I don't see a lot of new GNU software coming from NASA....
No, that's becuase the four full-time open source developers I know of at NASA (at the Numerical Aerodynamic Simulation Laboratory) are working on NetBSD, which is under the Berkeley license, not the GPL.
cjs
OSI sucks ... and so does this article
on
OSI vs Taco Bell
·
· Score: 1
Oh, and let's not forget: Alan Turing, one of the best and most influential theoriticians in computing the world has ever seen, was a `fag.'
I'm not interested in chasing any of the anonymous folks off; I am, in fact, interested in what some of them have to say. But Slashdot really needs edited discussion threads. Quite often (as is the case with this one) I don't read any of the responses, because there are just too many of them.
Basically, it lets you layer two filesystems so that modifications you make get stored on one, but you read old data off the other. One example would be layering a read-write filesystem on top of a CD so that you could `change' the CD, without having to copy the whole CD to a read-write medium. Another would be having different developers with their own private layers above a shared source code directory; a developer can compile, change and delete files, and he'll affect his private workspace but not the other developers (until the changes are moved to the shared layer).
There's lots of cheap Sparc and other equipment out there if you're willing to live with the older models. (An IPX is dirt cheap these days; on the other hand, an SS5 is a lot more expensive for what you get because it's still a `current' model, even though modern PCs blow it away in terms of speed.)
However, if you want to get good deals and know what you're getting, do some research before you buy. For hardware information, the Sun Hardware FAQ is the biggest collection in one place. Another place worth looking is under the `Supported Hardware' link at the NetBSD project ; there's information there on many different systems, including Suns, and links to other sources of information. (This is also the only modern OS that runs on a lot of old equipment.)
Once you've had at least a cursory look through the resources available, spend a week or two reading through misc.forsale.computers.workstation to get an idea of what a decent price for this stuff is. There are a lot of bad (too high) prices posted there, but these usually get pointed out fairly quickly.
The last thing I'd recommend is to start cheap, to see if this is to your taste. If you decide that working with non-PC equipment is just too much of a hassle, or you get too little performance for your dollar, you're only out a couple of hundred dollars if you bought an IPX, rather than a couple of thousand if you bought an SS20.
Personally, I find old hardware quite rewarding. I currently have running at home 3 Sparcstations (two with rather nice 17" colour monitors that were dirt cheap!), a couple of Sun 3/60s (one colour, one mono, both with 19" monitors) and a VAXStation 2000, and my main mail server for cynic.net is an IPX. I've also had various other things kicking around in the past, though I recently cleaned out my collection. It's fun stuff to play with, and actually does useful work quite well in many circumstances.
Actually, I was just having a quick look at a copy of The Unix Haters Handbook the other day, since I'd not read it in years. I only read through a couple of dozen pages scattered about the book, but even then, something became apparent to me that never struck me four years ago when I first read it. The implication is around in several places, but it's blatently stated in paragraph 3 of page 128. Essentialy what it says is:
What this system really should do is transfer code from the server to the client and execute it remotely.
Well, looking back on the last few years of Java holes, macro viruses, and the ActiveX disaster, I think we can pretty safely say that this fellow was quite lacking in foresight. So enjoy the book, but take it with a big grain of salt. It's turning out that some of the things that seem horrible about Unix aren't as bad as one might think.
Now NeXT isn't really like BSD. If I'm wrong tell me.
You're wrong.:-)
The OS underlying NeXTStep was originally 4.3BSD. I expect the 4.4ised it at some point, and more recently Apple certainly grabbed a fair amount of NetBSD code to use in it. (This was quite a boon to the NetBSD project, because they contributed back a pile of bugfixes and whatnot.)
...I feel the hair-splitting wars between Linux distributions does not serve the Linux cause well.
Indeed, nor do I. Ironically enough, though plenty of Linux folks I talk to say that one of the `advantages' of Linux over *BSD is that `There's only one Linux.' Is this a case of not knowing oneself well enough (to the user, the differences between the BSDs are actually less apparent than the differences between, say, Slackware and Debian), or a case of thinking of Linux as Red Hat and nothing else, implicitly acknowledging the Microsoft view of the world (there can be only One True OS)?
It will be interesting to see where Linux ends up in the long run. Unlike the BSD folks, Linux folks tend to work very hard at proselytisation, and are a lot more likely to work hard at features that will convert Windows users, rather than other Unix users. (I'm not saying that this is better or worse, just different.) Thus, Linux is a lot more well known, despite being younger, and is indeed headed towards being what Windows is today.
I've looked at these things, but not even tried one, because they seem to be lacking in buttons. I use a four-button mouse with wheel in my right hand, and of the 17 keys I reach with my left hand without moving it, I use 11 quite regularly. Add to that the 12 function keys which I use for team communication, and a few other, harder to reach (they require hand movement) keys for various other things, and a controller without a couple of dozen buttons doesn't look so appealing.
I can see as firewire grows the price on SCSI will drop in order to keep up competition. Looking forward to 10GB Ultra SCSI 3 HD for $300.
If the SCSI drive is that cheap, why wouldn't the firewire drive be the same price, or nearly the same? After all, it's the same drive, just with a slightly different controller on it.
I expect that the drive manufacturers don't give a damn whether they sell firewire or SCSI drives.
On the other hand, I think you're right about controllers. If an NCR 53C875 controller goes for Cdn$115 these days, there's no reason Adaptec should be charging Cdn$250 for their card.
[MacOS X is]
a BSD-based system with the source code now secret.
What's so secret about it? Do you have trouble going to the NetBSD FTP site and downloading the NetBSD-current source? In there you will find a fair number of code commits from Apple developers. Apple not only took NetBSD code and paid people to fix bugs and add features, but they even paid someone to integrate this work back into NetBSD. I fail to see how the GPL would have improved the situation. (In fact, if this code had been GPL'd, Apple probably would just have gone elsewhere for the code, and NetBSD would not have gained what it now has.)
BSD is harming the free software movement more than helping it.
BSD and its license making its code easy to integrate into commercial products was probably the biggest factor in driving TCP/IP to be where it is today. It gave vendors a strong incentive to use that code and stick with fully open standards, rather than buy or write code that implements proprietary standards. If you'd rather that the Internet was much smaller and used Novell IPX, well, you're welcome to your opinion. But I think it's fairly obvious that open standards are even more important than open source, and the BSD licence encourages the promulgation of open standards much more than the GPL. With some programs that doesn't matter; with others it's very important.
In fact, you shouldn't be pointing out these quotes to him because, having gone back and checked out the links, the quotes are misleading about what he said. For example, the quote
Is a Linux takeover likely? Give me a break. Of course not.
is from an article entitled `How Linux Could Kill Windows NT.' In the article he says, essentially, `These things are what Linux needs; if it gets them, Linux can kill NT. I don't think it's likely, but I think it's possible.' (And I'd like to note that two of those three requirements he gave are coming to fruition.)
Personally, I don't agree with his `risk of getting fired' assessment; in my experience in the corporate world, you don't get fired: you simply don't get to install a non-standard technology in the first place. But in his other two quotes, he says Linux is not likely to take over the mainstream. But, as he says in the last quote, he's always admitted the possibility, and this is clear if you read the articles themselves, rather than just the quotes.
So my question here is, why is someone from the Linux community misrepresenting someone who, before Linux really struck the press as a potential mainstream item, was writing articles with titles like `How Linux Could Kill Windows NT'? With folks like this on the team, who needs Microsoft to kill Linux?
Slashdot Mirror
cjs
Oh, and let's not forget: Alan Turing, one of the best and most influential theoriticians in computing the world has ever seen, was a `fag.'
cjs
I'm not interested in chasing any of the anonymous folks off; I am, in fact, interested in what some of them have to say. But Slashdot really needs edited discussion threads. Quite often (as is the case with this one) I don't read any of the responses, because there are just too many of them.
cjs
Basically, it lets you layer two filesystems so that modifications you make get stored on one, but you read old data off the other. One example would be layering a read-write filesystem on top of a CD so that you could `change' the CD, without having to copy the whole CD to a read-write medium. Another would be having different developers with their own private layers above a shared source code directory; a developer can compile, change and delete files, and he'll affect his private workspace but not the other developers (until the changes are moved to the shared layer).
cjs
There's lots of cheap Sparc and other equipment out there if you're willing to live with the older models. (An IPX is dirt cheap these days; on the other hand, an SS5 is a lot more expensive for what you get because it's still a `current' model, even though modern PCs blow it away in terms of speed.)
However, if you want to get good deals and know what you're getting, do some research before you buy. For hardware information, the Sun Hardware FAQ is the biggest collection in one place. Another place worth looking is under the `Supported Hardware' link at the NetBSD project ; there's information there on many different systems, including Suns, and links to other sources of information. (This is also the only modern OS that runs on a lot of old equipment.)
Once you've had at least a cursory look through the resources available, spend a week or two reading through misc.forsale.computers.workstation to get an idea of what a decent price for this stuff is. There are a lot of bad (too high) prices posted there, but these usually get pointed out fairly quickly.
The last thing I'd recommend is to start cheap, to see if this is to your taste. If you decide that working with non-PC equipment is just too much of a hassle, or you get too little performance for your dollar, you're only out a couple of hundred dollars if you bought an IPX, rather than a couple of thousand if you bought an SS20.
Personally, I find old hardware quite rewarding. I currently have running at home 3 Sparcstations (two with rather nice 17" colour monitors that were dirt cheap!), a couple of Sun 3/60s (one colour, one mono, both with 19" monitors) and a VAXStation 2000, and my main mail server for cynic.net is an IPX. I've also had various other things kicking around in the past, though I recently cleaned out my collection. It's fun stuff to play with, and actually does useful work quite well in many circumstances.
cjs
Actually, I was just having a quick look at a copy of The Unix Haters Handbook the other day, since I'd not read it in years. I only read through a couple of dozen pages scattered about the book, but even then, something became apparent to me that never struck me four years ago when I first read it. The implication is around in several places, but it's blatently stated in paragraph 3 of page 128. Essentialy what it says is:
What this system really should do is transfer code from the server to the client and execute it remotely.
Well, looking back on the last few years of Java holes, macro viruses, and the ActiveX disaster, I think we can pretty safely say that this fellow was quite lacking in foresight. So enjoy the book, but take it with a big grain of salt. It's turning out that some of the things that seem horrible about Unix aren't as bad as one might think.
cjs
The OS underlying NeXTStep was originally 4.3BSD. I expect the 4.4ised it at some point, and more recently Apple certainly grabbed a fair amount of NetBSD code to use in it. (This was quite a boon to the NetBSD project, because they contributed back a pile of bugfixes and whatnot.)
cjs
It will be interesting to see where Linux ends up in the long run. Unlike the BSD folks, Linux folks tend to work very hard at proselytisation, and are a lot more likely to work hard at features that will convert Windows users, rather than other Unix users. (I'm not saying that this is better or worse, just different.) Thus, Linux is a lot more well known, despite being younger, and is indeed headed towards being what Windows is today.
cjs
I've looked at these things, but not even tried one, because they seem to be lacking in buttons. I use a four-button mouse with wheel in my right hand, and of the 17 keys I reach with my left hand without moving it, I use 11 quite regularly. Add to that the 12 function keys which I use for team communication, and a few other, harder to reach (they require hand movement) keys for various other things, and a controller without a couple of dozen buttons doesn't look so appealing.
cjs
I expect that the drive manufacturers don't give a damn whether they sell firewire or SCSI drives.
On the other hand, I think you're right about controllers. If an NCR 53C875 controller goes for Cdn$115 these days, there's no reason Adaptec should be charging Cdn$250 for their card.
cjs
cjs
In fact, you shouldn't be pointing out these quotes to him because, having gone back and checked out the links, the quotes are misleading about what he said. For example, the quote
is from an article entitled `How Linux Could Kill Windows NT.' In the article he says, essentially, `These things are what Linux needs; if it gets them, Linux can kill NT. I don't think it's likely, but I think it's possible.' (And I'd like to note that two of those three requirements he gave are coming to fruition.)Personally, I don't agree with his `risk of getting fired' assessment; in my experience in the corporate world, you don't get fired: you simply don't get to install a non-standard technology in the first place. But in his other two quotes, he says Linux is not likely to take over the mainstream. But, as he says in the last quote, he's always admitted the possibility, and this is clear if you read the articles themselves, rather than just the quotes.
So my question here is, why is someone from the Linux community misrepresenting someone who, before Linux really struck the press as a potential mainstream item, was writing articles with titles like `How Linux Could Kill Windows NT'? With folks like this on the team, who needs Microsoft to kill Linux?
cjs