Well I can agree that NetBSD would be just as useful in this role as the Linux kernel would be. I'm not suggesting that they deploy a full GNU setup on the thing, just a kernel. My point was that with investors, often times, Linux == cool and NetBSD == what is that?
For one, outside of the geek community who the hell knows what NetBSD is. There is a bit of appeal for companies to be associated with Linux, especially from the viewpoint of their shareholders. They will think, "hey, I've heard of this Linux thing, must be they're on the cutting edge of things." So they go out and buy another 1000 shares of the companies stock.
Besides that point, recent Linux development have been targetted towards making Linux more friendly on embedded hardware.
Personally, I rather like the fact that people are using Linux and will be giving(even if it is forced giving) their changes back to the community and increasing the general knowledge of different commputing platforms(source code is often excellent documentation on a new platform)
Obviously you never read the flame war between Tannenbaum and Linus that took place in 1991. Theoritical design is one thing, but writing something that works and works well is another.
But, if you don't like how things are going with the linux kernel, nobody is stopping you from starting your own fork of the linux kernel. Import the whole source tree into your own CVS repository, get some developers and get some work done.
Could have something to do with the fact that his wife, Tove, was pregnant and gave birth in late November I believe. I think that would take a bit more priority over tbe release of 2.4.0
You might want to take a look at reiserfs for your news spool. Reiserfs doesn't have inodes(not in the traditional sense anyways). What it does it creates a unique hash for each file and uses that for an inode, so you end up having basically an unlimited number of inodes.
Also reiserfs is not block oriented so you end up with little to no wasted disk space when dealing with small files. Also with extremely small files, the data is often stored with the metadata rather than allocating a chuck of disk space elsewhere.
On the other hand, a Christian could argue that Darwinist are teaching evolution as if it were fact.
Neither one is truly provable, as both are proven in the terms of the people attempting to prove it.
Consider that if you use scientific method to prove or disprove creationism, of course you are going to be led to disprove it, as your already biased. Even without the bias, creationism is illogical in the eyes of science.
With that being said, if you try to prove evolution within the context of Christianity, you are going to be led to disprove it as well, as you are already biased. And of course evolution is illogical in the eyes of a Christian.
Its really a mute point to argue the issue, as people are going to believe what they like, and most people are not huge fans of trying to be "converted". Next time think about that you try to shove evolution down a creationist throat. Do you want a creationist shoving creationism down your throat?
Yeah...try to compile a rather large package on your 486DX/33MHz box with 8mb of ram..see how long it takes to compile something large, like for example, emacs.
And if you say that you're compiling it yourself so you know its not backdoored, well do you check ever single line of code for backdoors? I didn't think so.
So what you're saying is that things need to be dumbed down for the lowest common demoninator of our society? Oh thats right, god forbid we educate people on critical thinking skills like logic, cause and effect, context clues etc. So I say, if we want the lowest common demonitator, why not make a "shiny things" interface. No text anywhere, because god forbid people who can't read won't be able to use it. We won't need a keyboard either, as nobody knows how to type. I'm sorry you can have that if you like, but I'll pass on the OS for masses of people requiring thorazine.
You know, there is nothing stopping you from going
to Washington yourself and lobbying yourself. If your congresscritter doesn't know about your concerns, how are they supposed to know how you feel?
Perhaps some organization(FSF maybe, or perhaps Software in the Public Interest, maybe they already do lobby though?), should get some people around to lobby in DC for us free software people.
Take a look at the drivers for some of the Adaptec SCSI controllers, you'll see that there is a wad of code uploaded to the controller that we have no source for either...
I assume that since this is using a corba interface one could just as easily mount a postgresql database or anything else that is exportable via corba.
Somehow this all reminds me of how Plan9 does things, where *everything* is a file, that includes TCP connections etc, etc...
Now what would be more useful is using corba as a replacement for NFS and SUN RPC. Now that might be of interest...
A lot of mail servers allows a user to do fun things with their email address like this:
Say if a user has an email address like anonymous@slashdot.org and he wanted to track where some junk came from, so on slashdot his publishs his email like anonymous-coward@slashdot.org some mail servers(sendmail and postfix do this i think qmail does too) will get dropped in the mailbox of anonymous@slashdot.org
Its pretty simple and doesn't require a lot of screwing around with things. Of course not all mail servers are setup to do this..
I believe one can opt-out of the junk mail that the postal service delivers to everyone. I think there is a form you have to fill out though. Your best bet would be to talk to somebody at your local post office. Now this of course does nothing for the junk you get from people who have your exact address though...
I've you actually followed reiserfs development any you would know this. The issue was the fact that reiserfs knows how to handle filenames longer that 255, but the VFS in the linux kernel does not. So, reiserfs that is in 2.4.1-pre7 limits this to 255 characters..
As proof for you tiny little mind...
mkdir "$(perl -e 'print "x" x 768')"
mkdir: cannot create directory `xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxx': File name too long
You are also better off using the 3.5.x utils as reiserfs 3.6.x does not have a functional fsck yet(its in the works though). Not that you'll need it too often, but just in case.
MKLinux was a port of the Linux kernel to the Mach microkernel. Basically the linux kernel ran as a single server on the Mach microkernel. Which would really suck as you lose all the advantages of a microkernel and a monolithic kernel by doing that.
Of course Mklinux is pretty much dead with PPC Linux around now...
I think that a lot of the problem is due to XUL, take a look at Galeon to see how much faster it runs. Galeon is using the rendering engine(I think they are calling it Gecko, but I'm too lazy to check now) from Mozilla, via the gtkmozembed library.
I primarly use galeon at this point and I've been very happy with it. Almost as quick as NS4.x and a similiar memory footprint.
Heh..it is kinda funny. I know that there has been a patch or two on the linux-kernel mailing list doing zerocopy TCP, which the tcp checksumming being done on the ethernet controller(if it supports that of course). But from what I've read they have been able to max out a gigabit ethernet link with a p3@500mhz.
I don't see any of the BSDs doing that these days..
Not to mention a really dumb idea. Sure leave your door open, but don't expect anything left in the morning. Sure you *should* be able to do this, and you *should* be able to trust people. And most people can be trusted, but not everyone can.
There should be some sort of social responsiblity to keep computing equipment on the internet in a maintained, orderly fashion. If you do not want to do that, do not place it on the internet.
How about this one, sign a contract with your ISP that you either:
A. Agree to keep your equipment orderly, secure and maintained and that you agree to pay punitive damages(based on income) in a failure to do so that becomes exploited, unless you can prove that you acted in good faith maintain your equipment.
or
B. Agree to allow the ISP to filter certain types of outgoing traffic from your equipment. (Oversized or excessive ICMP packets, TCP packets with bad flags or excessive SYN packets, basically any type of data that is not normal).
These all seem reasonable to protect both the ISPs networks and the users.
As the IDE subsystem developer for Linux, how will you deal with this misfeature? Will you merely work around it in software, or will you stay true to the SPECS and implement the copy protect feature? Or perhaps have the copy protect a CONFIG option? Or will this be a layer below the kernel(in the chipset) and in such case, hack around that too(XORing the file as it goes to disk obscuring any goofy signatures and reversing the operation on the way back?).
So by the same token, people should be allowed to drive automobiles even if they don't know how to drive them? Or they should be allowed to use the telephone if they don't know basic telephone safety(Like don't give your address etc etc out to
strangers and all the other things you learned as a kid).
My point is that computers attached to the internet are not just "toys" but they are serious pieces of electronic equipment. Equipment that most people trust their finanical records and other aspects of their personal life to. At least software should go with conservative defaults for the uneducated. The people who do know what they are doing know how to change the defaults.
It would be fairly trival to have a bug of worm that gets into a system via a bug in outlook(or more often than not an education problem, like files named pr0n.jpg.exe etc..) and then phones home with all of the goodies to some random webserver in siberia. Oh and it installed a nice backdoor or something.
People just need to be educated about the risks, like with the box tossing up a message about loading remote images saying that "Loading images from remote servers that are received via email can be considered a privacy threat, if you know that you will not have this problem, click ok, otherwise the safe choice is no".
I'm sick of people trying to candy coat things and saying that its completely safe to have your computer on the internet, because we all damn well know its is a risk.
Actually recent versions of PINE(greater than > 4.2 maybe?) do render HTML. Of course it doesn't autoload images or anything, but HTML isn't really the problem. Its fuckwitted software that likes to automagically load everything up at once under the guise of "of course the user will want to load this image". And the truth is, your average lamebrain windows user, even if confronted by the option of turning off images from remote sites, they would say no because they don't know a better and would think it meant that they would no longer be able to receive emails with porno pics of the "woman" he has be having cybersex with in some java based chat room(never mind the fact that this "woman" is really a 500pound biker guy who has a thing for giving enemas to people).
Slashdot Mirror
Well I can agree that NetBSD would be just as useful in this role as the Linux kernel would be. I'm not suggesting that they deploy a full GNU setup on the thing, just a kernel. My point was that with investors, often times, Linux == cool and NetBSD == what is that?
See my point now?
For one, outside of the geek community who the hell knows what NetBSD is. There is a bit of appeal for companies to be associated with Linux, especially from the viewpoint of their shareholders. They will think, "hey, I've heard of this Linux thing, must be they're on the cutting edge of things." So they go out and buy another 1000 shares of the companies stock.
Besides that point, recent Linux development have been targetted towards making Linux more friendly on embedded hardware.
Personally, I rather like the fact that people are using Linux and will be giving(even if it is forced giving) their changes back to the community and increasing the general knowledge of different commputing platforms(source code is often excellent documentation on a new platform)
Obviously you never read the flame war between Tannenbaum and Linus that took place in 1991. Theoritical design is one thing, but writing something that works and works well is another.
But, if you don't like how things are going with the linux kernel, nobody is stopping you from starting your own fork of the linux kernel. Import the whole source tree into your own CVS repository, get some developers and get some work done.
Could have something to do with the fact that his wife, Tove, was pregnant and gave birth in late November I believe. I think that would take a bit more priority over tbe release of 2.4.0
You might want to take a look at reiserfs for your news spool. Reiserfs doesn't have inodes(not in the traditional sense anyways). What it does it creates a unique hash for each file and uses that for an inode, so you end up having basically an unlimited number of inodes.
Also reiserfs is not block oriented so you end up with little to no wasted disk space when dealing with small files. Also with extremely small files, the data is often stored with the metadata rather than allocating a chuck of disk space elsewhere.
On the other hand, a Christian could argue that Darwinist are teaching evolution as if it were fact.
Neither one is truly provable, as both are proven in the terms of the people attempting to prove it.
Consider that if you use scientific method to prove or disprove creationism, of course you are going to be led to disprove it, as your already biased. Even without the bias, creationism is illogical in the eyes of science.
With that being said, if you try to prove evolution within the context of Christianity, you are going to be led to disprove it as well, as you are already biased. And of course evolution is illogical in the eyes of a Christian.
Its really a mute point to argue the issue, as people are going to believe what they like, and most people are not huge fans of trying to be "converted". Next time think about that you try to shove evolution down a creationist throat. Do you want a creationist shoving creationism down your throat?
Yeah...try to compile a rather large package on your 486DX/33MHz box with 8mb of ram..see how long it takes to compile something large, like for example, emacs.
And if you say that you're compiling it yourself so you know its not backdoored, well do you check ever single line of code for backdoors? I didn't think so.
So what you're saying is that things need to be dumbed down for the lowest common demoninator of our society? Oh thats right, god forbid we educate people on critical thinking skills like logic, cause and effect, context clues etc. So I say, if we want the lowest common demonitator, why not make a "shiny things" interface. No text anywhere, because god forbid people who can't read won't be able to use it. We won't need a keyboard either, as nobody knows how to type. I'm sorry you can have that if you like, but I'll pass on the OS for masses of people requiring thorazine.
Do you remember how to think?
You know, there is nothing stopping you from going
to Washington yourself and lobbying yourself. If your congresscritter doesn't know about your concerns, how are they supposed to know how you feel?
Perhaps some organization(FSF maybe, or perhaps Software in the Public Interest, maybe they already do lobby though?), should get some people around to lobby in DC for us free software people.
Take a look at the drivers for some of the Adaptec SCSI controllers, you'll see that there is a wad of code uploaded to the controller that we have no source for either...
Its sucks, but it works..
Never mind the fact that the RTL8139 is a piece of crap. Quit being cheap and buy yourself a 3com.
How about integrated into at least the upgrade? So that it will be able to mount and upgrade on currently existing reiserfs filesystems?
I assume that since this is using a corba interface one could just as easily mount a postgresql database or anything else that is exportable via corba.
Somehow this all reminds me of how Plan9 does things, where *everything* is a file, that includes TCP connections etc, etc...
Now what would be more useful is using corba as a replacement for NFS and SUN RPC. Now that might be of interest...
A lot of mail servers allows a user to do fun things with their email address like this:
Say if a user has an email address like anonymous@slashdot.org and he wanted to track where some junk came from, so on slashdot his publishs his email like anonymous-coward@slashdot.org some mail servers(sendmail and postfix do this i think qmail does too) will get dropped in the mailbox of anonymous@slashdot.org
Its pretty simple and doesn't require a lot of screwing around with things. Of course not all mail servers are setup to do this..
I believe one can opt-out of the junk mail that the postal service delivers to everyone. I think there is a form you have to fill out though. Your best bet would be to talk to somebody at your local post office. Now this of course does nothing for the junk you get from people who have your exact address though...
The ads got killed on inclusion to 2.4.1-preX
So don't worry about the ads anymore..
This bug has been fixed.
x xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxx': File name too long
I've you actually followed reiserfs development any you would know this. The issue was the fact that reiserfs knows how to handle filenames longer that 255, but the VFS in the linux kernel does not. So, reiserfs that is in 2.4.1-pre7 limits this to 255 characters..
As proof for you tiny little mind...
mkdir "$(perl -e 'print "x" x 768')"
mkdir: cannot create directory `xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
The said filesystem is reiserfs..
You are also better off using the 3.5.x utils as reiserfs 3.6.x does not have a functional fsck yet(its in the works though). Not that you'll need it too often, but just in case.
MKLinux was a port of the Linux kernel to the Mach microkernel. Basically the linux kernel ran as a single server on the Mach microkernel. Which would really suck as you lose all the advantages of a microkernel and a monolithic kernel by doing that.
Of course Mklinux is pretty much dead with PPC Linux around now...
I think that a lot of the problem is due to XUL, take a look at Galeon to see how much faster it runs. Galeon is using the rendering engine(I think they are calling it Gecko, but I'm too lazy to check now) from Mozilla, via the gtkmozembed library.
I primarly use galeon at this point and I've been very happy with it. Almost as quick as NS4.x and a similiar memory footprint.
Heh..it is kinda funny. I know that there has been a patch or two on the linux-kernel mailing list doing zerocopy TCP, which the tcp checksumming being done on the ethernet controller(if it supports that of course). But from what I've read they have been able to max out a gigabit ethernet link with a p3@500mhz.
I don't see any of the BSDs doing that these days..
Not to mention a really dumb idea. Sure leave your door open, but don't expect anything left in the morning. Sure you *should* be able to do this, and you *should* be able to trust people. And most people can be trusted, but not everyone can.
There should be some sort of social responsiblity to keep computing equipment on the internet in a maintained, orderly fashion. If you do not want to do that, do not place it on the internet.
How about this one, sign a contract with your ISP that you either:
A. Agree to keep your equipment orderly, secure and maintained and that you agree to pay punitive damages(based on income) in a failure to do so that becomes exploited, unless you can prove that you acted in good faith maintain your equipment.
or
B. Agree to allow the ISP to filter certain types of outgoing traffic from your equipment. (Oversized or excessive ICMP packets, TCP packets with bad flags or excessive SYN packets, basically any type of data that is not normal).
These all seem reasonable to protect both the ISPs networks and the users.
Aaron
As the IDE subsystem developer for Linux, how will you deal with this misfeature? Will you merely work around it in software, or will you stay true to the SPECS and implement the copy protect feature? Or perhaps have the copy protect a CONFIG option? Or will this be a layer below the kernel(in the chipset) and in such case, hack around that too(XORing the file as it goes to disk obscuring any goofy signatures and reversing the operation on the way back?).
Aaron
So by the same token, people should be allowed to drive automobiles even if they don't know how to drive them? Or they should be allowed to use the telephone if they don't know basic telephone safety(Like don't give your address etc etc out to
strangers and all the other things you learned as a kid).
My point is that computers attached to the internet are not just "toys" but they are serious pieces of electronic equipment. Equipment that most people trust their finanical records and other aspects of their personal life to. At least software should go with conservative defaults for the uneducated. The people who do know what they are doing know how to change the defaults.
It would be fairly trival to have a bug of worm that gets into a system via a bug in outlook(or more often than not an education problem, like files named pr0n.jpg.exe etc..) and then phones home with all of the goodies to some random webserver in siberia. Oh and it installed a nice backdoor or something.
People just need to be educated about the risks, like with the box tossing up a message about loading remote images saying that "Loading images from remote servers that are received via email can be considered a privacy threat, if you know that you will not have this problem, click ok, otherwise the safe choice is no".
I'm sick of people trying to candy coat things and saying that its completely safe to have your computer on the internet, because we all damn well know its is a risk.
Actually recent versions of PINE(greater than > 4.2 maybe?) do render HTML. Of course it doesn't autoload images or anything, but HTML isn't really the problem. Its fuckwitted software that likes to automagically load everything up at once under the guise of "of course the user will want to load this image". And the truth is, your average lamebrain windows user, even if confronted by the option of turning off images from remote sites, they would say no because they don't know a better and would think it meant that they would no longer be able to receive emails with porno pics of the "woman" he has be having cybersex with in some java based chat room(never mind the fact that this "woman" is really a 500pound biker guy who has a thing for giving enemas to people).