I can certainly see how it'd be cool as a hacking plaything to mess around with. I wouldn't mind getting a Linux PS2 just for that very reason.
As for the average Joe Customer wanting a Linux PS2 Kit..of course they have no need for it. The average Joe Customer isn't even a Linux user. You may as well ask "What does the average Joe Customer want Linux on a PC for?" also. The fact that it's a PS2 doesn't matter. It's obviously being sold to geeks and developers.
Robert Heinlein has coined/foreshadowed many tech terms and acheivements. He coined the term "free fall" and, IIRC, was a NASA advisor as well.
The best stuff (that applys to your research) would be the collection of Future History books. It's the early stuff, mostly juvenile, and not the adult (not _Adult_) novels that cover Lazarus Long, etc.
If you were replying to the story then you should not have replied to a specific post. There is a Reply button at the very top of the comments for starting a new comment thread. Misuse leads to confusion.
You either misunderstand or misremembered the explanation. No scripts break when root's shell is changed.
You shouldn't change root's shell because you shouldn't be using the root acount often enough for it to matter. If you use su to become root, use the -m flag so it keeps your existing shell and environment. That's what it's there for. Or, install sudo and use 'sudo bash' to become root. On all my servers I haven't 'logged in' as root in at least 6 months. Linux users use the root account far too often and then carry their bad habit to other OSes when they switch.
The other danger is not being able to use dynamically compiled shells in single user mode. You mentioned this.
I have always favored the BSD freenixes over Linux. One primary reason is that all code is maintained in publicly viewable CVS servers.
Linux, unfortunately, is not. To the best of my knowledge, Linus doesn't even use CVS privately. If you want to upgrade your kernel, you have to wait for new releases in the form of full or patch tarballs delivered to kernel.org like mana from heaven (Linus). There's no easy way to see arbitrary changes in any file at any time. There's no reading commit logs.
For that matter, there's no easy way to contribute. That is to say, there's not an _easier_ way. You have to mail your patches to some list or maintainer, etc. There's no public bug tracker.
I don't have any certifications and I never finished my CS degree (1.5 semesters was all I finished) and I was the Network & Unix admin at a regional ISP near Chicago for 2 years.
Being a geek by nature helped. I learned BASIC at 10 & ran a BBS in the 80's, etc. I already had some Unix and Net experience when I started at the ISP doing 1st level tech support. I owe a great deal to the boss I had (he was the owner and was an engineer at Bell Labs for 10 years before starting an ISP) who encouraged me to continue learning. I wrote scripts and read O'Reilly books. I setup a home network, I reg'd a domain and did my own DNS & Mail over nailed dialup.
Eventually I became Senior tech and then Administrator. I setup new domains and websites. I configured ciscos for new dedicated customers. There's more but this isn't a resume.;)
I also got involved in FreeBSD. I'm a contributer to the Ports Collection. I've been to both BSDCons.
Now I have the knowledge (thanks to O'Reilly) and the Experience (thanks to my former boss) to get an admin job anywhere that doesn't require a degree.:) But don't worry, there are alot of such places.
My current job is as a network application programmer using interpreted scripting languages. We're an ethernet provider for residential complexes & student housing, and my ISP experience is what got me the job.
I wouldn't mind a Cisco cert or two, and I could use some decent Solaris experience, but if I were to go back to school it would be for fun and not in a tech field.
NetBSD and FreeBSD simultaneously grew out of the 386BSD project which was headed by Bill Jolitz. It was a project that ported 4.4BSD to the i386.
There's no such thing as "FreeBSD regents". You're thinking of the Regents of the University of California, who owned the (open source) license to BSD. They're a bunch of university administrators and have nothing to do with operating system development.
In fact, NetBSD is technically older than "FreeBSD", as FreeBSD was then just a handful of people (4 or 5) who started releasing patches to 386BSD called the "386BSD Patchkit".
You're right about the rest.
For your (and anyone reading this) review, see/usr/share/misc/bsd-family-tree and visit http://www.daemonnews.org/200104/bsd_family.html.
FreeBSD follows a pretty standard development cycle, influenced by it's use of CVS.
There is the main development trunk, known as -CURRENT (or -HEAD and -MAIN in CVS terms). This is where all the really cool new stuff goes and as a result may be unstable at any given moment. Most work goes into this main trunk. This is 5.0 right now.
CVS allows branching of source. -CURRENT _used_ to be 4.0, but it progressed far enough to be -STABLE. A branch occured at that point, but -CURRENT continued growing straight up (becoming 5.0-CURRENT), and 4.X-STABLE started growing out to the side (just like a tree branch.)
When something in -CURRENT has been well tested and is deemed to be stable enough, it's merged into the branch below it. This is how -STABLE continues to grow. The latest -STABLE has reached 4.4 now.
CVS also supports tagging. It's primary use is to mark a snapshot of the CVS repository for -RELEASE.
It's always been done this way, and many, many software projects are developed this way.
This is unlike the Linux kernel development probably because Linus and friends don't use CVS (or if they do it's not public and it's not used in the same fashion.) Linux also has two different streams. The even numbered kernels are "stable", and the odd numbered kernels are "developmental". 2.2 used to be the stable branch, and 2.3 was the dev branch. 2.3 became the 2.4 stable branch, and 2.5 is now the dev branch. I only call them branches because I don't know what else to call them, not knowing how Linus actually manages source.
I was directly responding to your assertion that the end user has to know how to fetch, configure, build and install software.
They do not. Not with *BSD Ports or Packages or any of the Linux package systems.
Whether just running 'make' is a significant difference over './configure && make install' isn't the point. It's that the user doesn't have to know how, which you claimed they did.
You don't understand the Ports system very well then. The source is fetched _automatically_ and then it's built _automatically_ as well. No knowledge of transfer protocols is necessary, no knowledge of autoconf, automake, configure, etc is necessary. All of the innumerable problems were taken care of by the Maintainer so that you don't have to worrya bout them. A single command does all that is required. Don't like or know how to use a command line? Click a button that does it for you. There are already a few GUI frontends to Ports. The OpenPackages project will definately be able to provide the framework for a userfriendly, custom installed-software-at-a-whim system.
It's not even worrying about making money, though money IS necessary...unless you are a trust fund baby with buttloads of mommy and daddy's money to suck off of. The problem is that you (the collective you) do all this work for the sake of BSD only for a company like M$ to come along, fold YOUR work into their product and use it to make tons of money with a system that is totally incompatible with yours. They use your work to try to create a world in which your chosen BSD is all but useless but to a few.
My work is there, and it's still useful to anyone who used it in the first place. There's two scenarios here:
My Software is the only thing of it's kind. I find it useful, which is why I wrote it, and others find it useful, so they chip in if and when they can. If they don't, oh well.
My Software attracts the attention of Big Company. They copy (stealing is not possible unless they remove the hard drive from my CVS server) it, add some value (or not), and sell it as their own. Thousands of users who never heard of me or My Software now find Their Software useful. Their Software doesn't work with My Software. Big Company cashes in.
Life for me and my copy of My Software is unchanged. So what if Their Software doesn't work with My Software? In Scenario #1 They don't even exist. The only difference is that Thousands benefit from My Software, where only a few did before. The terms of My License may or may not require that I be acknowledged for it.
The Emphasis placed on your quoted text is my reponse to the bit about money. It (your argument) absolutely IS about money. (It's about something else too, which you either haven't realized or refuse to acknowledge - but I'll save it for later.) If it was money I was interested in I would have either made it closed source in the first place, or look, it's a BSD license, so I CAN!
I also assume that the incompatible software you're referring to is Kerberos. If this is the case then you misunderstand the Kerberos issue. Please search for my nick among the (now flattened) comments on this older/. story. (There is either a feature or a bug that is preventing me from linking to the specific ${cid}s. Probably related to the flattening.)
The bigger problem is that the BSD license selects for fragmentation rather than cohesiveness. Apple takes BSD, folds it into their OS, software developers write programs to work with Apple's BSD - none of it will run on your personal BSD. Your hard work was used to create a system that doesn't support your chosen OS. You are left out in the cold vis a vis software developers (no games or any other possibly desirous apps for you - just whatever you and your relatively few buddies can spin up in a weekend).
I think you're confused about a great many things regarding BSD and MacOS X. BSD hasn't been folded into MacOS. MacOS has been re-written on top of Darwin, which is Mach/FreeBSD. Any MacOS developers would be writing for MacOS. This means (IIRC) Carbon and Cocoa software, things which aren't possible on a vanilla BSD system anyway.
Darwin developers on the other hand, whether their dev OS is MacOS X or Darwin on PPC or x86 would be writing for the Unix OS that Darwin is. Any (non arch dependant) code would run with little modification (assuming it was OSS) on any *nix.
Me and my buddies were there before, and we'll be there after.
This may be perfectly fine for a few hardcore BSD users but it is NOT in the best interests of standards and general end-users.
Evil Mind Controlling Mega-Corps aside, how is the widespread use of my code not a benefit to others? As you say, they are subject to the whims of Evil Mega-Corp. They always will be whether I write BSD software or not. If this bit is an argument in favor of the GPL, I'm not seeing it. The sheep will use what the Evil Mega-Corp tells them they should use, be it _in spite_ of the GPL or _because_ of BSD.
I'm afraid I lost you on the next paragraph. Definate degradation detected.
M$ CAN and WILL embrace, extend, extinguish ANYONE who tries to compete against them, whether they are offereng some BSD variant or not. On the other hand, M$ cannot legally do this to linux. They cannot embrace and extend it, thus marginalizing it - if they tried, they would be de facto violating copyright law and would be liable.
Microsoft certainly could(we're talking possibility here, not probability) compete against and extinguish Linux. There's nothing illegal there. They can even embrace it. And they can even extend it. Only if they don't give those extensions back is it a violation of law. It's never going to happen of course, but for the purposes of this debate, it's not only _possible_, it's legal. I'm afraid you left yourself open with that "or not".
I really don't see how *BSD and its licensing is in any way an improvement on anything. It still allows monopolists like M$ to do business as usual whether they use BSD stuff or not.
The contradiction there is so strong it's almost painful. BSD and it's license allows monopolists to continue even if they don't use it? Come again?
Now for the something else. It's about control. That's it. It's not about money, though that's the first thing out of everyone's mouth. It's not even about Freedom (both kinds).
Is the GPL a good thing? Yes. Are Berkeley style licenses a good thing? Yes. Is one better than the other (The improvement you suggested)? Depends on the software.
Boy, that sure was a whole lot of well spoken points. I can't help but think they were meant for some other topic.
I clearly wasn't suggesting or advocating writing open source software for free instead of writing commercial, closed source software for pay.
If you don't want to write free (in both senses) software, then don't - and you won't here me, or ANY BSD advocate slight you for it. But if you do write free software, don't bitch about the money it's not making you. That's _all_ I said.
If BSD developers were out to gain anything of material value they wouldn't be writing free software in the first place. This argument by GNU advocates is weak in this regard. If you're all worried about someone selling your code and you not getting anything in return, why do you write it? It's about writing damned good code for your own personal use. Most projects start out of the needs of the original author. So some big scary company comes and gets a copy... so what? Original Author keeps writing his code as it suits him and (hopefully) anyone else who finds it useful.
WRT RedHat making money and OpenBSD and FreeBSD not.... *BSDs are software Projects, not software Companies. They're not OUT to make money. RedHat is. For that matter, Linux itself doesn't _make_ any money, and Linus didn't write it for that reason either. It got him a job and some nice free stock, but many companies who rely on the BSD projects hire full time BSD developers, so it works both ways.
I think it's time you reevaluated your goals. What do you want out of the software you write? The respect and admiration of your peers or a wad of cash?
It has never been known whether MS actually ever used the reference code.
As for the notion of items #4 and #5 being possible if the reference spec was published under the GPL, it's sort of like saying I can write something based on an RFC protocol, modified to work only between my own software, and call it by the name the RFC calls it.
You continue to be confused by the difference between specification (RFC) and implementation (reference code). It's nothing like saying anything of the sort.
Yes, if they didn't use the reference code, then they probably could've done the same thing.
Not probably, definately.
The GPL (were the Kerberos code published under this, or even the spec (if one can publish a spec under the GPL)) would have at least required Microsoft to publish the changes they made to the specification if they wanted to use it at all.
There is no copyright on RFCs. Furthermore, Microsoft MADE NO CHANGES to any such spec. How many times does it have to be repeated: "The original Kerberos spec had an empty, unused field." MS didn't add an extra field, or change the used fields, they simply used a field marked (By the Athena/Kerberos team) "reserved for future use."
There is no doubt that MS sucks the big one for making proprietary, non-compatible changes to an open standard.
The point being made here is that everyone continues to blame the BSD License for it, when it is demonstratably NOT SO.
1) The Kerberos Specification does not fall under a software license.
2) The Kerberos Specification includes an unused field.
3) The Kerberos reference implementation was released under a MIT style (ie, BSD style) license
4) MS used the unused field. They may or may not have used the reference implementation as a code base.
5) MS Kerberos only works with MS Kerberos. (Let's ignore current projects aiming to fix this.) MS keeps their changes to themselves.
I am not defending MS here. I am defending the BSD license. Even if the reference code was licensed under the GPL, #4 and #5 would still have been possible. Anyone who brings up the entire MS/Kerberos fiasco as an argument against the BSD license fails to realize this.
Who said he wasn't staying with FreeBSD? He was part of the patchkit team that birthed FreeBSD and he was the first full time employee of WC CDROM to work on FreeBSD. He ate then and he ate when he worked at BSDi and Wind River. He's just eating better maybe at Apple;), but he still plays the same role in the FreeBSD project that he always has.
Did you even read the message? For that matter, do you ever read any stories? "as I understand it"..where the hell have you been for the last 2 years?
They didn't do _anything_ with BSD'd Kerberos code. It's the Kerberos spec itself that has the empty field. They just put it to use. That's _ALL_. You admit this yourself but can't make the distinction. Whether they wrote their implementation themselves or used someone elses as a base, it's really irrelevant.
Slashdot Mirror
Hell, I remember when it _wasn't_.
VA Research was a fine name. 'VA Linux' just inflated the IPO rush.
You answered your own question:
I can certainly see how it'd be cool as a hacking plaything to mess around with. I wouldn't mind getting a Linux PS2 just for that very reason.
As for the average Joe Customer wanting a Linux PS2 Kit..of course they have no need for it. The average Joe Customer isn't even a Linux user. You may as well ask "What does the average Joe Customer want Linux on a PC for?" also. The fact that it's a PS2 doesn't matter. It's obviously being sold to geeks and developers.
Robert Heinlein has coined/foreshadowed many tech terms and acheivements. He coined the term "free fall" and, IIRC, was a NASA advisor as well.
The best stuff (that applys to your research) would be the collection of Future History books. It's the early stuff, mostly juvenile, and not the adult (not _Adult_) novels that cover Lazarus Long, etc.
If you were replying to the story then you should not have replied to a specific post. There is a Reply button at the very top of the comments for starting a new comment thread. Misuse leads to confusion.
You either misunderstand or misremembered the explanation. No scripts break when root's shell is changed.
You shouldn't change root's shell because you shouldn't be using the root acount often enough for it to matter. If you use su to become root, use the -m flag so it keeps your existing shell and environment. That's what it's there for. Or, install sudo and use 'sudo bash' to become root. On all my servers I haven't 'logged in' as root in at least 6 months. Linux users use the root account far too often and then carry their bad habit to other OSes when they switch.
The other danger is not being able to use dynamically compiled shells in single user mode. You mentioned this.
There are considerably _less_ debs than Ports.
I have always favored the BSD freenixes over Linux. One primary reason is that all code is maintained in publicly viewable CVS servers.
Linux, unfortunately, is not. To the best of my knowledge, Linus doesn't even use CVS privately. If you want to upgrade your kernel, you have to wait for new releases in the form of full or patch tarballs delivered to kernel.org like mana from heaven (Linus). There's no easy way to see arbitrary changes in any file at any time. There's no reading commit logs.
For that matter, there's no easy way to contribute. That is to say, there's not an _easier_ way. You have to mail your patches to some list or maintainer, etc. There's no public bug tracker.
When will it be Open? Or is Free enough?
I don't have any certifications and I never finished my CS degree (1.5 semesters was all I finished) and I was the Network & Unix admin at a regional ISP near Chicago for 2 years.
;)
:) But don't worry, there are alot of such places.
Being a geek by nature helped. I learned BASIC at 10 & ran a BBS in the 80's, etc. I already had some Unix and Net experience when I started at the ISP doing 1st level tech support. I owe a great deal to the boss I had (he was the owner and was an engineer at Bell Labs for 10 years before starting an ISP) who encouraged me to continue learning. I wrote scripts and read O'Reilly books. I setup a home network, I reg'd a domain and did my own DNS & Mail over nailed dialup.
Eventually I became Senior tech and then Administrator. I setup new domains and websites. I configured ciscos for new dedicated customers. There's more but this isn't a resume.
I also got involved in FreeBSD. I'm a contributer to the Ports Collection. I've been to both BSDCons.
Now I have the knowledge (thanks to O'Reilly) and the Experience (thanks to my former boss) to get an admin job anywhere that doesn't require a degree.
My current job is as a network application programmer using interpreted scripting languages. We're an ethernet provider for residential complexes & student housing, and my ISP experience is what got me the job.
I wouldn't mind a Cisco cert or two, and I could use some decent Solaris experience, but if I were to go back to school it would be for fun and not in a tech field.
There are errors in your history.
/usr/share/misc/bsd-family-tree and visit http://www.daemonnews.org/200104/bsd_family.html.
NetBSD and FreeBSD simultaneously grew out of the 386BSD project which was headed by Bill Jolitz. It was a project that ported 4.4BSD to the i386.
There's no such thing as "FreeBSD regents". You're thinking of the Regents of the University of California, who owned the (open source) license to BSD. They're a bunch of university administrators and have nothing to do with operating system development.
In fact, NetBSD is technically older than "FreeBSD", as FreeBSD was then just a handful of people (4 or 5) who started releasing patches to 386BSD called the "386BSD Patchkit".
You're right about the rest.
For your (and anyone reading this) review, see
FreeBSD follows a pretty standard development cycle, influenced by it's use of CVS.
There is the main development trunk, known as -CURRENT (or -HEAD and -MAIN in CVS terms). This is where all the really cool new stuff goes and as a result may be unstable at any given moment. Most work goes into this main trunk. This is 5.0 right now.
CVS allows branching of source. -CURRENT _used_ to be 4.0, but it progressed far enough to be -STABLE. A branch occured at that point, but -CURRENT continued growing straight up (becoming 5.0-CURRENT), and 4.X-STABLE started growing out to the side (just like a tree branch.)
When something in -CURRENT has been well tested and is deemed to be stable enough, it's merged into the branch below it. This is how -STABLE continues to grow. The latest -STABLE has reached 4.4 now.
CVS also supports tagging. It's primary use is to mark a snapshot of the CVS repository for -RELEASE.
It's always been done this way, and many, many software projects are developed this way.
This is unlike the Linux kernel development probably because Linus and friends don't use CVS (or if they do it's not public and it's not used in the same fashion.) Linux also has two different streams. The even numbered kernels are "stable", and the odd numbered kernels are "developmental". 2.2 used to be the stable branch, and 2.3 was the dev branch. 2.3 became the 2.4 stable branch, and 2.5 is now the dev branch. I only call them branches because I don't know what else to call them, not knowing how Linus actually manages source.
Sometimes it's ok to say things that people will disagree with and have it NOT be flame bait.
Cygwin isn't about Linux on Windows, it's about Unix on Windows. Anyone who thinks or says otherwise DOES have a narrow and biased view.
Moderation sucks.
--
I was directly responding to your assertion that the end user has to know how to fetch, configure, build and install software.
They do not. Not with *BSD Ports or Packages or any of the Linux package systems.
Whether just running 'make' is a significant difference over './configure && make install' isn't the point. It's that the user doesn't have to know how, which you claimed they did.
--
If you're going to introduce people to a whole new world, show them the right one and not your narrow little Linux biased view.
--
It's the size of your history, and it's shell dependant (Meaning OS INdependant.) It's completely unrelated to security.
--
You don't understand the Ports system very well then. The source is fetched _automatically_ and then it's built _automatically_ as well. No knowledge of transfer protocols is necessary, no knowledge of autoconf, automake, configure, etc is necessary. All of the innumerable problems were taken care of by the Maintainer so that you don't have to worrya bout them. A single command does all that is required. Don't like or know how to use a command line? Click a button that does it for you. There are already a few GUI frontends to Ports. The OpenPackages project will definately be able to provide the framework for a userfriendly, custom installed-software-at-a-whim system.
--
My work is there, and it's still useful to anyone who used it in the first place. There's two scenarios here:
Life for me and my copy of My Software is unchanged. So what if Their Software doesn't work with My Software? In Scenario #1 They don't even exist. The only difference is that Thousands benefit from My Software, where only a few did before. The terms of My License may or may not require that I be acknowledged for it.
The Emphasis placed on your quoted text is my reponse to the bit about money. It (your argument) absolutely IS about money. (It's about something else too, which you either haven't realized or refuse to acknowledge - but I'll save it for later.) If it was money I was interested in I would have either made it closed source in the first place, or look, it's a BSD license, so I CAN!
I also assume that the incompatible software you're referring to is Kerberos. If this is the case then you misunderstand the Kerberos issue. Please search for my nick among the (now flattened) comments on this older /. story. (There is either a feature or a bug that is preventing me from linking to the specific ${cid}s. Probably related to the flattening.)
The bigger problem is that the BSD license selects for fragmentation rather than cohesiveness. Apple takes BSD, folds it into their OS, software developers write programs to work with Apple's BSD - none of it will run on your personal BSD. Your hard work was used to create a system that doesn't support your chosen OS. You are left out in the cold vis a vis software developers (no games or any other possibly desirous apps for you - just whatever you and your relatively few buddies can spin up in a weekend).
I think you're confused about a great many things regarding BSD and MacOS X. BSD hasn't been folded into MacOS. MacOS has been re-written on top of Darwin, which is Mach/FreeBSD. Any MacOS developers would be writing for MacOS. This means (IIRC) Carbon and Cocoa software, things which aren't possible on a vanilla BSD system anyway.
Darwin developers on the other hand, whether their dev OS is MacOS X or Darwin on PPC or x86 would be writing for the Unix OS that Darwin is. Any (non arch dependant) code would run with little modification (assuming it was OSS) on any *nix.
Me and my buddies were there before, and we'll be there after.
This may be perfectly fine for a few hardcore BSD users but it is NOT in the best interests of standards and general end-users.
Evil Mind Controlling Mega-Corps aside, how is the widespread use of my code not a benefit to others? As you say, they are subject to the whims of Evil Mega-Corp. They always will be whether I write BSD software or not. If this bit is an argument in favor of the GPL, I'm not seeing it. The sheep will use what the Evil Mega-Corp tells them they should use, be it _in spite_ of the GPL or _because_ of BSD.
I'm afraid I lost you on the next paragraph. Definate degradation detected.
M$ CAN and WILL embrace, extend, extinguish ANYONE who tries to compete against them, whether they are offereng some BSD variant or not. On the other hand, M$ cannot legally do this to linux. They cannot embrace and extend it, thus marginalizing it - if they tried, they would be de facto violating copyright law and would be liable.
Microsoft certainly could(we're talking possibility here, not probability) compete against and extinguish Linux. There's nothing illegal there. They can even embrace it. And they can even extend it. Only if they don't give those extensions back is it a violation of law. It's never going to happen of course, but for the purposes of this debate, it's not only _possible_, it's legal. I'm afraid you left yourself open with that "or not".
I really don't see how *BSD and its licensing is in any way an improvement on anything. It still allows monopolists like M$ to do business as usual whether they use BSD stuff or not.
The contradiction there is so strong it's almost painful. BSD and it's license allows monopolists to continue even if they don't use it? Come again?
Now for the something else. It's about control. That's it. It's not about money, though that's the first thing out of everyone's mouth. It's not even about Freedom (both kinds).
Is the GPL a good thing? Yes. Are Berkeley style licenses a good thing? Yes. Is one better than the other (The improvement you suggested)? Depends on the software.
Dramatic finale omitted.
--
Boy, that sure was a whole lot of well spoken points. I can't help but think they were meant for some other topic.
I clearly wasn't suggesting or advocating writing open source software for free instead of writing commercial, closed source software for pay.
If you don't want to write free (in both senses) software, then don't - and you won't here me, or ANY BSD advocate slight you for it. But if you do write free software, don't bitch about the money it's not making you. That's _all_ I said.
--
If BSD developers were out to gain anything of material value they wouldn't be writing free software in the first place. This argument by GNU advocates is weak in this regard. If you're all worried about someone selling your code and you not getting anything in return, why do you write it? It's about writing damned good code for your own personal use. Most projects start out of the needs of the original author. So some big scary company comes and gets a copy... so what? Original Author keeps writing his code as it suits him and (hopefully) anyone else who finds it useful.
WRT RedHat making money and OpenBSD and FreeBSD not.... *BSDs are software Projects, not software Companies. They're not OUT to make money. RedHat is. For that matter, Linux itself doesn't _make_ any money, and Linus didn't write it for that reason either. It got him a job and some nice free stock, but many companies who rely on the BSD projects hire full time BSD developers, so it works both ways.
I think it's time you reevaluated your goals. What do you want out of the software you write? The respect and admiration of your peers or a wad of cash?
--
If the only problem was a switch/router config, why was a change of IP involved?
--
As for the notion of items #4 and #5 being possible if the reference spec was published under the GPL, it's sort of like saying I can write something based on an RFC protocol, modified to work only between my own software, and call it by the name the RFC calls it.
You continue to be confused by the difference between specification (RFC) and implementation (reference code). It's nothing like saying anything of the sort.
Yes, if they didn't use the reference code, then they probably could've done the same thing.
Not probably, definately.
The GPL (were the Kerberos code published under this, or even the spec (if one can publish a spec under the GPL)) would have at least required Microsoft to publish the changes they made to the specification if they wanted to use it at all.
There is no copyright on RFCs. Furthermore, Microsoft MADE NO CHANGES to any such spec. How many times does it have to be repeated: "The original Kerberos spec had an empty, unused field." MS didn't add an extra field, or change the used fields, they simply used a field marked (By the Athena/Kerberos team) "reserved for future use."
--
There is no doubt that MS sucks the big one for making proprietary, non-compatible changes to an open standard.
The point being made here is that everyone continues to blame the BSD License for it, when it is demonstratably NOT SO.
1) The Kerberos Specification does not fall under a software license.
2) The Kerberos Specification includes an unused field.
3) The Kerberos reference implementation was released under a MIT style (ie, BSD style) license
4) MS used the unused field. They may or may not have used the reference implementation as a code base.
5) MS Kerberos only works with MS Kerberos. (Let's ignore current projects aiming to fix this.) MS keeps their changes to themselves.
I am not defending MS here. I am defending the BSD license. Even if the reference code was licensed under the GPL, #4 and #5 would still have been possible. Anyone who brings up the entire MS/Kerberos fiasco as an argument against the BSD license fails to realize this.
--
The only thing he's leaving is Wind River. Linux works for Transmeta, but he's still at the top of the Linux pyramid.
JKH is still the chief PR spokesman for FreeBSD and it's release engineer and Core team member.
THAT HAS NOT CHANGED.
--
Who said he wasn't staying with FreeBSD? He was part of the patchkit team that birthed FreeBSD and he was the first full time employee of WC CDROM to work on FreeBSD. He ate then and he ate when he worked at BSDi and Wind River. He's just eating better maybe at Apple ;), but he still plays the same role in the FreeBSD project that he always has.
Did you even read the message? For that matter, do you ever read any stories? "as I understand it"..where the hell have you been for the last 2 years?
--
They didn't do _anything_ with BSD'd Kerberos code. It's the Kerberos spec itself that has the empty field. They just put it to use. That's _ALL_. You admit this yourself but can't make the distinction. Whether they wrote their implementation themselves or used someone elses as a base, it's really irrelevant.
--
Can no one tell the difference between quoted submissions and editors comments?
--