Except that Darwin is actually based on NeXTStep in whatever state it was when Apple bought NeXT, which is effectively 4.2BSD with some graphical add-ons.
The only relation FreeBSD 3.2 bears to the OS which Apple acquired via NeXTStep is that it comes from 4.xBSD way back when and has been updated in similar ways.
That said, NeXTStep is painfully out of date in some basic, security relevant ways (I have a NeXT 040 turbo cube running NeXTStep 3.3 sitting to my left and completely hidden behind ipf and ipnat on my NetBSD/i386 machine, so I'd say that yes, I am qualified to speak about it, thanks). So the chances that FreeBSD was used as a reference platform for dragging NeXStep into preparedness for the present-day Internet is not an unreasonable assumption, though I don't know the details of that myself.
Um, considering Frank van der Linden ported this on Wasabi's time and money and, when he contacted AMD to mention he was ready to go and make a press release, they were shocked at how quickly the port had happened, I really don't think this was part of AMD's marketing plan. I'd love to think that claiming "NetBSD already runs on it!" meant something, but I really don't think my own favorite OS gets quite enough press to be useful.
As far as practicality goes, if you don't see why companies will be jumping at 64 bit systems as quickly as they can get them, you obviously haven't been paying much attention to what first DEC and Sun, and now Intel have been doing lately, have you?
That would explain the cancellation of The Lone Gunmen. After all they used LINUX (or at least Langley and Byers did. . . ). M$ obviously paid Fox to take 'em off the air. ..
I want to be able to pay the artists money for their songs. Up until now, there simply is no way to give money if you want to download an electronic version. If they allow me to pay a reasonable price to download a song, then I will gladly pay it.
Unfortunately, that's not what will happen through this scheme.
You'll be giving money to the copyright owner, who is almost without fail the label. If you really think that RIAA is in business for the artists, you're completely deluding yourself.
Further, there exist artists (They Might Be Giants comes to mind) who have had ways for their audience to get their music electronically and compensate them directly for quite some time now. Granted, it's not the norm, but it'd be the right way to do it, rather than feeding RIAA more cash.
While adding another set of root nameservers to the standard root.cache sounds like a good idea on its face (and should be technically feasible, unless I'm missing something in my memory of how BIND 8 deals with cache), it won't work.
Why?
A fair portion of the Internet doesn't use BIND. Even on Unix systems, there are BIND replacements, just as there are sendmailreplacements. But even ignoring the Unix world, what about Windows 2k, etctera? I mean, sure, they're making plenty of modifications to what's Right on their own (domains segments beginning in "_", for instance), but the chances of Microsoft not going along with ICANN (especially if NSI shuffles some money behind MS stock) are awfully low.
Its funny to see a slashdot user say that linux isn't good for some platform. I have the frame of mind
that, initially, it might not be the best, but because of it being open sourced, and having thousands of
developers, it will become the best for what people use it for.
And, since the point of a PDA is for it to work quickly and behave as if it were a notepad, addressbook, etcetera, from where do you expect those thousands of developers to arise?
I have a palm. I use it a lot. I'm not about to replace it with something less useable that maybe has a touch more hack value. These things are not computers, and treating the market the same way is a mistake.
So, I'm not interested, but maybe other people will be? I doubt it. I'm an active contributer to NetBSD. I've whacked away mostly at making extraneous things behave on the macppc port. There are fewer active developers of the port than I can count on two hands. Definitely not thousands.
I'll grant you that I'm not talking about Linux, but development of NetBSD functions along similar principles... principles that are perhaps more applicable to the porting-[insert OS here]-to-a-PDA discussion, considering the number of platforms on which NetBSD runs (ls -l/usr/src/sys/arch | grep ^d | grep -v CVS | wc -l says 35). In order to even write applications for this new platform, you're really going to have to buy one yourself, because the platform is so drastically different from a standard Linux machine, and that's a pretty major barrier.
Especially for those of us perfectly content with PalmOS's functionality... things are pretty much right with it already, why change?
It seems more likely that your server processor is slow enough that the encryption time is significant.
... which sounds like restatement when I reread it.
Point is, I don't have the same problem that ClaudioLeite did, and I doubt that a majority of people connecting with an ssh1 client to an sshd2-on-sshd1 server do.
I connect quite regularly from an aging Gateway machine running Win98 via a dialup at my apartment to my NetBSD/macppc machine running sshd2 on Swarthmore College's campus. Granted, the dialup and the server are on the same LAN, but I see no greater latency connecting to my NetBSD machine with a G3) than I do connecting to the Computer Science Program's Solaris 2.6 box running sshd1 (which has two UltraSparcs).
It seems more likely that your server processor is slow enough that the encryption time is significant.
Considering that debate hinges primarily on the hubris of Theo de Raadt, who has a long history of being fairly disagreeable rather permanently, that doesn't seem very likely.
OpenBSD is good software, which I use in several places for several organizations, and Theo seems like a pretty nice guy most of the time, but he definitely has difficulties controling his temper, especially when he doesn't get his way.
... except that MacOS X descends from NeXTStep, which is different (kernally) from modern BSD in a lot of (both bad and not so bad) ways.
That said, Darwin's fairly up to date, but it's not the same code (or, most likely, the same design ethic) as BSD.
Even more, MacOS X uses the Mach Microkernel (as MkLinux does and LinuxPPC does not), which makes it very different from NetBSD/macppc (which, as plenty of people have noted, already exists and already runs on iMacs just fine, thank you), which boots through the Mac's OpenFirmware (hold down command-option-O-F on a PCI-arch mac).
Jordan Hubbard is welcome to look at the iMac, but it sure seems like reinventing the wheel, since NetBSD/macppc works just fine as far as I can tell:
uriel:~% uname -mprs
NetBSD 1.4R macppc powerpc
uriel:~% uptime
5:50PM up 11 days, 33 mins, 4 users, load averages: 0.19, 0.10, 0.09... and that's just because I have no UPS during power outtages.
Really flashy page, but it loads fast enough, even as slashdotted as it presumably is right now.
At a glance, I'd say all the old files are still there (searching for the latest 1000 files gives you just what you asked for... not that it's fun waiting for Netscape to render that), and it's all presented well.
Their search engine's pretty basic in that you can't give terms higher or lower precedence, or exclude them entirely, unfortunately, but maybe that will change. (If I want to know about holes in screen, I do a search for "screen"... and the first thing returned - with four stars - is "nt.screensaver.token.txt"... so I try "screen -NT", same first choice, more NT stuff in the list below.)
The archives page is very cool, pulling important stuff out into categories that are easy to browse, so maybe it's not really necessary that the search engine be perfect.
This, google's release, and my new Palm IIIx all make me happy.:^>
zanONi wrote: Would someone be kind to post their new IP ?
Not to flog a dead horse here, but it's really easy to get this information for yourself, which you might have done and included it in your original post if you wanted to be helpful.
Evan Vetere wrote: My karma just got knocked down two points in the past hour; this indicates to me that I am doing a bad job moderating. News to me - I thought I'd always been sublimely fair and impartial.
That could well have been an old, moderated-up comment of yours trailing off the back end of the system, rather than any effect of meta-moderation on your moderation.
WNight wrote: The code is often the smaller part of making a new util. Design, both in identifying the problem, and in deciding on the implementation, are as much of a problem as the actual coding.
Ah, yes, but you've glossed over an important point there: one big thing that's different between Gnu utilities/the Linux kernel and 4.4 BSD/System V utilities and kernel is the implementation. This is the bulk of the "actual coding".
When you write a piece of software to mimic another piece of software, your front end may look very much the same, but your back end could well be going about things differently.
That said, the Gnu tools weren't built as if in a clean room, so they probably mimic implementation pretty closely. I was referring mostly, however, to the Linux kernel, which we all know does some things in a very (and at times purposely) different way.
(Why can't we have a thread that mentions GNU without someone using the term fanatic? They have stated a goal and are working towards that goal, not getting distracted in the meme of the moment isn't fanaticism.)
I don't really want to get in a debate over defintions of fanaticism (and it would just be silly, anyway), but suffice to say that I didn't mean that all (or even most... or, really, even more than a handful) of the members of the Gnu project were fanatics. I was referring to one particular individual, whose initials start with R and end with S. Anyway, who said fanaticism was necessarily a negative concept? It definitely gets things done...
Anonymous Coward wrote: Go to theos.org and read the huge file Theo has there of the email exchange between NetBSD core and himself. "Kicked out" seems like a pretty good summary of the situation to me.
Second, the file you describe isn't up right now, but I imagine it belongs under theos.com/deraadt somewhere.
Third, I think I've read it before, and it displays some flaring tempers, but the basic issue was that Theo wanted to go a direction the Core didn't. Also, Theo has made it pretty clear that this is ancient history. The Open- and NetBSD projects are on amicable terms now and regularly kick code back and forth between their CVS repositories.
First off, let me say that it's good to see an article even addressing these issues in the Wall Street Journal, and as one poster said, they did a pretty good job for mainstream media. That said, the author tries to go into details and teach a history that he actually doesn't know, and that some/. readers also may not know, because it's pretty convoluted. These are arguably minor points, but I think they're important.
The BSD programs and Linux actually share a common lineage, a collective development process and a rambunctious cast of characters.
The free programs are all variants of the venerable Unix system invented by AT&T Corp.
This is basically untrue. All four BSDs (including BSDi's BSD/OS) stem from the AT&T Unix sources, Linux was written entirely without access to those sources. It behaves similarly in a lot of ways, but vastly differently in others (arp and routing tables, for instance).
This isn't to say that either Unix/BSD's or Linux's way is better (I personally prefer the methods that have been around and proven for twenty-odd years, but that's me).
The author may have been trying to straighten out this mis-statement when he wrote:
The Linux saga is already the stuff of modern legend. In 1991, Linus Torvalds, a 21-year-old student in Helsinki, began writing an operating system essentially from scratch so he could have something to use on his home computer. The programs FreeBSD, NetBSD and OpenBSD, by contrast, are the descendants of code written in the late 1970s and early 1980s at UC Berkeley.
... but I'm not sure that really clarifies things for the average reader, and has some factual failings of its own (Linus wrote a kernel, not an operating system, and operating system needs basic software, Gnu had it, we all know the drill and the fanatics involved).
OpenBSD was started in 1995 by Theo de Raadt, a mountain biking 31-year-old Canadian after being kicked out of the NetBSD movement.
Okay, so maybe Theo didn't leave NetBSD under the friendliest of circumstances, but to claim he was "kicked out" isn't really fair. He had disagreements about what the focus of the program should be, so he broke off to pursue the focus he felt was more important. This doesn't make either focus invalid, just points up the fact that you can't have one set of people focusing on both spreading platform support and securing all OS processes. The above comments imply that there's some kind of lasting enmity between the Open- and NetBSD projects, which simply isn't true.
All of this said, the point an earlier poster made about how this is a pretty good article, and that the mainstream media is doing a much better job than they once did is quite valid. I'm also gladdened to see this article wasn't just more slobbering over RedHat... I've seen quite enough of that to last me the rest of my days.
I mean, it just popped up at the top of my/. page when I came to read this afternoon.
Does this have the same selection criteria as regular moderation, or is every (logged-in)/.er seeing this? Every day? Only for a testing period?
Also, as I've just been reading what Hofstadter has to say about recursion and meta-stuff in Godel, Escher, Bach: an Eternal Golden Braid (meta-genies in his case, Chapter V and thereabouts), how long before there's meta-meta-moderation, and so forth?
How deeply into this self-checking are we going to go before the moderation is really no more effectual than the posting? I mean, how many people actually read any given/. comment? Right now, very few. How many will see the effect of a given moderation? A whole lot more... right now. But for how long?
On a separate-but-related note, when I explained/.'s (fairly) new moderation system to a friend, he commented, "That's sort of oddly big-brother, isn't it?"
My only response was "Well, yeah, in a publicly-held sort of way."
Slashdot Mirror
Heh. And I was just about to make a "Feh, bash" post.
/. posts very well.
I like using zsh's RPROMPT, though the results don't show up in html
(I keep RPROMPT set to '[%!]'.)
--
Wow. I've really fallen completely out of the gaming circle.
I know Bungie's off at M'soft developing Xbox stuff, but I was unaware they'd sold the rights to Myth. When'd that happen?
--
Except that Darwin is actually based on NeXTStep in whatever state it was when Apple bought NeXT, which is effectively 4.2BSD with some graphical add-ons.
The only relation FreeBSD 3.2 bears to the OS which Apple acquired via NeXTStep is that it comes from 4.xBSD way back when and has been updated in similar ways.
That said, NeXTStep is painfully out of date in some basic, security relevant ways (I have a NeXT 040 turbo cube running NeXTStep 3.3 sitting to my left and completely hidden behind ipf and ipnat on my NetBSD/i386 machine, so I'd say that yes, I am qualified to speak about it, thanks). So the chances that FreeBSD was used as a reference platform for dragging NeXStep into preparedness for the present-day Internet is not an unreasonable assumption, though I don't know the details of that myself.
--
As far as practicality goes, if you don't see why companies will be jumping at 64 bit systems as quickly as they can get them, you obviously haven't been paying much attention to what first DEC and Sun, and now Intel have been doing lately, have you?
--
That would explain the cancellation of The Lone Gunmen. After all they used LINUX (or at least Langley and Byers did. . . ). M$ obviously paid Fox to take 'em off the air. . .
;^>
Um, yeah, or else the show just really sucked...
I want to be able to pay the artists money for their songs. Up until now, there simply is no way to give money if you want to download an electronic version. If they allow me to pay a reasonable price to download a song, then I will gladly pay it.
Unfortunately, that's not what will happen through this scheme.
You'll be giving money to the copyright owner, who is almost without fail the label. If you really think that RIAA is in business for the artists, you're completely deluding yourself.
Further, there exist artists (They Might Be Giants comes to mind) who have had ways for their audience to get their music electronically and compensate them directly for quite some time now. Granted, it's not the norm, but it'd be the right way to do it, rather than feeding RIAA more cash.
--
(that is, on the home row.)
If they had SDL they could run this Marathon... :P
Uh, we do and, incidentally, I am. (Too lazy to bother getting a uname and an Aleph One shot on the screen at the same time, though.)
Matter of fact, seeing as we have full binary Linux compatibility on the i386 architecture, I could have just installed all of this by RPM.
Trust me, you ain't got anything we ain't also got.
While adding another set of root nameservers to the standard root.cache sounds like a good idea on its face (and should be technically feasible, unless I'm missing something in my memory of how BIND 8 deals with cache), it won't work.
Why?
A fair portion of the Internet doesn't use BIND. Even on Unix systems, there are BIND replacements, just as there are sendmail replacements. But even ignoring the Unix world, what about Windows 2k, etctera? I mean, sure, they're making plenty of modifications to what's Right on their own (domains segments beginning in "_", for instance), but the chances of Microsoft not going along with ICANN (especially if NSI shuffles some money behind MS stock) are awfully low.
Or, perhaps, Atari.
Its funny to see a slashdot user say that linux isn't good for some platform. I have the frame of mind that, initially, it might not be the best, but because of it being open sourced, and having thousands of developers, it will become the best for what people use it for.
/usr/src/sys/arch | grep ^d | grep -v CVS | wc -l says 35). In order to even write applications for this new platform, you're really going to have to buy one yourself, because the platform is so drastically different from a standard Linux machine, and that's a pretty major barrier.
And, since the point of a PDA is for it to work quickly and behave as if it were a notepad, addressbook, etcetera, from where do you expect those thousands of developers to arise?
I have a palm. I use it a lot. I'm not about to replace it with something less useable that maybe has a touch more hack value. These things are not computers, and treating the market the same way is a mistake.
So, I'm not interested, but maybe other people will be? I doubt it. I'm an active contributer to NetBSD. I've whacked away mostly at making extraneous things behave on the macppc port. There are fewer active developers of the port than I can count on two hands. Definitely not thousands.
I'll grant you that I'm not talking about Linux, but development of NetBSD functions along similar principles... principles that are perhaps more applicable to the porting-[insert OS here]-to-a-PDA discussion, considering the number of platforms on which NetBSD runs (ls -l
Especially for those of us perfectly content with PalmOS's functionality... things are pretty much right with it already, why change?
Been done.
Repeatedly.
It seems more likely that your server processor is slow enough that the encryption time is significant.
... which sounds like restatement when I reread it.
Point is, I don't have the same problem that ClaudioLeite did, and I doubt that a majority of people connecting with an ssh1 client to an sshd2-on-sshd1 server do.
Hm.
I connect quite regularly from an aging Gateway machine running Win98 via a dialup at my apartment to my NetBSD/macppc machine running sshd2 on Swarthmore College's campus. Granted, the dialup and the server are on the same LAN, but I see no greater latency connecting to my NetBSD machine with a G3) than I do connecting to the Computer Science Program's Solaris 2.6 box running sshd1 (which has two UltraSparcs).
It seems more likely that your server processor is slow enough that the encryption time is significant.
Considering that debate hinges primarily on the hubris of Theo de Raadt, who has a long history of being fairly disagreeable rather permanently, that doesn't seem very likely.
OpenBSD is good software, which I use in several places for several organizations, and Theo seems like a pretty nice guy most of the time, but he definitely has difficulties controling his temper, especially when he doesn't get his way.
... except that MacOS X descends from NeXTStep, which is different (kernally) from modern BSD in a lot of (both bad and not so bad) ways.
That said, Darwin's fairly up to date, but it's not the same code (or, most likely, the same design ethic) as BSD.
Even more, MacOS X uses the Mach Microkernel (as MkLinux does and LinuxPPC does not), which makes it very different from NetBSD/macppc (which, as plenty of people have noted, already exists and already runs on iMacs just fine, thank you), which boots through the Mac's OpenFirmware (hold down command-option-O-F on a PCI-arch mac).
Not that folks can't figure it out for themselves, but the Mac PowerPC port is at http://www.netbsd.org/Ports/macppc".
... and that's just because I have no UPS during power outtages.
Jordan Hubbard is welcome to look at the iMac, but it sure seems like reinventing the wheel, since NetBSD/macppc works just fine as far as I can tell:
uriel:~% uname -mprs NetBSD 1.4R macppc powerpc uriel:~% uptime 5:50PM up 11 days, 33 mins, 4 users, load averages: 0.19, 0.10, 0.09
(It's a PowerMac 7500 with a G3 processor card.)
Really flashy page, but it loads fast enough, even as slashdotted as it presumably is right now.
:^>
At a glance, I'd say all the old files are still there (searching for the latest 1000 files gives you just what you asked for... not that it's fun waiting for Netscape to render that), and it's all presented well.
Their search engine's pretty basic in that you can't give terms higher or lower precedence, or exclude them entirely, unfortunately, but maybe that will change. (If I want to know about holes in screen, I do a search for "screen"... and the first thing returned - with four stars - is "nt.screensaver.token.txt"... so I try "screen -NT", same first choice, more NT stuff in the list below.)
The archives page is very cool, pulling important stuff out into categories that are easy to browse, so maybe it's not really necessary that the search engine be perfect.
This, google's release, and my new Palm IIIx all make me happy.
zanONi wrote:
Would someone be kind to post their new IP ?
Not to flog a dead horse here, but it's really easy to get this information for yourself, which you might have done and included it in your original post if you wanted to be helpful.
Observe:
turmeric:~ 499$ whois securify.com
[netsol garbage]
Registrant:
Securify Inc. (SECURIFY-DOM)
[information not relevant here]
Domain servers in listed order:
WEST.MAINSTREET.NET 207.5.0.40
EAST.MAINSTREET.NET 199.245.73.9
turmeric:~ 500$ nslookup packetstorm.securify.com WEST.MAINSTREET.NET
Server: west.mainstreet.net
Address: 207.5.0.40
Name: packetstorm.securify.com
Addresses: 209.143.242.114, 209.143.242.115
turmeric:~ 501$
Evan Vetere wrote:
My karma just got knocked down two points in the past hour; this indicates to me that I am doing a bad job moderating. News to me - I thought I'd always been sublimely fair and impartial.
That could well have been an old, moderated-up comment of yours trailing off the back end of the system, rather than any effect of meta-moderation on your moderation.
WNight wrote:
The code is often the smaller part of making a new util. Design, both in identifying the problem, and in deciding on the implementation, are as much of a problem as the actual coding.
Ah, yes, but you've glossed over an important point there: one big thing that's different between Gnu utilities/the Linux kernel and 4.4 BSD/System V utilities and kernel is the implementation. This is the bulk of the "actual coding".
When you write a piece of software to mimic another piece of software, your front end may look very much the same, but your back end could well be going about things differently.
That said, the Gnu tools weren't built as if in a clean room, so they probably mimic implementation pretty closely. I was referring mostly, however, to the Linux kernel, which we all know does some things in a very (and at times purposely) different way.
(Why can't we have a thread that mentions GNU without someone using the term fanatic? They have stated a goal and are working towards that goal, not getting distracted in the meme of the moment isn't fanaticism.)
I don't really want to get in a debate over defintions of fanaticism (and it would just be silly, anyway), but suffice to say that I didn't mean that all (or even most... or, really, even more than a handful) of the members of the Gnu project were fanatics. I was referring to one particular individual, whose initials start with R and end with S. Anyway, who said fanaticism was necessarily a negative concept? It definitely gets things done...
Anonymous Coward wrote:
Go to theos.org and read the huge file Theo has there of the email exchange between NetBSD core and himself. "Kicked out" seems like a pretty good summary of the situation to me.
First off, that's theos.com
Second, the file you describe isn't up right now, but I imagine it belongs under theos.com/deraadt somewhere.
Third, I think I've read it before, and it displays some flaring tempers, but the basic issue was that Theo wanted to go a direction the Core didn't. Also, Theo has made it pretty clear that this is ancient history. The Open- and NetBSD projects are on amicable terms now and regularly kick code back and forth between their CVS repositories.
This is basically untrue. All four BSDs (including BSDi's BSD/OS) stem from the AT&T Unix sources, Linux was written entirely without access to those sources. It behaves similarly in a lot of ways, but vastly differently in others (arp and routing tables, for instance).
This isn't to say that either Unix/BSD's or Linux's way is better (I personally prefer the methods that have been around and proven for twenty-odd years, but that's me).
The author may have been trying to straighten out this mis-statement when he wrote:
Okay, so maybe Theo didn't leave NetBSD under the friendliest of circumstances, but to claim he was "kicked out" isn't really fair. He had disagreements about what the focus of the program should be, so he broke off to pursue the focus he felt was more important. This doesn't make either focus invalid, just points up the fact that you can't have one set of people focusing on both spreading platform support and securing all OS processes. The above comments imply that there's some kind of lasting enmity between the Open- and NetBSD projects, which simply isn't true.
All of this said, the point an earlier poster made about how this is a pretty good article, and that the mainstream media is doing a much better job than they once did is quite valid. I'm also gladdened to see this article wasn't just more slobbering over RedHat... I've seen quite enough of that to last me the rest of my days.
I mean, it just popped up at the top of my /. page when I came to read this afternoon.
/.er seeing this? Every day? Only for a testing period?
/. comment? Right now, very few. How many will see the effect of a given moderation? A whole lot more... right now. But for how long?
/.'s (fairly) new moderation system to a friend, he commented, "That's sort of oddly big-brother, isn't it?"
Does this have the same selection criteria as regular moderation, or is every (logged-in)
Also, as I've just been reading what Hofstadter has to say about recursion and meta-stuff in Godel, Escher, Bach: an Eternal Golden Braid (meta-genies in his case, Chapter V and thereabouts), how long before there's meta-meta-moderation, and so forth?
How deeply into this self-checking are we going to go before the moderation is really no more effectual than the posting? I mean, how many people actually read any given
On a separate-but-related note, when I explained
My only response was "Well, yeah, in a publicly-held sort of way."
Hrm.
I wrote:
Or Hemos's, for that matter?
ahem