OpenBSD's "only one hole in over 7 years" claim is made using some very, umm, "non-standard" definitions. The last time I checked, using the same definitions FreeBSD could actually claim a longer period without any security holes.
Of course, this is a completely meaningless (and misleading) statistic since both FreeBSD and OpenBSD have had numerous root holes in the default installation published and fixed.
I bet there were some juicy tidbits to be discovered by a sniffer on that wireless network! I wonder how many passwords and email correspondence were captured by the intelligence agencies of the world present at that conference.
Yes, physics people publish in journals (after first publishing the e-print on xxx.arxiv.org). Yes, physics people cite journal articles (together with the arxiv reference). And yes, non-physics people read journals (because they are 5-10 years behind the trend towards e-publishing). But physics people don't read articles in paper journals, instead they read the e-print version on xxx.arxiv.org.
As a theoretical physicist myself, I think I am speaking with some authority here:)
The other highly suspcious aspect of this whole affair is that the work was never published online prior to being submitted to a journal.
Some background for non-physicists: thesedays the primary venue for publishing new works is the arXiv. Several hundred papers per day are uploaded here in various categories, and it is the de facto standard library of modern research in physics.
After publishing your work on the arXiv, physicists around the work can and will read your paper and submit feedback. Typically, after publication on the arXiv you might submit your work to a paper-based journal, but this is only a secondary procedure, and the only real point is to give you bonus points for your resume.
Here is the main point:
No-one reads paper journals any more!
The fact that the Bogdanov papers were never uploaded to the arxiv meant that apart from the 2 referees (who basically seem to have abdicated responsibility), no-one had, or ever would have read their work!.
If the authors were serious researchers, they would have submitted their work to the arxiv so it could be read and critiqued by their peers.
Undergraduate physics education is the USA is a travesty; however fortunately this is not true in other countries. This is presumably why the vast majority of PhD students in physics in the USA are foreign students. This is especially true in theoretical physics: most american students just cannot compete with their foreign counterparts.
There _are_ universities with good undergraduate physics programs (and good students coming out of the bad universities), but they certainly seem to be in the minority.
No, but to people who do understand the field in which the Bogdanov brothers claim to be researching (I am a PhD student in string theory and include myself in this set), their paper is complete gibberish.
Their paper is full of unfounded assertions strung together, combined with definitions and other assertions that are patently false. The paper does not follow a logical chain of reasoning that allows the reader to repeat or verify their conclusions (which are also not clearly stated).
Furthermore, when confronted by other theoretical physicists (on the sci.physics.research Usenet group) with specific, detailed questions about their work, the authors have systematically refused to answer, or selectively answered with further vague or absurd statements. To me, this is the real clincher: they have completely failed to demonstrate a technical understanding of the field in which they claim to be working.
It is clear to everyone in the field that these papers are nonsense and should never have been published. The only actual supporters of the authors seem to be from non-scientists or physicists who are unqualified to judge the work itself.
Your expectations are misguided and unfair. I don't know of ANY free software project that provides support for arbitrarily old releases -- unless you pay for it. Conversely, if you're willing to pay for support then you can surely find someone to support OpenBSD for you.
Hmm, that seems a bit bass-ackwards: you're on a much slower connection, and you want to download a 700M ISO image instead of ~100M of gzipped files (for a base install)?
You're much better off downloading the distribution sets you want to install to a local source (NFS, MSDOSFS, etc), since you'll only be downloading what you actually want to install, and not all the extra crap on the iso.
No, you're confusing things. FreeBSD added the 48-bit block addressing *128PB) capabilities a month ago, for which the Linux patch has just been announced.
The ability to address 48-bit blocks on a disk has nothing to do with the size of filesystems which can be constructed on that disk, or the files which can be created within that filesystem. This is a concept known as 'layering':-)
Well, actually it's not file sizes; FreeBSD has had 64-bit file size forever, and Linux has that too thesedays I believe. That's a completely separate issue to the ATA driver being able to address 48-bit physical media, which is what the support is for. And actually the Linux ATA driver maintainer only released these patches a few days ago (and they're not yet integrated into any of the kernel distributions), so they are much more recent than FreeBSD's support.
File sizes
Filesystem sizes
Partition sizes
Physical media sizes {-- this is what we're talking about: the limitations on these four things are more or less independent
I read the message, and this seems almost identical to the jail(8) functionality which FreeBSD has had for a number of years. Jail is very very useful for a number of purposes, but this doesn't seem to be anything new, just a "feature catch-up".
Well, it's good to see that Linux has caught up, but the article is not correct that Linux is the first OS to support 48-bit ATA; FreeBSD has had this support for over a month now.
See for example:
this file which is one of the files containing the ATA-6r2 code, committed to FreeBSD on October 6.
I got two FreeBSD CD sets delivered today from DN and they each came with a 'BSD' car sticker similar to the country-of-registration stickers you see in europe. Kinda cool:)
Jordan Hubbard is *not* the "lead programmer". He's an evangelist and does very little coding.
You've also misunderstood the reason for pushing back 5.0-RELEASE; the changes being made were just too big, and the system isn't ready for a release yet. We could have released an unstable, crap 5.0-RELEASE now, or we can wait and get it right.
#1) Jordan is as active as he's ever been in FreeBSD despite the move in his daytime job to Apple. He's even paid to work on FreeBSD during the week. The fact of the matter is that Jordan isn't an active developer anyway, he's more of a publicist and manager. If you're choosing to move away from FreeBSD because you think the PR spokesman is no longer dedicated to his job, well, that seems like a misguided decision to make.
#2) The "14 developers who haven't been working on SMPng" have NOT left FreeBSD and continue to do their work in the other areas of the system they work on. The announcement merely stated the lack of current developer activity on the SMPng project within FreeBSD. These people have not left FreeBSD and there certainly hasn't been any massive behind-the-scenes "rift" in the developer community, as some of the replies to this thread seem to be assuming.
FreeBSD developer activity continues to increase, and in fact has probably never been better in the history of the project
FWIW, FreeBSD is trialling binary update packages for security holes starting with 4.3-RELEASE. If things go well, perhaps OpenBSD will adopt a similar system.
Slashdot Mirror
...for Brown Pants Linnex!
OpenBSD's "only one hole in over 7 years" claim is made using some very, umm, "non-standard" definitions. The last time I checked, using the same definitions FreeBSD could actually claim a longer period without any security holes.
Of course, this is a completely meaningless (and misleading) statistic since both FreeBSD and OpenBSD have had numerous root holes in the default installation published and fixed.
Support for RFC 3514 was committed to FreeBSD already.
I bet there were some juicy tidbits to be discovered by a sniffer on that wireless network! I wonder how many passwords and email correspondence were captured by the intelligence agencies of the world present at that conference.
It wasn't posted by a FreeBSD committer or core team member.
I don't see your point.
:)
Yes, physics people publish in journals (after first publishing the e-print on xxx.arxiv.org). Yes, physics people cite journal articles (together with the arxiv reference). And yes, non-physics people read journals (because they are 5-10 years behind the trend towards e-publishing). But physics people don't read articles in paper journals, instead they read the e-print version on xxx.arxiv.org.
As a theoretical physicist myself, I think I am speaking with some authority here
The other highly suspcious aspect of this whole affair is that the work was never published online prior to being submitted to a journal.
Some background for non-physicists: thesedays the primary venue for publishing new works is the arXiv. Several hundred papers per day are uploaded here in various categories, and it is the de facto standard library of modern research in physics.
After publishing your work on the arXiv, physicists around the work can and will read your paper and submit feedback. Typically, after publication on the arXiv you might submit your work to a paper-based journal, but this is only a secondary procedure, and the only real point is to give you bonus points for your resume.
Here is the main point:
No-one reads paper journals any more!
The fact that the Bogdanov papers were never uploaded to the arxiv meant that apart from the 2 referees (who basically seem to have abdicated responsibility), no-one had, or ever would have read their work!.
If the authors were serious researchers, they would have submitted their work to the arxiv so it could be read and critiqued by their peers.
Undergraduate physics education is the USA is a travesty; however fortunately this is not true in other countries. This is presumably why the vast majority of PhD students in physics in the USA are foreign students. This is especially true in theoretical physics: most american students just cannot compete with their foreign counterparts.
There _are_ universities with good undergraduate physics programs (and good students coming out of the bad universities), but they certainly seem to be in the minority.
Their "model" is not even stated, much less stated self-consistently :-)
Nonsense is impossible to understand by anyone.
No, but to people who do understand the field in which the Bogdanov brothers claim to be researching (I am a PhD student in string theory and include myself in this set), their paper is complete gibberish.
Their paper is full of unfounded assertions strung together, combined with definitions and other assertions that are patently false. The paper does not follow a logical chain of reasoning that allows the reader to repeat or verify their conclusions (which are also not clearly stated).
Furthermore, when confronted by other theoretical physicists (on the sci.physics.research Usenet group) with specific, detailed questions about their work, the authors have systematically refused to answer, or selectively answered with further vague or absurd statements. To me, this is the real clincher: they have completely failed to demonstrate a technical understanding of the field in which they claim to be working.
It is clear to everyone in the field that these papers are nonsense and should never have been published. The only actual supporters of the authors seem to be from non-scientists or physicists who are unqualified to judge the work itself.
Your expectations are misguided and unfair. I don't know of ANY free software project that provides support for arbitrarily old releases -- unless you pay for it. Conversely, if you're willing to pay for support then you can surely find someone to support OpenBSD for you.
The way I counted them was by building a new INDEX file (one entry per port) and then running wc -l on it :-)
Or FreeBSD's capability bits, available in 5.0.
Hmm, that seems a bit bass-ackwards: you're on a much slower connection, and you want to download a 700M ISO image instead of ~100M of gzipped files (for a base install)?
You're much better off downloading the distribution sets you want to install to a local source (NFS, MSDOSFS, etc), since you'll only be downloading what you actually want to install, and not all the extra crap on the iso.
No, you're confusing things. FreeBSD added the 48-bit block addressing *128PB) capabilities a month ago, for which the Linux patch has just been announced.
:-)
The ability to address 48-bit blocks on a disk has nothing to do with the size of filesystems which can be constructed on that disk, or the files which can be created within that filesystem. This is a concept known as 'layering'
Well, actually it's not file sizes; FreeBSD has had 64-bit file size forever, and Linux has that too thesedays I believe. That's a completely separate issue to the ATA driver being able to address 48-bit physical media, which is what the support is for. And actually the Linux ATA driver maintainer only released these patches a few days ago (and they're not yet integrated into any of the kernel distributions), so they are much more recent than FreeBSD's support.
File sizes
Filesystem sizes
Partition sizes
Physical media sizes {-- this is what we're talking about: the limitations on these four things are more or less independent
I read the message, and this seems almost identical to the jail(8) functionality which FreeBSD has had for a number of years. Jail is very very useful for a number of purposes, but this doesn't seem to be anything new, just a "feature catch-up".
Well, it's good to see that Linux has caught up, but the article is not correct that Linux is the first OS to support 48-bit ATA; FreeBSD has had this support for over a month now.
See for example: this file which is one of the files containing the ATA-6r2 code, committed to FreeBSD on October 6.
I got two FreeBSD CD sets delivered today from DN and they each came with a 'BSD' car sticker similar to the country-of-registration stickers you see in europe. Kinda cool :)
Jordan Hubbard is *not* the "lead programmer". He's an evangelist and does very little coding.
You've also misunderstood the reason for pushing back 5.0-RELEASE; the changes being made were just too big, and the system isn't ready for a release yet. We could have released an unstable, crap 5.0-RELEASE now, or we can wait and get it right.
FreeBSD 4.5
No, it wasn't just data structures and variable names, but *comments*. Check the files yourself from the links in the story.
Please stop trolling.
#1) Jordan is as active as he's ever been in FreeBSD despite the move in his daytime job to Apple. He's even paid to work on FreeBSD during the week. The fact of the matter is that Jordan isn't an active developer anyway, he's more of a publicist and manager. If you're choosing to move away from FreeBSD because you think the PR spokesman is no longer dedicated to his job, well, that seems like a misguided decision to make.
#2) The "14 developers who haven't been working on SMPng" have NOT left FreeBSD and continue to do their work in the other areas of the system they work on. The announcement merely stated the lack of current developer activity on the SMPng project within FreeBSD. These people have not left FreeBSD and there certainly hasn't been any massive behind-the-scenes "rift" in the developer community, as some of the replies to this thread seem to be assuming.
FreeBSD developer activity continues to increase, and in fact has probably never been better in the history of the project
FWIW, FreeBSD is trialling binary update packages for security holes starting with 4.3-RELEASE. If things go well, perhaps OpenBSD will adopt a similar system.