around though some sort of massive catastrophe.
True. And to add to the story, what they released as OSS diddnt even compile. The was a bunch of necesary 3rd party code in it that was removed. And they (Netscap) / we (outside developers (but not me)) spent about a year trying to work with the old code before they/we gave up.
But the question is not if the current Mozilla codebase contains any Netscape <=4.x code in it, but if the release fo the 4.x code spawned a community that is still active. And that is a resounding YES.
The thing about glibc, gcc, the Linux kernel, etc, is that you can take a course, or read a book, on libc, compiler writing, Unix internals. In theory, any upper year CS undergrad should be able to contribute to those projects. But there is no manual on writing a web browser. It is hard because it is new. The Mozilla codebase is huge. Contributing to the project thus requires a steep learning curve, and implicitly, highly motivated coders. A lot of that motivation comes from "because it is Netscape" and now "because it is Mozilla", beyond just "its an OSS web browser".
Most EULAs specificly talk about being valid for a single computer, the GPL does not. One legal interpertation of the GPL for "internal" uses would be that the entire company is using the software. Bob in accounting is not the licensee, MegaGlobalCorp is. Once Bob steps into MegaGlobalCorp property he, in this context, ceses to exist as an individual. Sally in HR is not going to demand for the source code changes that Bob has made, because within MegaGlobalCorp, they are the same thing. Neither Bob nor Sally has agreed to the GPL, and any work Bob does at work is MegaGlobalCorps work, not his.
This kind of incident is exactly why USMC pilots (indeed, all USMC officers save for MDs and chaplins) spend at least one year as a platoon leader.
Proof: "Friendly fire" incident in Afghanistan where Canadian soldiers were killed by the hands of a US Pilot. I put "friendly fire" in quotes because to me, it doesn't even qualify as that. Im prepared to accept combat deaths, even friendly fire ones. Anyway: the Canadians were on a known live fire range (and the Americans knew, even if that information didnt make it to the pilot and air controlers). The pilot reported fire, and was specificly told to hold fire. The fire was coming from a range well known to be used by friendly ground forces, including US Forces. He fired anyway, killing 4, wounding a dozen others. He claimed he was "under attack", but how a machine gun could threatin a F18 at 20,000ft, I dont know.
Final end result: loss of 1/2 months pay for 2 months, letter of reprimand. In other words: exactly nothing.
If the US DoD were concerned about friendly fire there would have been a minimum of 3 people, if not incarcerated, then discharged from the service. The pilot, the person (or persons) who didn't provide sufficient briefings, and the commanders for letting that happen.
Two things:
- Password "hashes" are just that. Special types of hashes known as cryptographic hashes, but hashes none the less. They are not encryption in the traditionl sense of the word.
- It is, in a sense, possible to "break" a crypyographic hash. You simply put all possible inputs into the hash function and record their outputs. Indeed, a service that can return the input for a given crypt()ed output was recently featured on slashdot. It only worked over a limited range of possible inputs, but over the very likely range of inputs (alphanumeric, <10 characters, IIRC). Dictionaly attacks on password files are common, both by sysadmins and crackers alike. It is only a matter of determination (and storage capacity) to produce such a table for all possible inputs to crypt(), and its modern, stronger, friends. Now, you dont necessaraly have the "correct" input returned when you do such a search, but you get a correct enough input.. And for cryptographic hashes a correct enough input is correct.
Interesting question: how long would it take for Distributed Net to search the entire hashspace of MD5, SHA-1?But RS-232 should give up and die. If you have exactly two devices that are 232 and you connect them once and leave it alone it is great. But given how lax the spec is - or perhaps how lax all the implementaions are - if you have any more then two devices, you're screwed. You need an exponental number of cables and/or a drawer full of gender changers, null modem adapters, and some goat blood to get things working.
Even if you get things connected up together so they kinda-sorta work, there is no gaurentee that they do actually work. Hardware vs. Software flow control. 7 bit. 8 bit. Stop bits. Bacon bits. Ever plug a Sun into a terminal, and then unplug the power from the terminal? Ever see a high traffic Sun server go into boot prom mode? Whose idea was it to start putting 232 over RH-45 terminated cables? Everyones. You think they would have gotten together and made the problem better. But no, every manufacturer has there own idea how they should work, many with multiple incompatable ideas from the same company! Thus there are even more possibilities with pinouts there then with D9/D25. Please make it stop!
The world of RS232 is so compleatly fucked that it is one technolgy that I am very much waiting to see die off.
Simple Mail Transfer Protocol. To subit mail you should use Message Submission (RFC 2476). MSA is more or less SMTP that is both less and more restrictive. The grandparents point is that you, a client, should not be using SMTP at all. You should not be forced to use your ISPs SMTP server... And indeed, if you point is to send outgoing mail from one of your domains, you shouldnt be using there SMTP server at all. You should MSA the mail to your server (which is SPF published as acceptable).
"Easy to learn". and "easy to learn correctly" are not the same thing.
HTML is NOT easy. Writing a valid HTML documenta 20-30 lines long by hand is possible, but for any non-trivial page it is near impossible. (that there are no good HTML editors is besides the point). Any website containing more then 1 page should use some kind of automation to create HTML, be that continiously generated dynamic pages, or generating the HTML once per change. Spreading the rumor that HTML is easy to learn and that anyone can, and should, use HTML is a disservice to humanity.
If you are a non-technical person, or even a technical person whose job is not specificly HTML creation software writing, then you should not generate HTML by hand. Use some kind of CMS. Use some HTML editor. Use Docbook and "compile" HTML.
Why are Wiki's popular? Because they use a markup language that actually is easy, one that is hard to screw up.
Please, pleae, please, dont continue to spread the lie that HTML is easy enough for anyone to learn.
This summer I got a new (uncomputer) job, and for direct deposit, I gave them my personal account information. Unfortunatly, Ive been using my computer "business" account 100% of the time for a couple of years - my personal account was closed. (hmm, would have been nice if they had told me).
My first pay stub I get has my last name spelt wrong (no supervisor, I did not spell my own name wrong on the forms you gave me. I was not the one who typed it into a computer. Perhaps that person was the moron who got it wrong.) I have some cash, so I dont check that the pay went in. Next pay I do. Invalid PIN. Talk to a person: that account was closed. Any attempts at deposits? Nothing that I can see.
Talk to work. We havent gotten a rejection back. Cant help you. (changed to my "business" account at this point)
Return to $MY_EX_BANK and bash some heads. The pass it up the chain.
Get a call: Yes, we got a deposit request on that account. Rejected it, account closed. But we sent the rejection back WITH YOUR NAME SPELT CORRECTLY. $WORK_BANK got confused and rejected the rejection. Money sitting in limbo. No klaxons going off in a machine room somewhere.
Harass work again. Yes, Im prepared to believe you that you havent gotten a rejection notice. Harass your bank. You put $100,000,000/year through them, you should be able to get some answeres. I no longer have an account with $MY_EX_BANK and they got back to me.
Work calls $MY_EX_BANK and hears that there was a rejection. Still waiting to hear back from $WORK_BANK.
ARGH!!!! Still waiting on 4 weeks of pay.
So tell me, collection of computer geeks, what kind of moron built the banking network that rejections are keyed on peoples names, not account numbers, or transaction IDs? Why did a message box not pop up on someones computer when the network got a transaction it couldnt grok?
And for fuck sake. My name is Jeff Warnica. You know how many Warnica's there are in North America? Like 1000. How many in Nova Scotia? <15. How many Jeff Warnica's? 1. Not almost one. Exactly 1. What the fuck happens to people with the name John Smith?
Linus himself sums it up quite nicely:
The Linux kernel is no where near alone in using the GPL, but not assiginig IP to FSF. I dont want to speculate, but it is possible that Linux was the first to do so, but these days it is quite common. Of the 19 packages in my /usr/src with a license/copy file in the top level, 7 of them use the GPL. None of them are (c) FSF. Compleatly unscientific, but anyway...
To suggest that the FSF is even in the top 100, ney top 1000, owners of IP is simply foolish. I doubt they would even be in the top 25 if you limit that down to software only. Besides, how do you measure that? LoC? Market worth? Revenue? Hours invested?