This reminds me of a funny situation which happened on the 7 FISL early this year.
At one of the conference rooms, while everybody was waiting for the speaker to start, suddenly out of somewhere comes a Windows boot sound.
Everybody started yelling at and criticizing the person with that laptop (why would someone use Windows at FISL?).
That until the person explained: he was using Windows just because some of his laptop's hardware was not supported, was using it just for VMWare (where Linux would be running), and in fact the Windows install didn't look like it got much use (apart from a couple of VMWare icons, it looked like a new install).
He was loudly applauded when VMWare started (booting Linux, of course).
Yes... When I read it (in the original, as I'm Brazilian) I was stunned... A politician that understands Free Software! And knows the OSD (aka DFSG)!
Well, since he seems to understand it, he must know how to speak English (well, most people here know Spanish or English as a second language, since it is a requisite in many jobs). And if he knows all those things about the OSD, he might also know you're one of the ones that created it in the first place. So write to him yourself. Tell him who you are. And ask if he might want some help. Though I can't see how you could make a difference, it's always nice to try.
One of the reviews said that the current MTRR won't work (they had to disable it). They also said AMD had contacted the MTRR maintainer (!), so the support will probably be added soon.
First, stop trolling. We all know you are the same troll that appeared in the first thread and on yesterday's article. You can disguise as an AC but you can't disguise your writing style.
"Being agressive is not evil", you say. Fine. The Linux crowd is also very agressive. When they find a non-free program they like, they try to clone it. This might piss the ones who wrote the non-free app, but then, like you said, it's capitalism...
"This pisses people off because you ruined their business." Like when some version of Windows had a fake error message when run in some non-Microsoft version of DOS? Even with the code to detect the OS hidden under several layers of anti-RE code, someone found out.
"My point is if you do not like it, write for another OS." This is exactly what we are doing. And this is exactly what Microsoft wants to prevent (remember the Halloween documents?). Also, this sentence showed you wrote before thinking; you're ranting.
After this point, you go on and on with non-clear thinking (first you say it's simple, then you say it isn't; then you rant about supporting other OSs when Free Software is the most ported kind of software (and we even have Windows versions of most of them)). Then you say that the problem with the LINUX crowd (why all caps?) is that they think that it is a new way of life/thinking (no we don't think that; we know it's older than proprietary software). Then you go on and say it's business plain and simple. Funny, where's my paycheck;-P ?
Then you say that we should search for money only. But Free Software is not market-driven, it is driven by the needs of its users. And last, you mention someone I've never heard about, but fail to provide a link.
I hope you realize your cover blew up (posting as an AC is useless to disguise you're the same guy as some other random guy when you can't disguise your style and your way of thinking) and that you should stop. You won't win without a rational argument.
I could go on and on, but I left it as an exercise to the fellow slashdotters.
When the CPU has been idle for some time, Linux (and other well-engineered OSs) issue the hlt instruction. It puts the CPU in a 'sleep' state from which it gets out in any interrupt. After returning from the interrupt, it already 'forgot' it was in an sleep state and continues working normally.
When the CPU is idle, this means no proccess has anything to do, and the only things left to do are answering to keyboard, timers, modems, sound DMA exaustion, IDE read/write completed, mouse movement, and other things like that. And they all use interrupts, so it gets out of the halt state without losing any cycles (besides, the hlt is only used after the cpu has been idle for some time, so if you are fully using the cpu no hlt will appear).
When the proccesses are idle, they are all waiting for some event (blocking reads/writes and sleeping for some specified time being the most common). All those events can only happen after a ISR (so it'll wake up after a time interrupt, or after a disk interrupt, or after a keyboard interrupt, or after a NIC interrupt, etc.). So the CPU has actually nothing useful to do until an ISR happens.
Once I wrote (under DOS, before I got Linux) a.com with only 18 hlts. When I run it, it waited for a second (timer is 18.2/sec under DOS, 100/sec under Linux/i386) and exited (ret to int 20 on psp).
So you want to categorize everything relevant to the Internet, since the first days of Eniac and maybe even some trivia about what came before (telegraph, phone systems, etc.). The problem here is: the net is so big and touches so many topics, iteracting with each other in a way that truly resembles a web.
For example, we have the phreakers' history. Sure, they shaped a lot of our current world (if it weren't for them, and from the ones who borrowed the name 'hackers' and started compromissing lots of sites' security, we would have almost no security on our systems today; no flames please).
Then we have the other 'hackers' history, this one portrayed in the now-ESR-maintained Jargon File. It also has a lot of history of its own, including the recent events (which are a lot).
If we zoom in a bit, we get (for example) the Debian's history. How it started, how it grow, the issues it faced, etc; and the Slashdot's tale, a virtual pub with its anti-M$ zealots, censorship attempts, the First Posters, MEEPT!, the gnulix_guy, and lots of other people.
Then we go back out and drift a bit. We get the tales of an awful clone of a brain-dead OS, the os called CP/M and the clone is... Well you can guess it. This is also long, we had years and years of borging and fighting, from the bad to the worst to the not-so-bad (5.0+), then (in parallel) the almost unknown old Windozes, the best-of-all Win3.1x, and the progressive bloat of the 9x family.
We also have a corner for the spammers history, entangled with the USENET and email history. We have a corner for the UNIX wars and their siblings (*BSD). We have a corner for rec.pets.cats and the crosspost flammage that went there.
All of this is significant for some groups and completely irrelevant to other ones. However, if we left them out, it would be discrimination.
I think this is way too big for just a small group of persons; we need a bazaar style. The problem is that the Internet time is a lot faster than the real world's, i.e., more happens in a specific amount of time in the net than in the Real World. This is because the Net is composed from the lifes not of just a geographically close group, but from millions of people from around the world.
We would need some way of categorizing the facts. This would be hard. Some are time spans, some are specific dates, some are 'circa 1984'. We would also need to have lots and lots of subcategories, and some way to 'zoom in' from the less specific general timeline with only the most relevant facts to a very specific timeline with all the 1000+ events in a single week or in a specific but small topic in a larger period of time (like the dpkg history since Dec 1999).
To sum it up: you are nuts. We don't have enough diskspace. We would need the space of a Boeing-747 full of DVDs. (OK, I'm exagerating a bit, but not much). If we could do it, we would get lagged faster than you can say 'Benedict'.
Sorry if I got carried away, it's late. And if I mispelled something, it was on purpose.
Did anybody read an old RFC which described exactly the same problem (bandwidth hogs) and the solution to it (fair queueing)? It basically says that the routers should act as if they had one incoming queue for each possible source IP (probably it is only one queue, but the router would act _as if_ it were many). Then it goes round-robin getting one packet from each IP and putting it on the output links. If some single source tries to send more packets than the link can handle, packets from _just this source_ will get dropped. This is a non-issue for OSs with congestion control (like Linux) since they can detect the drops and use them to adjust the outgoing speed.
This way, if a single node is sending too much data most of it will get dropped and (if it has a good stack) it will reduce its sending speed automatically. Other users and the download speed wouldn't be affected (assuming they use congestion control too).
Parallel voting for names/scores
on
Slashdot Notes
·
· Score: 2
I think it would be better this way:
1. Moderators receive moderation points as in the old system and use them in the score _only_.
2. Moderators (might be the same ones, or not) receive special pseudo-moderation points that are spent on the categories. Each category would have its "score". For example: moderator A votes +1 to 'Offtopic' and +1 to 'Troll' in one comment; moderator B votes -1 to 'Troll' and +1 to 'Insightful' in the same comment. Then they are combined and the result is (Score: whatever, Insightful) if the score is high and (Score: whatever, Offtopic) if the score is very low. Those categories would always be between -1 and +5, just like the normal moderation.
Also, another cool misfeature that would be worth adding to/. would be personal stories (aka Time For Another Harddisk;-) ). This would allow people to post articles and have superuser status (i.e. Rob-like) in a special 'article area' that would be accessed wia the 'User Info' link. People who abuse it (like trying to fill up the whole/. HDD subsystem) would lose their privileges of creating new articles. This would allow every people to have his own slashdot, to post stories that wouldn't go through to the main page (no p0rn of course), but if popular enough could be linked to (or even appear directly) in the main site. Then we could add a 'Most Active Current User Stories' slashbox so people could read interesting threads other people liked (a bit like a daily-updated HOF).
Anyway, it's getting late here. Tomorrow I'll read the comments.
I also seems to have a kind of ADD. Read the Jargon File; most hackers seem to have it. I think it's good, too. Sometimes I get in a kind of deep hack mode, though I guess I need more training;-) . I also read very fast, in that mode. It's interesting: you start reading slowly, then you get 'into' what you are reading and finish it pretty fast.
And I also have that problem of 'getting lost' in nowhere (I lose the ability to concentrate, it's sometimes a problem but truly useful when you're in a waiting room for 1 hour and you have already read all the magazines there.) I'm 'training' how to change to those states (normal, 'lost', hyperfocus).
BTW, a good slashdot pool seems to have surfaced here.
Lots of thoughtful comments above. People seem to really think much about it.
But I do believe that what we need is not only Artificial Intelligence. We also need Artificial Emotion. The entity must have it own will, its own desire to live. And without emotion you can't do that.
Thinking in rational ways only can't go really far - you wouldn't have really a life, you wouldn't be really more than a program. We need to learn how to code the complex and nonlogic sensations that we feel in an artificial entity's pseudo-brain. Then we would reach not only the full AI, but also something beyond it - the full Artificial Life.
Somebody must have been predicting this. This is the second most asked question on Slashdot - the first one is 'Why kids kill?'. I think they are about exactly the same thing.
Another thread is the recurrent 'We should make a free hackers' country' - and it's the same thing. I think we should stop and consider the relationship between those threads. They appear everytime not out of randomness, but because they have something to tell us.
I hope they do not leave that backdoor they left on q1 and q2 (the one in which a packet with a forged ip source and a special password could run any console command) and avoid other bugs (like the old buffer overflows).
So you wanted that the website that announced this only allowed people with passwords. What's next? See "The Right To Read", from RMS (couldn't find the URL).
If you have the decoders, you know what the decoders would expect, and then you could simply create your own encoder to create a new stream the the decoders would understand.
In this case, it's possible that you come up with an encoder that's even BETTER than the closed source ones.
Of course, it's possible to get an encoded file and its output (well, lots of encoded files and lots of outputs) and _guess_ the file format. You are not doing anything wrong - you're not reverse-engineering their code, you are just comparing the inputs and outputs and analysing them. You could get it right. Or totally wrong. Anyway, you got the decoder.
Last time I checked the MPEG standard, you had to buy 3+ books each $50+. No pdf docs.
I want specs that I can get for free just for the sake of it, not specs that cost me too much money so that I have to think before I buy it if I need it. The answer is usually no, so I have one less spec. And if I buy one, I cannot scan it and put it online - the specs itself are not 'free' (as in freedom), even if their content is.
Sorry for the rant, but I got pissed a while before when I saw a spec that cost $500 (the homerf one), and since then all paid specs make me nervous.
We need to shut down the hardware to avoid mandelbugs. Never asked why all the linux boot loaders turn off the floppy drive motor? To avoid undeterministic state that could cause unexpected bugs depending on the exact point of operation, system heat condition, and the phase of the moon.
The Right Way to do this is to make a full dump of the system state (processes etc.), load the new kernel in the memory (compressed), install a trampoline in memory, and jump to it. The trampoline would:
a) Do the required hardware shutdown (to avoid undetermined state). Doing this without breaking a modem connection is left as a exercise to the reader. b) Disable NMI and interrupts c) Dump the kernel structures (ps tables, etc) d) Change the structures to match the format in the new kernel (the 1st hard step) e) Load the new kernel with a special flag. It will initialize, reinitialize the hardware, check the flag and jump back to the trampoline. f) The trampoline disables NMI and interrupts (again) g) The trampoline restore the kernel structs (now in the new kernel's format), calling special __initfunc functions in the kernel to alloc memory and setup the new values. It also reloads the modules the same way. h) The program states is undumped (the 2nd hard step) by the kernel (called from the trampoline). i) The trampoline jumps to the kernel, which cleans up and frees the trampoline and its data. j) The kernel sends a SIGKUPD (kernel update) to all processes and runs schedule. k) The update program notices the signal, and call the bootloader to setup the boot block (so if you have a power outage here the kernel to run will be the new one).
Note: the process dumps are to the disk (root directory, probably)
The hard parts: 0. Generating the trampoline. 1. Converting the kernel structures (doing this automatically, with version gaps of 2-3 minor numbers, is pretty hard). 2. Restoring the processes state. Some sutble things have changed, and will crash many processes. The signal to make them re-exec themselves helps a bit, but legacy (ahem) apps will suffer.
Do you think we upgrade just because we have nothing else to do? (Well, sometimes you're right:-).) But people upgrade because they _know_ that the fix for the latest exploitz (to use your way of saying plurals) and _awful_ bugs are there. Every time I check the diffs, I see some _obvious_ bug fixes (if you can't recognize one, you're not a programmer) that could _have_ hurt me.
Update early, update often (sorry couldn't resist:-) ).
Since you're against Linux, what you said is FUD. I'm gonna send a bunch of crazy Linux zealots and Linus minions to deploy a nuclear warhead on you.
While I agree that linux users do sometimes overreact, I think that almost everything that says linux don't work is also a kind of reaction - 'If it doesn't fit what I currently believe as right it is wrong and I should by all means try to destroy it, but after the Media says it's right I will rethink and say to all, "forget what I said before"'.
Slashdot Mirror
Some people have already been thinking about the legal implications of 3D printers. You might be interested in the following paper:
"The Intellectual Property Implications of Low-Cost 3D Printing"
This reminds me of a funny situation which happened on the 7 FISL early this year.
At one of the conference rooms, while everybody was waiting for the speaker to start, suddenly out of somewhere comes a Windows boot sound.
Everybody started yelling at and criticizing the person with that laptop (why would someone use Windows at FISL?).
That until the person explained: he was using Windows just because some of his laptop's hardware was not supported, was using it just for VMWare (where Linux would be running), and in fact the Windows install didn't look like it got much use (apart from a couple of VMWare icons, it looked like a new install).
He was loudly applauded when VMWare started (booting Linux, of course).
There already is an alternate root for the EU: Open Root Server Network
Yes... When I read it (in the original, as I'm Brazilian) I was stunned... A politician that understands Free Software! And knows the OSD (aka DFSG)!
Well, since he seems to understand it, he must know how to speak English (well, most people here know Spanish or English as a second language, since it is a requisite in many jobs). And if he knows all those things about the OSD, he might also know you're one of the ones that created it in the first place. So write to him yourself. Tell him who you are. And ask if he might want some help. Though I can't see how you could make a difference, it's always nice to try.
It's always sad to lose an old friend. Farewell, Mir. May you rest in peace.
Yes there are some of them in some of the reviews, and it has IIRC 1+Mk/sec (while the K6 is at ~ 0.5Mk/sec)
One of the reviews said that the current MTRR won't work (they had to disable it). They also said AMD had contacted the MTRR maintainer (!), so the support will probably be added soon.
OK, we'll need to do some things to add Athlon support to Linux:
a) Kernel - we need to add the Athlon to the cpuid lists so it won't come with the CPU type blank
b) gcc - we will have to edit the i386.md, i386.c and i386.h so gcc will know the instruction delays when you use -mcpu=k7 (generates faster code)
c) gas - we will have to add the new K7 instructions to it
d) rc5des (offtopic) - we need k7 cores
e) Xfree86 - we need optimized drivers for K7 (and for other archs, AFAIK we do not have any CPU-specific optimization in the drivers
Maybe we would need some SMP changes too, and UDMA and AGP support for the new chipsets.
Any more changes?
First, stop trolling. We all know you are the same troll that appeared in the first thread and on yesterday's article. You can disguise as an AC but you can't disguise your writing style.
;-P ?
"Being agressive is not evil", you say. Fine. The Linux crowd is also very agressive. When they find a non-free program they like, they try to clone it. This might piss the ones who wrote the non-free app, but then, like you said, it's capitalism...
"This pisses people off because you ruined their business." Like when some version of Windows had a fake error message when run in some non-Microsoft version of DOS? Even with the code to detect the OS hidden under several layers of anti-RE code, someone found out.
"My point is if you do not like it, write for another OS." This is exactly what we are doing. And this is exactly what Microsoft wants to prevent (remember the Halloween documents?). Also, this sentence showed you wrote before thinking; you're ranting.
After this point, you go on and on with non-clear thinking (first you say it's simple, then you say it isn't; then you rant about supporting other OSs when Free Software is the most ported kind of software (and we even have Windows versions of most of them)). Then you say that the problem with the LINUX crowd (why all caps?) is that they think that it is a new way of life/thinking (no we don't think that; we know it's older than proprietary software). Then you go on and say it's business plain and simple. Funny, where's my paycheck
Then you say that we should search for money only. But Free Software is not market-driven, it is driven by the needs of its users. And last, you mention someone I've never heard about, but fail to provide a link.
I hope you realize your cover blew up (posting as an AC is useless to disguise you're the same guy as some other random guy when you can't disguise your style and your way of thinking) and that you should stop. You won't win without a rational argument.
I could go on and on, but I left it as an exercise to the fellow slashdotters.
When the CPU has been idle for some time, Linux (and other well-engineered OSs) issue the hlt instruction. It puts the CPU in a 'sleep' state from which it gets out in any interrupt. After returning from the interrupt, it already 'forgot' it was in an sleep state and continues working normally.
.com with only 18 hlts. When I run it, it waited for a second (timer is 18.2/sec under DOS, 100/sec under Linux/i386) and exited (ret to int 20 on psp).
When the CPU is idle, this means no proccess has anything to do, and the only things left to do are answering to keyboard, timers, modems, sound DMA exaustion, IDE read/write completed, mouse movement, and other things like that. And they all use interrupts, so it gets out of the halt state without losing any cycles (besides, the hlt is only used after the cpu has been idle for some time, so if you are fully using the cpu no hlt will appear).
When the proccesses are idle, they are all waiting for some event (blocking reads/writes and sleeping for some specified time being the most common). All those events can only happen after a ISR (so it'll wake up after a time interrupt, or after a disk interrupt, or after a keyboard interrupt, or after a NIC interrupt, etc.). So the CPU has actually nothing useful to do until an ISR happens.
Once I wrote (under DOS, before I got Linux) a
I hope someone understood.
So you want to categorize everything relevant to the Internet, since the first days of Eniac and maybe even some trivia about what came before (telegraph, phone systems, etc.). The problem here is: the net is so big and touches so many topics, iteracting with each other in a way that truly resembles a web.
For example, we have the phreakers' history. Sure, they shaped a lot of our current world (if it weren't for them, and from the ones who borrowed the name 'hackers' and started compromissing lots of sites' security, we would have almost no security on our systems today; no flames please).
Then we have the other 'hackers' history, this one portrayed in the now-ESR-maintained Jargon File. It also has a lot of history of its own, including the recent events (which are a lot).
If we zoom in a bit, we get (for example) the Debian's history. How it started, how it grow, the issues it faced, etc; and the Slashdot's tale, a virtual pub with its anti-M$ zealots, censorship attempts, the First Posters, MEEPT!, the gnulix_guy, and lots of other people.
Then we go back out and drift a bit. We get the tales of an awful clone of a brain-dead OS, the os called CP/M and the clone is... Well you can guess it. This is also long, we had years and years of borging and fighting, from the bad to the worst to the not-so-bad (5.0+), then (in parallel) the almost unknown old Windozes, the best-of-all Win3.1x, and the progressive bloat of the 9x family.
We also have a corner for the spammers history, entangled with the USENET and email history. We have a corner for the UNIX wars and their siblings (*BSD). We have a corner for rec.pets.cats and the crosspost flammage that went there.
All of this is significant for some groups and completely irrelevant to other ones. However, if we left them out, it would be discrimination.
I think this is way too big for just a small group of persons; we need a bazaar style. The problem is that the Internet time is a lot faster than the real world's, i.e., more happens in a specific amount of time in the net than in the Real World. This is because the Net is composed from the lifes not of just a geographically close group, but from millions of people from around the world.
We would need some way of categorizing the facts. This would be hard. Some are time spans, some are specific dates, some are 'circa 1984'. We would also need to have lots and lots of subcategories, and some way to 'zoom in' from the less specific general timeline with only the most relevant facts to a very specific timeline with all the 1000+ events in a single week or in a specific but small topic in a larger period of time (like the dpkg history since Dec 1999).
To sum it up: you are nuts. We don't have enough diskspace. We would need the space of a Boeing-747 full of DVDs. (OK, I'm exagerating a bit, but not much). If we could do it, we would get lagged faster than you can say 'Benedict'.
Sorry if I got carried away, it's late. And if I mispelled something, it was on purpose.
Did anybody read an old RFC which described exactly the same problem (bandwidth hogs) and the solution to it (fair queueing)? It basically says that the routers should act as if they had one incoming queue for each possible source IP (probably it is only one queue, but the router would act _as if_ it were many). Then it goes round-robin getting one packet from each IP and putting it on the output links. If some single source tries to send more packets than the link can handle, packets from _just this source_ will get dropped. This is a non-issue for OSs with congestion control (like Linux) since they can detect the drops and use them to adjust the outgoing speed.
This way, if a single node is sending too much data most of it will get dropped and (if it has a good stack) it will reduce its sending speed automatically. Other users and the download speed wouldn't be affected (assuming they use congestion control too).
I think it would be better this way:
/. would be personal stories (aka Time For Another Harddisk ;-) ). This would allow people to post articles and have superuser status (i.e. Rob-like) in a special 'article area' that would be accessed wia the 'User Info' link. People who abuse it (like trying to fill up the whole /. HDD subsystem) would lose their privileges of creating new articles. This would allow every people to have his own slashdot, to post stories that wouldn't go through to the main page (no p0rn of course), but if popular enough could be linked to (or even appear directly) in the main site. Then we could add a 'Most Active Current User Stories' slashbox so people could read interesting threads other people liked (a bit like a daily-updated HOF).
1. Moderators receive moderation points as in the old system and use them in the score _only_.
2. Moderators (might be the same ones, or not) receive special pseudo-moderation points that are spent on the categories. Each category would have its "score". For example: moderator A votes +1 to 'Offtopic' and +1 to 'Troll' in one comment; moderator B votes -1 to 'Troll' and +1 to 'Insightful' in the same comment. Then they are combined and the result is (Score: whatever, Insightful) if the score is high and (Score: whatever, Offtopic) if the score is very low. Those categories would always be between -1 and +5, just like the normal moderation.
Also, another cool misfeature that would be worth adding to
Anyway, it's getting late here. Tomorrow I'll read the comments.
I also seems to have a kind of ADD. Read the Jargon File; most hackers seem to have it. I think it's good, too. Sometimes I get in a kind of deep hack mode, though I guess I need more training ;-) . I also read very fast, in that mode. It's interesting: you start reading slowly, then you get 'into' what you are reading and finish it pretty fast.
And I also have that problem of 'getting lost' in nowhere (I lose the ability to concentrate, it's sometimes a problem but truly useful when you're in a waiting room for 1 hour and you have already read all the magazines there.) I'm 'training' how to change to those states (normal, 'lost', hyperfocus).
BTW, a good slashdot pool seems to have surfaced here.
Lots of thoughtful comments above. People seem to really think much about it.
But I do believe that what we need is not only Artificial Intelligence. We also need Artificial Emotion. The entity must have it own will, its own desire to live. And without emotion you can't do that.
Thinking in rational ways only can't go really far - you wouldn't have really a life, you wouldn't be really more than a program. We need to learn how to code the complex and nonlogic sensations that we feel in an artificial entity's pseudo-brain. Then we would reach not only the full AI, but also something beyond it - the full Artificial Life.
Somebody must have been predicting this. This is the second most asked question on Slashdot - the first one is 'Why kids kill?'. I think they are about exactly the same thing.
Another thread is the recurrent 'We should make a free hackers' country' - and it's the same thing. I think we should stop and consider the relationship between those threads. They appear everytime not out of randomness, but because they have something to tell us.
I hope they do not leave that backdoor they left on q1 and q2 (the one in which a packet with a forged ip source and a special password could run any console command) and avoid other bugs (like the old buffer overflows).
1. RedHat is a company.
;-)
2. Companies are evil.
therefore,
3. RedHat is evil.
Jeez, RedHat is doing the Right Thing (or at least trying to do it). But they made the wrong thing by being a company. So, they should not be trusted.
If you don't like it, use Debian. Or create your own distro based on pristine upstream source. But stop whining.
And to those that like it: be ready to change to something else if it blows up
So you wanted that the website that announced this only allowed people with passwords. What's next? See "The Right To Read", from RMS (couldn't find the URL).
If you have the decoders, you know what the decoders would expect, and then you could simply create your own encoder to create a new stream the the decoders would understand.
In this case, it's possible that you come up with an encoder that's even BETTER than the closed source ones.
Of course, it's possible to get an encoded file and its output (well, lots of encoded files and lots of outputs) and _guess_ the file format. You are not doing anything wrong - you're not reverse-engineering their code, you are just comparing the inputs and outputs and analysing them. You could get it right. Or totally wrong. Anyway, you got the decoder.
Last time I checked the MPEG standard, you had to buy 3+ books each $50+. No pdf docs.
I want specs that I can get for free just for the sake of it, not specs that cost me too much money so that I have to think before I buy it if I need it. The answer is usually no, so I have one less spec. And if I buy one, I cannot scan it and put it online - the specs itself are not 'free' (as in freedom), even if their content is.
Sorry for the rant, but I got pissed a while before when I saw a spec that cost $500 (the homerf one), and since then all paid specs make me nervous.
Just my $0.02.
We need to shut down the hardware to avoid mandelbugs. Never asked why all the linux boot loaders turn off the floppy drive motor? To avoid undeterministic state that could cause unexpected bugs depending on the exact point of operation, system heat condition, and the phase of the moon.
The Right Way to do this is to make a full dump of the system state (processes etc.), load the new kernel in the memory (compressed), install a trampoline in memory, and jump to it. The trampoline would:
a) Do the required hardware shutdown (to avoid undetermined state). Doing this without breaking a modem connection is left as a exercise to the reader.
b) Disable NMI and interrupts
c) Dump the kernel structures (ps tables, etc)
d) Change the structures to match the format in the new kernel (the 1st hard step)
e) Load the new kernel with a special flag. It will initialize, reinitialize the hardware, check the flag and jump back to the trampoline.
f) The trampoline disables NMI and interrupts (again)
g) The trampoline restore the kernel structs (now in the new kernel's format), calling special __initfunc functions in the kernel to alloc memory and setup the new values. It also reloads the modules the same way.
h) The program states is undumped (the 2nd hard step) by the kernel (called from the trampoline).
i) The trampoline jumps to the kernel, which cleans up and frees the trampoline and its data.
j) The kernel sends a SIGKUPD (kernel update) to all processes and runs schedule.
k) The update program notices the signal, and call the bootloader to setup the boot block (so if you have a power outage here the kernel to run will be the new one).
Note: the process dumps are to the disk (root directory, probably)
The hard parts:
0. Generating the trampoline.
1. Converting the kernel structures (doing this automatically, with version gaps of 2-3 minor numbers, is pretty hard).
2. Restoring the processes state. Some sutble things have changed, and will crash many processes. The signal to make them re-exec themselves helps a bit, but legacy (ahem) apps will suffer.
The 'hard part number one' is the worst one.
Do you think we upgrade just because we have nothing else to do? (Well, sometimes you're right :-) .) But people upgrade because they _know_ that the fix for the latest exploitz (to use your way of saying plurals) and _awful_ bugs are there. Every time I check the diffs, I see some _obvious_ bug fixes (if you can't recognize one, you're not a programmer) that could _have_ hurt me.
:-) ).
Update early, update often (sorry couldn't resist
Since you're against Linux, what you said is FUD. I'm gonna send a bunch of crazy Linux zealots and Linus minions to deploy a nuclear warhead on you.
While I agree that linux users do sometimes overreact, I think that almost everything that says linux don't work is also a kind of reaction - 'If it doesn't fit what I currently believe as right it is wrong and I should by all means try to destroy it, but after the Media says it's right I will rethink and say to all, "forget what I said before"'.