The difference is that, last I checked, in Linux root can unset these flags.
In any 4.4BSD derived operating system, the flags can only be unset by root if the security level is under 1. The end result? You've got a bunch of files that the kernel will not let you modify (or depending on the flag, delete, do anything but append to, etc.) without dropping into single user. That means that even if someone gains root, his job is made difficult unless he has local access: placing trojans over important binaries becomes impossible, as does (more importantly) deleting the logs that logged your entry. See page 263 of Design and Implementation of the 4.4BSD Operating System for more information.
I think that predicting the death of the monopoly of Microsoft is a bit presumptious. Given it's current share, even winning the new PC market (which I doubt for other reasons that will come later) won't lead to the death of the monopoly.
Microsoft's dominance of the desktop has reached such a critical point where it's hard to expect the entire world to change. Predicting that [one | some] free operating system(s) will cause Windows to lose it's majority (a prerequsite for Microsoft losing it's monopoly) in six months is a bit like predicting that the US will convert to metric in six months.
Sure, we use base 10 for everything, so one could argue that the US will switch to metrics shortly because it makes sense. But we (as a nation) are so comfortable with the english system, as we are with the Windows platform, that change will be slow and painful. ESR says that the fact that PCs are getting cheaper is a good indicator that PC manufacturers will get fed up, but this I doubt.
I say this because PCs have always gotten cheaper, and hardware is getting cheaper as well. Given how most people see windows as indispensible, they have no objection to paying for a license. If the so-called "MS Tax" ends up being a much larger percentage of the total cost of the computer, I think the logical response from Microsoft would be to lower their prices; not to keep charging an amount people won't pay until they die.
Microsoft keeps up with things; they aren't about to lose their monopoly due to changes that were expected (cheaper, smaller, faster) but will rather require many more unexpected events to appear on the horizon: but note, MS has their radar up.
I'm no clinical psychologist (but I did get an A in HS Introduction to psychology!) but I would guess that the reason people work better in a "war-room" is that they feel they have a purpose. Besides the more obvious reasons mentioned in the article, such as encouring cooperation, when you're working closely with other people what you're doing seems a bit more important.
Sitting down facing a screen, it doesn't really care if you work on the programming task at hand or if you play a couple rounds of xmame. With your peers all with you, you can't let the team down. So, by creating a team atmosphere, the end result is probably a constant fear of not wanting to screw things up for everyone;)
On a lighter note, does anyone out there work in a "war room" type enviornment? It sounds like somewhere I'd like to work, but only if the chairs were leather and really comfortable;)
It should be seen as a good thing: I'd much rather see two companies arguing over standards than spending their time finding ways to make their work totally incompatible and break all standards, making all code relevant to the older version worthless.
The fact that standards exist are wonderful; debate over them is proper to make sure that all aspects are considered. Even if Microsoft wants only what's best for Microsoft, by causing BizTalk to be better off, the XML spec is not about to become more restricted. (Just their DTD [XML buddies: am I getting the terminology right?], which they could do anyway)
Perhaps I don't fully understand the case, but it seems to me that a temp worker knows he's not getting benefits. Hence the reason that they are usually paid more per hour.
It doesn't seem that there was anything even close to a promise, let alone an exchange of consideration, in which Microsoft promised people benefits, or that they would only be employed for a certain duration of time (they could quit whenever they wanted)
Where were the temp workers wronged? They weren't expecting benefits, they didn't receive them. They received employment at the amount that they work for..
Any objection to this system would be an objection to the temp agencies as a whole- not what Microsoft did. Temp workers certainly weren't wronged; given that they don't receive benefits they generally carry home more dollars per week than their full-time equivilent (of equal experience, etc)
Re:I disagree with the concept of imprisoning him
on
Spammer Pleads Guilty
·
· Score: 1
The money from the fine would go to the state. In theory that results in the state needing that less money, and consequently lower taxes...
But you're right, the STATE wastes all of our money;) What we need is reform at every level.. local, statewide, and federal.
I disagree with the concept of imprisoning him
on
Spammer Pleads Guilty
·
· Score: 1
Why? Because it's prison, and he's of no use to anyone in prison. Look, the world has problems. Poverty is everywhere. The majority of the world (some say as high as 75%) can be classified as "hungry"
And we're putting this guy in prison. He's not dangerous. Heck, he's technically minded. His action was inspired by capitalism, and therefore the logical punishment is to fine him. A big, hefty fine. And heck, while we're at it, we should be using these fines for a (real) good cause. Let's fine this guy a million dollars, and then use that to feed the hungry.
Now, there's no point filling up our prisons. In a capitalist society, economic punishment is effective enough! And, it lets him spend his time working- although he won't see the benefits of his labor until the debt is paid off, the rest of us will.
But backporting is not easy! The time would be better spent getting 2.4 into better shape, making those who MUST live on the cutting edge be able to do so.
2.4 is no longer cutting-edge technology, it's getting old by many standards. All of the BSDs have developed wonderful USB stacks and driver support over time, and all of these "shiny new" features of 2.4 are starting to get old. So, it's overdue for release (a year overdue!): it's time to stabilize it and then get more cutting-edge things to keep up with the Joneses:)
I understand porting back little bugfixes and all, but revamping USB support? That sounds like wasted effort to me. Heck, why don't we just backport ALL of 2.4?;\
2.4 is long overdue, Linus was first hoping for fall of 1999, and now here we are: almost in 2001. Now I don't mean to be mean, but if perhaps a bit more time was spent working forward than unnecessairly backwards, we'd all be a lot better off.
If you want to get into 2.4, then you'll need to convince people to switch- and backporting every feature in 2.4 to 2.2 is not the way to do that. I think that if we're not at the point already where we can just through 2.2 into occasional maitenence, then something is wrong. It's akin to backporting SMP support to 0.99 before it even made it into 2.0.
The "forks" of UNIX have given us UNIX on machines that aren't PDP mainfraimes. They've given us virtual memory, demand paging, page replacement, networking support (sockets!), FREE software! They've given us job control, reliable signals, groups (for permissions), termcap, vi! They've given us NFS and RPC!
Look, there are some unixes that haven't offered much more over other versions. And they aren't terribly popular, either. But even if you look at one of the most "blah" UNIXes out there, you'll still see valuable contributions: like cscope (anyone who's used it, loves it)
If big providers started not allowing blacklisted hosts to pass, then they'd save on bandwith and help reduce the problem for everyone:)
A lot of people wonder if it's a bit unfair, but I've never met anyone who stayed on the RBL list for long and did not deserve to. So, Verizon, don't drop the ball and utilize the RBL!
I know I don't accept email from those blacklisted, and I certainly don't forward it- and if everyone did the same, the world would be a better place. </end cheesy sally struthers informerical>
In my opinion, you may very well see real-time Linux as the domain of hardware hobbyists, but I have severe doubts it'd make it into professional systems.
No, I'm not insulting its quality, but I question it's size. How small can you get a real-time linux kernel? In most professional applications, cost is a big issue. If you have only 4k bytes to work with, you'll probably end up writing the routines needed in straight asm for the processor and sending that to the chip.
I'll be watching how this product develops, and especially take notes of where it is deployed. Good luck, and don't forget to enjoy your work!
Wow, congratulations. I'm just wondering how a company can expand that quickly- I imagine that a good portion would be spent on advertising, and then some more spent for investments/acquisitions, and probably some left liquid.
It will be interesting to see how this newfound capital will affect an already established company such as BSDi (note: I'm aware that BSDi is a young company, but it's really just an expanded BSDI: which is an established company)
No, oxidization is the loss of an electron. It can happen if there's anything around to receieve the electrons from the solder: most likely the result of the flux in the solder breaking down into something a bit more reactive. Or any other chemicals the board may have been exposed to breaking down and then reacting.
Don't you remember highschool chemistry? LEO the lion says GER! (now I know I'm not the only one who learned that it's a mnemoic device to learn- Loss Electrons Oxidization and Gain Electrons Reduction)
TdR: I think there's a constant overestimation of how big the Linux development community actually is. I don't think they have thousands of developers working on the hard problems, just on the simple problems and the GUI things. But in the actual operating system, I would be surprised if they have more than 200 people, the kernel, the libraries, the basic utilities. On the other hand, we have about 40 persons who work on just those parts all the time.
It's interesting to see how there's a lot of speculation aroud Linux, with not a lot of basis in fact. For instance, could anyone tell me how many people actively contribute to the parts that Mr. DeRaadt mentioned? His guess is as good as any.
More interesting are the fairly outlandish estimations. Look at the Linux Counter 162,506 users registered, and a guess of sixteen million Linux users total. Where are they all hiding? Only 1% have registered? What makes you say that?
It's the same in a lot of the open source world; speculation gets you nowhere. Let's look at what we can quantitvely measure, such as downloads and registered users, and end the guessing games.
Numbers of developers are meaningless: all that matters is the total quality of the development process.
Thirty five years! Congratulations to everyone at NASA for making contact, and more importantly to the engineers who designed and built this satelite!
NASA gets a lot of bad press for say, not doing metric conversions, but this clearly is an example of excellent professionals doing their best. A lot of solder joints will oxidize and go bad before thirty five years.. this goes to show that the NASA engineers were not considering how long the probe was wanted when they built it, but rather built it for its maximum life. If only VCRs and such were built like that: today's consumer electronics have a bunch of cheap, light plastic parts:(
Do what I do with my work, such as chessd -- release it under
a very lax license. All of my work, regardless of the project, is either
under the public domain or a 3-clause BSD license.
What's there to enforce? The copyright reproduction clause on the BSD license? First of all, it's not worth my hassle. Secondly, if someone
does use a significant portion of my code, they'll probably reduce the copyright somewhere anyway- a line of ink is much cheaper than hiring an lawyer for even an hour.
When you turn what we're doing, which is programming for our own enjoyment, into some political cause, you complicate the matter with having a legal
headache and worrying about enforcing things. Relax, this is supposed to be fun, and don't use a license that reads
like a contract- we get our share of legalese by watching the Florida recount..
I know that you can purchase either BSD/OS or FreeBSD from BSDi with
support; and customers use both (especially the former, as it's been offered
by BSDi [then BSDI] longer) in areas that need
even better uptime than cash registers.
It's easy to find someone to offer you support, but you will need a big name for trust. LinuxCare, LinuxOne, RedHat, I'm sure one of them could offer you what you need. Shop around!
Engineers pride themselves in their ability to complete the job within the specifications in the best way possible. I don't think most would bother looking for new empolyment if they could, within reason, implement the incompatiblity.
Err, no, BSD was the original BSD. I think I see what you're saying: BSDI was founded by Bostic with other former CSRG members. It was a way to continue development of BSD after the Uni cut funding.
Realize, however, that there is a clear distinction between BSD and BSD/OS. One's very free, the other is not.
First of all, the statistic seems maybe a bit on the high side, but it's not totally unacceptable.
The statistic is there is a 1 in 10,000 chance of ANYBODY getting hurt. That puts the odds of YOU getting hurt at :
(1/10,000) * (1/6,000,000,000) = 1/60,000,000,000,000. Now those are odds I like!
Re:Programmers Make Computers Slower Year by Year
on
Netscape 6 Vs. 4.7x
·
· Score: 1
You do not know what you are talking about.
Spreadsheet updates have always been cutting edge benchmark phenomena. When you are recalculating a spreadsheet with several hundred thousand or million entries (yes, people do build those) and you have to recalculate it again and again and again and again and again as you refine your model, you quickly learn the difference between 30 seconds and 2 seconds.
But not to the point where you would gain anything from using asm as opposed to what your C compiler generates. Frankly, there's only so many ways you can add, subtract, multiply and divide.
C is also a crappy language for beginning programmers and for application programming in general, because it achieves the performance it gets by abandoning all error checking, so that writing to the 11th element of a 10-element array crashes your program instead of giving a sensible error message. While there is a place for C it is overused both at low levels where asm is really needed and at high levels where speed is not critical and a richer programming UI would speed development and bugriddance.
According to N869, writing beyond the nth element of an array of n elements size will invoke implementation-specific behavior. Same for accessing beyond the space allocated for you on the heap (Or, to put it in N869 terms, accessing a pointer beyond ptr+n where ptr is the pointer returned by malloc and n is the argument passed to it) That's the way it's been for as long as I know. Your complaint is with common implementations of C, not the language itself. There's a big difference.
Re:Programmers Make Computers Slower Year by Year
on
Netscape 6 Vs. 4.7x
·
· Score: 1
I don't know where you got the figure that much of Excel's calculation engine is written in asm. I'd like to see that.
Secondly, the calculation routines in a spreadsheet are not time-critical enough to the point where they should be written in asm. A good C compiler would write nearly as good code provided a fine math library and a not-braindead-programmer. And remember, the actual FP calculations that may go on in a spreadsheet pale in comparison to those that go on say, in Doom.
Now, yes there are times for asm, but on the PC they are quite rare (high performance 3D modelling engines) and can usually be done in C and yield roughly equal performance with a good compiler. But to say that we should learn asm to write better programs is absurd.
Slashdot Mirror
I apologize.
Does anyone have any idea? (and wow, an intelligent AC post!)
In any 4.4BSD derived operating system, the flags can only be unset by root if the security level is under 1. The end result? You've got a bunch of files that the kernel will not let you modify (or depending on the flag, delete, do anything but append to, etc.) without dropping into single user. That means that even if someone gains root, his job is made difficult unless he has local access: placing trojans over important binaries becomes impossible, as does (more importantly) deleting the logs that logged your entry. See page 263 of Design and Implementation of the 4.4BSD Operating System for more information.
Microsoft's dominance of the desktop has reached such a critical point where it's hard to expect the entire world to change. Predicting that [one | some] free operating system(s) will cause Windows to lose it's majority (a prerequsite for Microsoft losing it's monopoly) in six months is a bit like predicting that the US will convert to metric in six months.
Sure, we use base 10 for everything, so one could argue that the US will switch to metrics shortly because it makes sense. But we (as a nation) are so comfortable with the english system, as we are with the Windows platform, that change will be slow and painful. ESR says that the fact that PCs are getting cheaper is a good indicator that PC manufacturers will get fed up, but this I doubt.
I say this because PCs have always gotten cheaper, and hardware is getting cheaper as well. Given how most people see windows as indispensible, they have no objection to paying for a license. If the so-called "MS Tax" ends up being a much larger percentage of the total cost of the computer, I think the logical response from Microsoft would be to lower their prices; not to keep charging an amount people won't pay until they die.
Microsoft keeps up with things; they aren't about to lose their monopoly due to changes that were expected (cheaper, smaller, faster) but will rather require many more unexpected events to appear on the horizon: but note, MS has their radar up.
Sitting down facing a screen, it doesn't really care if you work on the programming task at hand or if you play a couple rounds of xmame. With your peers all with you, you can't let the team down. So, by creating a team atmosphere, the end result is probably a constant fear of not wanting to screw things up for everyone ;)
On a lighter note, does anyone out there work in a "war room" type enviornment? It sounds like somewhere I'd like to work, but only if the chairs were leather and really comfortable ;)
The fact that standards exist are wonderful; debate over them is proper to make sure that all aspects are considered. Even if Microsoft wants only what's best for Microsoft, by causing BizTalk to be better off, the XML spec is not about to become more restricted. (Just their DTD [XML buddies: am I getting the terminology right?], which they could do anyway)
It doesn't seem that there was anything even close to a promise, let alone an exchange of consideration, in which Microsoft promised people benefits, or that they would only be employed for a certain duration of time (they could quit whenever they wanted)
Where were the temp workers wronged? They weren't expecting benefits, they didn't receive them. They received employment at the amount that they work for..
Any objection to this system would be an objection to the temp agencies as a whole- not what Microsoft did. Temp workers certainly weren't wronged; given that they don't receive benefits they generally carry home more dollars per week than their full-time equivilent (of equal experience, etc)
But you're right, the STATE wastes all of our money ;) What we need is reform at every level.. local, statewide, and federal.
Do your part, vote. And not for BushGore ;)
And we're putting this guy in prison. He's not dangerous. Heck, he's technically minded. His action was inspired by capitalism, and therefore the logical punishment is to fine him. A big, hefty fine. And heck, while we're at it, we should be using these fines for a (real) good cause. Let's fine this guy a million dollars, and then use that to feed the hungry.
Now, there's no point filling up our prisons. In a capitalist society, economic punishment is effective enough! And, it lets him spend his time working- although he won't see the benefits of his labor until the debt is paid off, the rest of us will.
2.4 is no longer cutting-edge technology, it's getting old by many standards. All of the BSDs have developed wonderful USB stacks and driver support over time, and all of these "shiny new" features of 2.4 are starting to get old. So, it's overdue for release (a year overdue!): it's time to stabilize it and then get more cutting-edge things to keep up with the Joneses :)
2.4 is long overdue, Linus was first hoping for fall of 1999, and now here we are: almost in 2001. Now I don't mean to be mean, but if perhaps a bit more time was spent working forward than unnecessairly backwards, we'd all be a lot better off.
If you want to get into 2.4, then you'll need to convince people to switch- and backporting every feature in 2.4 to 2.2 is not the way to do that. I think that if we're not at the point already where we can just through 2.2 into occasional maitenence, then something is wrong. It's akin to backporting SMP support to 0.99 before it even made it into 2.0.
The "forks" of UNIX have given us UNIX on machines that aren't PDP mainfraimes. They've given us virtual memory, demand paging, page replacement, networking support (sockets!), FREE software! They've given us job control, reliable signals, groups (for permissions), termcap, vi! They've given us NFS and RPC!
Look, there are some unixes that haven't offered much more over other versions. And they aren't terribly popular, either. But even if you look at one of the most "blah" UNIXes out there, you'll still see valuable contributions: like cscope (anyone who's used it, loves it)
Relax..
A lot of people wonder if it's a bit unfair, but I've never met anyone who stayed on the RBL list for long and did not deserve to. So, Verizon, don't drop the ball and utilize the RBL!
I know I don't accept email from those blacklisted, and I certainly don't forward it- and if everyone did the same, the world would be a better place. </end cheesy sally struthers informerical>
No, I'm not insulting its quality, but I question it's size. How small can you get a real-time linux kernel? In most professional applications, cost is a big issue. If you have only 4k bytes to work with, you'll probably end up writing the routines needed in straight asm for the processor and sending that to the chip.
I'll be watching how this product develops, and especially take notes of where it is deployed. Good luck, and don't forget to enjoy your work!
It will be interesting to see how this newfound capital will affect an already established company such as BSDi (note: I'm aware that BSDi is a young company, but it's really just an expanded BSDI: which is an established company)
Don't you remember highschool chemistry? LEO the lion says GER! (now I know I'm not the only one who learned that it's a mnemoic device to learn- Loss Electrons Oxidization and Gain Electrons Reduction)
It's interesting to see how there's a lot of speculation aroud Linux, with not a lot of basis in fact. For instance, could anyone tell me how many people actively contribute to the parts that Mr. DeRaadt mentioned? His guess is as good as any.
More interesting are the fairly outlandish estimations. Look at the Linux Counter 162,506 users registered, and a guess of sixteen million Linux users total. Where are they all hiding? Only 1% have registered? What makes you say that?
It's the same in a lot of the open source world; speculation gets you nowhere. Let's look at what we can quantitvely measure, such as downloads and registered users, and end the guessing games.
Numbers of developers are meaningless: all that matters is the total quality of the development process.
NASA gets a lot of bad press for say, not doing metric conversions, but this clearly is an example of excellent professionals doing their best. A lot of solder joints will oxidize and go bad before thirty five years.. this goes to show that the NASA engineers were not considering how long the probe was wanted when they built it, but rather built it for its maximum life. If only VCRs and such were built like that: today's consumer electronics have a bunch of cheap, light plastic parts :(
What's there to enforce? The copyright reproduction clause on the BSD license? First of all, it's not worth my hassle. Secondly, if someone does use a significant portion of my code, they'll probably reduce the copyright somewhere anyway- a line of ink is much cheaper than hiring an lawyer for even an hour.
When you turn what we're doing, which is programming for our own enjoyment, into some political cause, you complicate the matter with having a legal headache and worrying about enforcing things. Relax, this is supposed to be fun, and don't use a license that reads like a contract- we get our share of legalese by watching the Florida recount..
It's easy to find someone to offer you support, but you will need a big name for trust. LinuxCare, LinuxOne, RedHat, I'm sure one of them could offer you what you need. Shop around!
Engineers pride themselves in their ability to complete the job within the specifications in the best way possible. I don't think most would bother looking for new empolyment if they could, within reason, implement the incompatiblity.
Realize, however, that there is a clear distinction between BSD and BSD/OS. One's very free, the other is not.
The statistic is there is a 1 in 10,000 chance of ANYBODY getting hurt. That puts the odds of YOU getting hurt at : (1/10,000) * (1/6,000,000,000) = 1/60,000,000,000,000. Now those are odds I like!
But not to the point where you would gain anything from using asm as opposed to what your C compiler generates. Frankly, there's only so many ways you can add, subtract, multiply and divide.
C is also a crappy language for beginning programmers and for application programming in general, because it achieves the performance it gets by abandoning all error checking, so that writing to the 11th element of a 10-element array crashes your program instead of giving a sensible error message. While there is a place for C it is overused both at low levels where asm is really needed and at high levels where speed is not critical and a richer programming UI would speed development and bugriddance.
According to N869, writing beyond the nth element of an array of n elements size will invoke implementation-specific behavior. Same for accessing beyond the space allocated for you on the heap (Or, to put it in N869 terms, accessing a pointer beyond ptr+n where ptr is the pointer returned by malloc and n is the argument passed to it) That's the way it's been for as long as I know. Your complaint is with common implementations of C, not the language itself. There's a big difference.
Secondly, the calculation routines in a spreadsheet are not time-critical enough to the point where they should be written in asm. A good C compiler would write nearly as good code provided a fine math library and a not-braindead-programmer. And remember, the actual FP calculations that may go on in a spreadsheet pale in comparison to those that go on say, in Doom.
Now, yes there are times for asm, but on the PC they are quite rare (high performance 3D modelling engines) and can usually be done in C and yield roughly equal performance with a good compiler. But to say that we should learn asm to write better programs is absurd.