Since then Apple engineers have kept sync with individual packages with a goal to be able to keep in step with more and more of the OS until they are A) using the latest stable branc and B) able to incorporate entire new releases with about 3-months of lag time.
I think this is the whole reason that we've not seen Apple contribute back much code: Darwin hasn't really been keeping up.
This is understandable. Apple are a good year away from actually upgrading MacOS X, and what you'll find happening is that they'll spend the next six to twelve months on performance enhancements and bug fixing. Then only will we see them trying to merge changes from FreeBSD, finding that the patches don't apply cleanly, and going through the process of feeding back changes to FreeBSD.
Given Apple's corporate culture, it's quite understandable that they didn't want to get involved in selling CD's of a free OS, and cheesy coffee mugs.;-)
Let's see, in the past while "FreeBSD" has been moved from Walnut Creek CDROM, to BSDi and then to Wind River, and now back to Walnut Creek CDROM (aka FreeBSD Mall). This has had one big influence on the design of FreeBSD, and that was the fine grained SMP locking for 5.0-CURRENT. BSDi released the source code of BSD/OS to FreeBSD developers, so that the kernel hackers could get into the niggly details of how they implemented their locking, which was known to work well on SMP systems.
Wind River had little influence, although they did a lot of work on getting the FreeBSD handbook into a state where it could be published.
The main design influences come from personal coders. The biggest recent influence on the design (remember the OS is not new) has probably been Whistle Communications, who have a number of people working on FreeBSD and have contributed a number of interesting subsystems, like NetGraph, and the kernel threading code going into 5.0-CURRENT. Yahoo! have a big influence on bug fixing, because they tend to work their boxes hard.
But on the whole, most of the code comes from individuals, although the most productive coders are those being paid to work full time. There were a number of such people at WC/BSDi/Wind River. Discussions occur on the mailing lists (mostly freebsd-arch).
Bandwidth has been a problem with the moves, because the main FreeBSD servers were on Walnut Creek CDROM servers. When Wind River bought BSDi, they didn't buy the CDROM bussiness, which lead to the downfall of ftp.freesoftware.com (aka ftp.freebsd.org), which used to be ftp.cdrom.com. Hopefully this change will lead to that coming back... ftp.freebsd.org is currently hosted in the Netherlands.
The problem now is that because of the 'official sponsor' vacumn created by Wind River when they announced that they were dropping FreeBSD, two other groups stepped forward. The first was the FreeBSD Foundation, which is a non profit established to fund FreeBSD (see recent announcements concerning Java for FreeBSD), and the second is Daemon News, who recently announced their own FreeBSD CD distribution, via "BSD Mall". Confused yet? How big a problem this is is yet to be seen, but anyway, the only real reason for CD's is to give them to Linux users...;-)
FreeBSD has always been a one stop OS, and this is going to confuse things. If you're a Linux geek, try FreeBSD. You'll find that a one stop OS is nice; you don't have to hunt for patches, or wait till your distribution gets the latest kernel, or worry about matching glibc with your kernel... With FreeBSD you decide if you want to run -STABLE or -CURRENT, and you just track it. The only time you have to worry about versions is with external packages.
I don't know about the new 4.5 CD's but the 4.4-RELEASE CD's say "This CDROM may be duplicated and redistributed" right under the copyright notice on the disc.
The 3.2 CD's don't have this, so I don't know when this was changed. Probably at the same time the project begun releasing ISO images on the FTP site.
I think you need to look a lot more carefully at Mark Shuttleworth. While his motives aren't altruistic (if I was an unemployed, ex.com'er, and had $575mil in my pocket, I'd also be going into space), he nearly didn't get to go because he insisted on being allowed to conduct scientific experiments, and held a competition in South Africa to find the best experients. NASA's primary complaint is that they didn't want him doing experiements... It was OK if he just sat there and enjoyed the view.
I'd hardly call his trip wasting the precious US investment (which would have been a lot less if the Russians had built it, because they don't squander money like NASA).
The BSD software world works kind of in reverse of the Linux world... In Linux
you have a central kernel, which is pretty much common across all distributions.
From here the big distributions tack on an installer of their choice and a mix of
software which they think makes up 'an operating system'. From here, other, smaller,
distributions tweak that set of packages until they have something which they consider
a workable system. Then from there the user is free to customize, although they are
normally locked into the package format of the base distribution (apt, rpm, etc.) unless
they build by hand.
In the BSD world, you have the ports/packages collection, which builds packages
from source for whatever you are running. While at the moment these are similar but
not quite the same, there is quite a lot of momentum for the Open Packages which is attempting to
unify all of these collections. Since these compile from source, they adapt to the
operating system underneath.
The three BSDs share a lot of the utilities which make up an operating system.
Like the compilers (which come from GNU), OpenSSH and many of the utilities (ls,
ppp, etc.) are either maintained across all of the BSDs by the same maintainer(s) or are
regularly merged.
Where the main differences lie is in the kernels and in the approach to the core
OS. OpenBSD and NetBSD are fairly similar in their kernels, an FreeBSD is quite
different (especially 5.x). This also where you see the philosophies of the
developers... FreeBSD tries to be as fast as possible on commodity hardware.
NetBSD tries to run on as much hardware as possible, and OpenBSD tries to be as
secure as possible.
There is a fairly continual cross pollination of ideas between the BSDs: FreeBSD
gets security fixes from OpenBSD, or code to port to other platforms rom NetBSD.
They get performance enhancements like softupdates.
What is common between the BSDs is the approach. There are two rules from
BSD development:
POLA: the Principal of Least Astonishment. Whatever the OS does must be the
least astonishing thing. So when a new version comes out it shouldn't require you to
reformat your disks...
Do The Right Thing: The code must not only work, it must do the right thing.
So it's not good enough to hack up a firewire driver which just reads video from one
camera - the driver must be designed to abstract the firewire bus, so disks or
networking could work too.
It is these two philosophies which make Linux people think that the BSDs are
bigoted. The BSDs reject a bunch of Linux's ideas because they are hacked together,
and seem to work by magic.
Merging the BSDs would be a big mistake, because it would mean that there would
not be room for differing ideas.
FreeBSD is still very much alive, and development of both 4-STABLE and -CURRENT continue as ever. AFAIK, none of the people being laid off are core team members or even really active source developers. Most of them work(ed) on documentation and improving the FreeBSD product line (CD box sets, etc.).
At the moment the release date for 5.x has been pushed out until late next year, partly because we've lost a few developers to real work, but also because we bit off more than we could chew... Rewriting the kernel for preemptive fine grain threading is a big task.
Other aspects of the project continue to be very actively developed. The Ports collection is almost at 6000 ports.
It is really sad to see people laid off, but this is just a side effect of the dotcom crash.
HP-UX will still be the only non-Linux Unix shipping on Itanium when McKinley rolls around
The FreeBSD port is coming along to... The latest commit messages suggest that it is booting in the emulator and close to booting on real hardare. Obviously there's a lot of work still to be done, but judging from work on the Alpha port, booting is the biggest hurdle...
To any of the HP-UX developers reading this, there's plenty of work in FreeBSD's kernel land. Year's of Unix experience and a love of clean design are definite bonuses.
Afraid you're wrong. It's a troll from end to end, and not even a very good one...
As a core consultant developer for the *BSD kernel for 6 months last year
No one by this name is involved in any BSD, certainly not at the core level. Also, there are no core team members in common between Free, Net and OpenBSD.
There are many issues which have not been resolved and are not being publicized to the public.
The FreeBSD project does all of it's development in public mailing lists...
1) The implementation of threads still uses fine grain kernel level locking
The 4.x kernel does not have fine grained locking, this is being developed in 5.x.
which does not adhere to POSIXX IEEE 811.2b level requirements
POSIX is an IEEE OS standard.
IEEE 801.11b is an IEEE wireless networking standard. A seasoned kernel hacker would know the difference...
certified for level 4 security.
There is no such things. Secure systems conform to data books such as the 'orange book'.
The hash implementation which was used for prior backdoor's still exists and the modules which access it have not been auditied by third party engineers. This is a serious security violation which the dev team refuses to address. In fact they are doing all they can to sweep it underground, hoping people will just forget about it.
4.4 uses a IETF standard algorithm for sequence number generation (hash algorithms cant be backdoored), and this replaces the algorithm in 4.3 and earlier versions which did have a problem with sequence number guessing.
There is still no credible evidence that the new implementation of the TCP/IP stack is an improvement over the broken one they are trying to replace from the 4.3.xx series.
The TCP/IP stack in 4.4 is the same as in 4.3 (there was no 4.3.xx) and is the best performing TCP/IP stack around (even compared to the new Linux stack).
and the potential for data loss was rated as QQQ on the topenhiemer algorithm.
The stack does not loose data, and there is no such thing as a topenhiemer algorithm to rate it as a QQQ.
I am currently petitioning the core dev team to remove my code from the project due to my differences with them
No one is currently petitioning the FreeBSD core team to remove any code.
For some reason all sorts of people crawl out of the woodwork, and begin trolling on FreeBSD stories. Normally, like this post, they've read the last few news items from the FreeBSD web site, managed to store a few terms, and then try to put them into some or other attack on FreeBSD...
The FreeBSD ports system is maintained seperately to the OS itself, and so you can generally install what ports you want. For popular software, the ports are normally updated within about one week of the release of a new version, although this varies heavily, epecially if the new version has some problems on FreeBSD.
For a lot of ports, you'll find that there are two versions in the ports tree, a "stable" version and a "devel" version. For example, the stable version of Apache is currently 1.3.20, and the devel version 2.0.16.
If your machine is slow, then you can install packages. These are built fairly frequently for the -STABLE branch, and can be found at http://www.freebsd.org/ports/. Or you can use 'pkg_add -r apache' (for example), which will fetch the latest stable package for apache and install it.
Hope this helps. If you have more questions, then try reading Chapter 4 of the FreeBSD Handbook (linked in the story above).
Not a drawing program, but one that I'm sure will find very wide use is GNUPlot. The CVS versions (http://sourceforge.net/projects/gnuplot/) have a SVG terminal. I've not tried it yet, but I'm hoping to use it in a project where I need high quality graphs on a web page. I noticed that in the latest GhostScript/GSView under windows that one of the conversion options is PS->SVG via the GNUPlot driver...
These patents mention a lot of stuff about a 'valid user'. I'm not a Palm owner (I'll take the pen and paper method anyday), I've never seen a Palm try to validate the user... You just switch it on. This alone would invalidate a large number of the claims.
They didn't even have to look at the numbers... A 39% drop in total sales of music would have meant we'd be seeing thousands of people out of work, empty shops, going out of bussiness sales, etc.
Your analysis of the numbers is still lacking however, because you are comparing this years figures with last years. To be accurate you should compare the 2000 figures, with 2000 figures without the influence of Napster. This is, however, not possible, so the best course would be to compare 2000 figures with projections for 2000, based on inflation and total growth in the US economy. What you'll probably find is that inflation was about 4% on luxury consumer items, but that sales of luxury items are down 6% from last year (because of the general slowdown in the US economy - espcially in the last quater). As a result the 2% drop in total sales was most likely purely in response to people having less disposable income around Christmas.
Judging from most people's reactions, I doubt Napster is having any effect at all on total music sales. It's causing some people to buy more and other to buy less. All that Napster is doing is changing how the game is played. But we already knew that, and we know that is is the real reason for the RIAA lawsuit...
1) Do you ever plan on moving away from the slow and resource intensive method of VMS style paging for memory address resolution
FreeBSD's paging code is extremely fast, which is why FreeBSD performs so well under load. It is fairly resource intensive, but the requirements for page tables etc are proportional to your RAM size, so FreeBSD will still run in low memory configurations.
2) Are there plans to rewrite the TCP/IP stack to be multi threaded
Once again, this is a buzz word issue - the TCP/IP stack performance is very good (ie can staturate whatever network you happen to plug in). But the entire kernel is being multi-threaded for 5.0, to provide fine grained SMP support.
3) Will BSD ever migrate away from UFS to a more modern file system?
The UFS file system is being continously upgraded. It has features which Linux and most other commercial FSs would love - like softupdates, and new utilities to grow filesystems (and shink them too hopefully soon). Just because Linux has had to rewrite it's FS because of poor reliability doesn't mean that the BSDs have a bad file system.
4) With serious POSIX compatablity issues are there plans to use code from POSIX compliant OS's to become more commercially attractive to major corporations
POSIX compatibility is also something which is always being improved. But I think that you're wrong about POSIX compatibility being an issue for major corporations. They are far more concerned with stable APIs, and at the moment they want stable APIs for things like windowing services. This is why people code for Windows, not POSIX compliance.
Actually, FreeBSD has been so successful that BSDi (the makers of BSD/OS - the commercial BSD) have made big investments in FreeBSD and have opened their entire source tree to FreeBSD developers.
FreeBSD continues to develop at a rate at least equal to Linux, and often beats Linux on support for new technologies (especially in stable releases). Examples from this release are ATA-100, USB, free RSA (OpenSSL and Open SSH).
I'm not sure which director you are trying to refer to... Since you're posting anonymously, why not name the person.
This is not something "new" for FreeBSD. For some time the base system has included RSA based encryption (for OpenSSH). What is new is that there is now only one distribution, not a USA version and an International version.
What occurred in the past was that the RSA code could come from two sources - a USA patented version, which required a licence for commercial use, and an international free version (which was also cleaner and faster). If you were a USA resident you were required to install the librsa port to obtain RSA based encryption.
Since the changes in the patent, there is now no need for the RSA Data Security library, and so the international library is used in all cases, and we now longer have to have two seperate distributions, and all of the Makefile goop to handle having two slightly different libs for USA/non USA.
It also means that RSA can be used commercially without a licence.
Regards,
-Jeremy (reg@FreeBSD.org)
Re:Are their not legal issues to OpenSSH/SSL???
on
FreeBSD 4.0 Released
·
· Score: 1
FreeBSD has always had two sets of crypto. One for the USA and one for international users. Typically the restricted code is not part of the release, and you need to specifically say that you are inside/outside the USA and the installer will fetch the right code.
For 4.0, with the integration of OpenSSL, the issue is not so much US export restrictions. It is the patent restrictions on RSA within the US. The solution is to have three possible states: No RSA, RSARef (only legal in the US and then only for noncommercial use or with a licence) and the international RSA code. 4.0 will dynamically link in RSA if you have the correct version, and print out an error message if you try to use RSA without the libraries installed.
You mean something like this: From: drs@netcom.com (Data Rentals and Sales) Subject: Re: Please help me translate this to... Keywords: COBOL C translate help plea Date: Mon, 30 Aug 1993 18:28:59 GMT In article rstout@netcom.com (Rick Stout) writes: >...and for my chapter on >programming languages I need examples of a simple program in >each of the most common languages. > >Well, I had no problem with C and C++ (being a C programmer trying >to learn C++). People from other comp.lang groups helped me with >Pascal, FORTRAN, and BASIC, but as I said, nobody reads alt.cobol. If you want a sample program in multiple languages, try this one: ----------------8_$$; echo "hello polyglots"; rm _$$; exit print C stop run. -*, 'hello polyglots' C C print. C display "hello polyglots". ( C */ int i;/* C */ main () {/* C */ i=printf ("hello polyglots\n");/* C *) (* C *) begin (* C *) writeln ('hello polyglots'); (* C *) (* ) C * ) pop 60 360 ( C * ) pop moveto (hello polyglots) show ( C * ) pop showpage (( C *) end.(* ) C)pop% program polyglot. *){*/} ----------------8 _ Kevin D Quitt 96.37% of all statistics are made up
Also included are: - Better USB support, including support for USB network adapters. - IPv6, including IPsec, an IPv6 firewall and IPv6 enabled netowrking tools. - OpenSSL now part of base system. - and a bunch more little things...
A panel discussion was held at the recent Transportation Research Board meeting in Washington DC (Concurrent session No. 414). Unfortunately notes from panel discussions are seldom published. http://www.nationalacademies.org/trb/ftp/am/All_Se ssions.pdf
For those worried about their cars being tracked, I suggest starting at:
http://www.its.dot.gov/
Intelligent Transportation Systems are currently a multi billion dollar area of research. With any luck, in some years time you wont be able to get a speeding ticket, because you wont be driving. It will be too dangerous for humans to drive on the highways of the future - they don't have dependable reactions.
cd/usr/ports/x11/gnome && make install clean; some day...
-Jeremy
Re:Let's have more integration between *BSD and Li
on
Intel using FreeBSD
·
· Score: 1
GNOME wasn't that bad. Not nearly as bad as some of the Sun/Solaris software I've looked at. Quite a few patches around bad or missing #defines and #includes, a few problems with threads. We complicated our lives a bit by trying to get it all back into a BSD rather than a SysV type directory layout.
I spent a lot of time on getting the two CD players to work, since they used Linux style ioctl's, and I hacked around some termcap/terminfo stuff in the PPP utils to get it to compile (no one's ever complained that it doesn't work, but I suspect that's because no one's ever tried it;)
It's a few days out of date (No 3.3.5 yet), but I'm sure the avid gamers will figure it out... Or look at http://glx.on.openprojects.net/
Like most third party software, the source code used for FreeBSD (the other BSD's) and Linux is the same, and so the feature set is pretty much the same.
Slashdot Mirror
Since then Apple engineers have kept sync with individual packages with a goal to be able to keep in step with more and more of the OS until they are A) using the latest stable branc and B) able to incorporate entire new releases with about 3-months of lag time.
I think this is the whole reason that we've not seen Apple contribute back much code: Darwin hasn't really been keeping up.
This is understandable. Apple are a good year away from actually upgrading MacOS X, and what you'll find happening is that they'll spend the next six to twelve months on performance enhancements and bug fixing. Then only will we see them trying to merge changes from FreeBSD, finding that the patches don't apply cleanly, and going through the process of feeding back changes to FreeBSD.
Given Apple's corporate culture, it's quite understandable that they didn't want to get involved in selling CD's of a free OS, and cheesy coffee mugs. ;-)
Regards,
-Jeremy
Let's see, in the past while "FreeBSD" has been moved from Walnut Creek CDROM, to BSDi and then to Wind River, and now back to Walnut Creek CDROM (aka FreeBSD Mall). This has had one big influence on the design of FreeBSD, and that was the fine grained SMP locking for 5.0-CURRENT. BSDi released the source code of BSD/OS to FreeBSD developers, so that the kernel hackers could get into the niggly details of how they implemented their locking, which was known to work well on SMP systems.
Wind River had little influence, although they did a lot of work on getting the FreeBSD handbook into a state where it could be published.
The main design influences come from personal coders. The biggest recent influence on the design (remember the OS is not new) has probably been Whistle Communications, who have a number of people working on FreeBSD and have contributed a number of interesting subsystems, like NetGraph, and the kernel threading code going into 5.0-CURRENT. Yahoo! have a big influence on bug fixing, because they tend to work their boxes hard.
But on the whole, most of the code comes from individuals, although the most productive coders are those being paid to work full time. There were a number of such people at WC/BSDi/Wind River. Discussions occur on the mailing lists (mostly freebsd-arch).
Bandwidth has been a problem with the moves, because the main FreeBSD servers were on Walnut Creek CDROM servers. When Wind River bought BSDi, they didn't buy the CDROM bussiness, which lead to the downfall of ftp.freesoftware.com (aka ftp.freebsd.org), which used to be ftp.cdrom.com. Hopefully this change will lead to that coming back... ftp.freebsd.org is currently hosted in the Netherlands.
The problem now is that because of the 'official sponsor' vacumn created by Wind River when they announced that they were dropping FreeBSD, two other groups stepped forward. The first was the FreeBSD Foundation, which is a non profit established to fund FreeBSD (see recent announcements concerning Java for FreeBSD), and the second is Daemon News, who recently announced their own FreeBSD CD distribution, via "BSD Mall". Confused yet? How big a problem this is is yet to be seen, but anyway, the only real reason for CD's is to give them to Linux users... ;-)
FreeBSD has always been a one stop OS, and this is going to confuse things. If you're a Linux geek, try FreeBSD. You'll find that a one stop OS is nice; you don't have to hunt for patches, or wait till your distribution gets the latest kernel, or worry about matching glibc with your kernel... With FreeBSD you decide if you want to run -STABLE or -CURRENT, and you just track it. The only time you have to worry about versions is with external packages.
Regards,
-Jeremy
I don't know about the new 4.5 CD's but the 4.4-RELEASE CD's say "This CDROM may be duplicated and redistributed" right under the copyright notice on the disc.
The 3.2 CD's don't have this, so I don't know when this was changed. Probably at the same time the project begun releasing ISO images on the FTP site.
Regards,
-Jeremy
If you can't wait, then here's a recent online chat with him, from a South African TV show.
Regards,
-Jeremy
I think you need to look a lot more carefully at Mark Shuttleworth. While his motives aren't altruistic (if I was an unemployed, ex .com'er, and had $575mil in my pocket, I'd also be going into space), he nearly didn't get to go because he insisted on being allowed to conduct scientific experiments, and held a competition in South Africa to find the best experients. NASA's primary complaint is that they didn't want him doing experiements... It was OK if he just sat there and enjoyed the view.
I'd hardly call his trip wasting the precious US investment (which would have been a lot less if the Russians had built it, because they don't squander money like NASA).
Regards,
-Jeremy
The BSD software world works kind of in reverse of the Linux world... In Linux you have a central kernel, which is pretty much common across all distributions. From here the big distributions tack on an installer of their choice and a mix of software which they think makes up 'an operating system'. From here, other, smaller, distributions tweak that set of packages until they have something which they consider a workable system. Then from there the user is free to customize, although they are normally locked into the package format of the base distribution (apt, rpm, etc.) unless they build by hand.
In the BSD world, you have the ports/packages collection, which builds packages from source for whatever you are running. While at the moment these are similar but not quite the same, there is quite a lot of momentum for the Open Packages which is attempting to unify all of these collections. Since these compile from source, they adapt to the operating system underneath.
The three BSDs share a lot of the utilities which make up an operating system. Like the compilers (which come from GNU), OpenSSH and many of the utilities (ls, ppp, etc.) are either maintained across all of the BSDs by the same maintainer(s) or are regularly merged.
Where the main differences lie is in the kernels and in the approach to the core OS. OpenBSD and NetBSD are fairly similar in their kernels, an FreeBSD is quite different (especially 5.x). This also where you see the philosophies of the developers... FreeBSD tries to be as fast as possible on commodity hardware. NetBSD tries to run on as much hardware as possible, and OpenBSD tries to be as secure as possible.
There is a fairly continual cross pollination of ideas between the BSDs: FreeBSD gets security fixes from OpenBSD, or code to port to other platforms rom NetBSD. They get performance enhancements like softupdates.
What is common between the BSDs is the approach. There are two rules from BSD development:
It is these two philosophies which make Linux people think that the BSDs are bigoted. The BSDs reject a bunch of Linux's ideas because they are hacked together, and seem to work by magic.
Merging the BSDs would be a big mistake, because it would mean that there would not be room for differing ideas.
Regards,
-Jeremy
Hmm, no.
FreeBSD is still very much alive, and development of both 4-STABLE and -CURRENT continue as ever. AFAIK, none of the people being laid off are core team members or even really active source developers. Most of them work(ed) on documentation and improving the FreeBSD product line (CD box sets, etc.).
At the moment the release date for 5.x has been pushed out until late next year, partly because we've lost a few developers to real work, but also because we bit off more than we could chew... Rewriting the kernel for preemptive fine grain threading is a big task.
Other aspects of the project continue to be very actively developed. The Ports collection is almost at 6000 ports.
It is really sad to see people laid off, but this is just a side effect of the dotcom crash.
Regards,
-Jeremy
HP-UX will still be the only non-Linux Unix shipping on Itanium when McKinley rolls around
The FreeBSD port is coming along to... The latest commit messages suggest that it is booting in the emulator and close to booting on real hardare. Obviously there's a lot of work still to be done, but judging from work on the Alpha port, booting is the biggest hurdle...
To any of the HP-UX developers reading this, there's plenty of work in FreeBSD's kernel land. Year's of Unix experience and a love of clean design are definite bonuses.
Regards,
-Jeremy
Afraid you're wrong. It's a troll from end to end, and not even a very good one...
As a core consultant developer for the *BSD kernel for 6 months last year
No one by this name is involved in any BSD, certainly not at the core level. Also, there are no core team members in common between Free, Net and OpenBSD.
There are many issues which have not been resolved and are not being publicized to the public.
The FreeBSD project does all of it's development in public mailing lists...
1) The implementation of threads still uses fine grain kernel level locking
The 4.x kernel does not have fine grained locking, this is being developed in 5.x.
which does not adhere to POSIXX IEEE 811.2b level requirements
POSIX is an IEEE OS standard.
IEEE 801.11b is an IEEE wireless networking standard.
A seasoned kernel hacker would know the difference...
certified for level 4 security.
There is no such things. Secure systems conform to data books such as the 'orange book'.
The hash implementation which was used for prior backdoor's still exists and the modules which access it have not been auditied by third party engineers. This is a serious security violation which the dev team refuses to address. In fact they are doing all they can to sweep it underground, hoping people will just forget about it.
4.4 uses a IETF standard algorithm for sequence number generation (hash algorithms cant be backdoored), and this replaces the algorithm in 4.3 and earlier versions which did have a problem with sequence number guessing.
There is still no credible evidence that the new implementation of the TCP/IP stack is an improvement over the broken one they are trying to replace from the 4.3.xx series.
The TCP/IP stack in 4.4 is the same as in 4.3 (there was no 4.3.xx) and is the best performing TCP/IP stack around (even compared to the new Linux stack).
and the potential for data loss was rated as QQQ on the topenhiemer algorithm.
The stack does not loose data, and there is no such thing as a topenhiemer algorithm to rate it as a QQQ.
I am currently petitioning the core dev team to remove my code from the project due to my differences with them
No one is currently petitioning the FreeBSD core team to remove any code.
For some reason all sorts of people crawl out of the woodwork, and begin trolling on FreeBSD stories. Normally, like this post, they've read the last few news items from the FreeBSD web site, managed to store a few terms, and then try to put them into some or other attack on FreeBSD...
Regards,
-Jeremy
The FreeBSD ports system is maintained seperately to the OS itself, and so you can generally install what ports you want. For popular software, the ports are normally updated within about one week of the release of a new version, although this varies heavily, epecially if the new version has some problems on FreeBSD.
For a lot of ports, you'll find that there are two versions in the ports tree, a "stable" version and a "devel" version. For example, the stable version of Apache is currently 1.3.20, and the devel version 2.0.16.
If your machine is slow, then you can install packages. These are built fairly frequently for the -STABLE branch, and can be found at http://www.freebsd.org/ports/. Or you can use 'pkg_add -r apache' (for example), which will fetch the latest stable package for apache and install it.
Hope this helps. If you have more questions, then try reading Chapter 4 of the FreeBSD Handbook (linked in the story above).
Regards,
-Jeremy
Regards,
-Jeremy
These patents mention a lot of stuff about a 'valid user'. I'm not a Palm owner (I'll take the pen and paper method anyday), I've never seen a Palm try to validate the user... You just switch it on. This alone would invalidate a large number of the claims.
Regards,
-Jeremy
They didn't even have to look at the numbers... A 39% drop in total sales of music would have meant we'd be seeing thousands of people out of work, empty shops, going out of bussiness sales, etc.
Your analysis of the numbers is still lacking however, because you are comparing this years figures with last years. To be accurate you should compare the 2000 figures, with 2000 figures without the influence of Napster. This is, however, not possible, so the best course would be to compare 2000 figures with projections for 2000, based on inflation and total growth in the US economy. What you'll probably find is that inflation was about 4% on luxury consumer items, but that sales of luxury items are down 6% from last year (because of the general slowdown in the US economy - espcially in the last quater). As a result the 2% drop in total sales was most likely purely in response to people having less disposable income around Christmas.
Judging from most people's reactions, I doubt Napster is having any effect at all on total music sales. It's causing some people to buy more and other to buy less. All that Napster is doing is changing how the game is played. But we already knew that, and we know that is is the real reason for the RIAA lawsuit...
-Jeremy
Only important questions if you are trolling...
1) Do you ever plan on moving away from the slow and resource intensive method of VMS style paging for memory address resolution
FreeBSD's paging code is extremely fast, which is why FreeBSD performs so well under load. It is fairly resource intensive, but the requirements for page tables etc are proportional to your RAM size, so FreeBSD will still run in low memory configurations.
2) Are there plans to rewrite the TCP/IP stack to be multi threaded
Once again, this is a buzz word issue - the TCP/IP stack performance is very good (ie can staturate whatever network you happen to plug in). But the entire kernel is being multi-threaded for 5.0, to provide fine grained SMP support.
3) Will BSD ever migrate away from UFS to a more modern file system?
The UFS file system is being continously upgraded. It has features which Linux and most other commercial FSs would love - like softupdates, and new utilities to grow filesystems (and shink them too hopefully soon). Just because Linux has had to rewrite it's FS because of poor reliability doesn't mean that the BSDs have a bad file system.
4) With serious POSIX compatablity issues are there plans to use code from POSIX compliant OS's to become more commercially attractive to major corporations
POSIX compatibility is also something which is always being improved. But I think that you're wrong about POSIX compatibility being an issue for major corporations. They are far more concerned with stable APIs, and at the moment they want stable APIs for things like windowing services. This is why people code for Windows, not POSIX compliance.
Regards,
-Jeremy
This site has some interesting looking lights. Especially the 7 LED custom Petzl headlamp, which is unfortunately (1) expensive, and (2) out of stock.
I don't own any of their products, but was thinking of buying one of the various headlamps.
Regards
-Jeremy
This is complete FUD!
Actually, FreeBSD has been so successful that BSDi (the makers of BSD/OS - the commercial BSD) have made big investments in FreeBSD and have opened their entire source tree to FreeBSD developers.
FreeBSD continues to develop at a rate at least equal to Linux, and often beats Linux on support for new technologies (especially in stable releases). Examples from this release are ATA-100, USB, free RSA (OpenSSL and Open SSH).
I'm not sure which director you are trying to refer to... Since you're posting anonymously, why not name the person.
-Jeremy
For information on the "ring of fire" see:
http://www.africaone.com/
They plan an 80Gb cable loop around Africa by 2003.
For general information on the Internet in Africa, the best source of information is:
http://www3.sn.apc.org/africa/index.htm l
-Jeremy
This is not something "new" for FreeBSD. For some time the base system has included RSA based encryption (for OpenSSH). What is new is that there is now only one distribution, not a USA version and an International version.
What occurred in the past was that the RSA code could come from two sources - a USA patented version, which required a licence for commercial use, and an international free version (which was also cleaner and faster). If you were a USA resident you were required to install the librsa port to obtain RSA based encryption.
Since the changes in the patent, there is now no need for the RSA Data Security library, and so the international library is used in all cases, and we now longer have to have two seperate distributions, and all of the Makefile goop to handle having two slightly different libs for USA/non USA.
It also means that RSA can be used commercially without a licence.
Regards,
-Jeremy (reg@FreeBSD.org)
FreeBSD has always had two sets of crypto. One for the USA and one for international users. Typically the restricted code is not part of the release, and you need to specifically say that you are inside/outside the USA and the installer will fetch the right code.
For 4.0, with the integration of OpenSSL, the issue is not so much US export restrictions. It is the patent restrictions on RSA within the US. The solution is to have three possible states: No RSA, RSARef (only legal in the US and then only for noncommercial use or with a licence) and the international RSA code. 4.0 will dynamically link in RSA if you have the correct version, and print out an error message if you try to use RSA without the libraries installed.
-Jeremy
You mean something like this: From: drs@netcom.com (Data Rentals and Sales) Subject: Re: Please help me translate this to... Keywords: COBOL C translate help plea Date: Mon, 30 Aug 1993 18:28:59 GMT In article rstout@netcom.com (Rick Stout) writes: >...and for my chapter on >programming languages I need examples of a simple program in >each of the most common languages. > >Well, I had no problem with C and C++ (being a C programmer trying >to learn C++). People from other comp.lang groups helped me with >Pascal, FORTRAN, and BASIC, but as I said, nobody reads alt.cobol. If you want a sample program in multiple languages, try this one: ----------------8_$$; echo "hello polyglots"; rm _$$; exit print C stop run. -*, 'hello polyglots' C C print. C display "hello polyglots". ( C */ int i; /* C */ main () { /* C */ i=printf ("hello polyglots\n"); /* C *) (* C *) begin (* C *) writeln ('hello polyglots'); (* C *) (* ) C * ) pop 60 360 ( C * ) pop moveto (hello polyglots) show ( C * ) pop showpage (( C *) end .(* ) C)pop% program polyglot. *){*/} ----------------8 _ Kevin D Quitt 96.37% of all statistics are made up
Nice list, but it's a bit old.
The compiler is GCC 2.95.2...
Also included are:
- Better USB support, including support for USB network adapters.
- IPv6, including IPsec, an IPv6 firewall and IPv6 enabled netowrking tools.
- OpenSSL now part of base system.
- and a bunch more little things...
-Jeremy
This is nothing really new. The technology for the location of cellular phones has been around since they were first implemented.
e /1zs01!.pdf
e ssions.pdf
A pilot project to asses the viability of doing this in Washington DC was underway in 1994.
http://www.itsdocs.fhwa.dot.gov/jpodocs/repts_t
A panel discussion was held at the recent Transportation Research Board meeting in Washington DC (Concurrent session No. 414). Unfortunately notes from panel discussions are seldom published.
http://www.nationalacademies.org/trb/ftp/am/All_S
For those worried about their cars being tracked, I suggest starting at:
http://www.its.dot.gov/
Intelligent Transportation Systems are currently a multi billion dollar area of research. With any luck, in some years time you wont be able to get a speeding ticket, because you wont be driving. It will be too dangerous for humans to drive on the highways of the future - they don't have dependable reactions.
Regards,
-Jeremy
GNOME not ported?
/usr/ports/x11/gnome && make install clean; some day...
cd
-Jeremy
GNOME wasn't that bad. Not nearly as bad as some of the Sun/Solaris software I've looked at. Quite a few patches around bad or missing #defines and #includes, a few problems with threads. We complicated our lives a bit by trying to get it all back into a BSD rather than a SysV type directory layout.
;)
I spent a lot of time on getting the two CD players to work, since they used Linux style ioctl's, and I hacked around some termcap/terminfo stuff in the PPP utils to get it to compile (no one's ever complained that it doesn't work, but I suspect that's because no one's ever tried it
-Jeremy
Try http://www.freebsd.org/~3d/
It's a few days out of date (No 3.3.5 yet), but I'm sure the avid gamers will figure it out... Or look at http://glx.on.openprojects.net/
Like most third party software, the source code used for FreeBSD (the other BSD's) and Linux is the same, and so the feature set is pretty much the same.
Regards,
-Jeremy