Maybe because IBM has a few billion invested in Linux since y2k, and a pending lawsuit? Do you think thry would just gamble on that let alone publish the results?
No conspiracy at all, just profit margins . I don't have to worry about that as an individual, so I do OK and still save with decent parts. The hour or so of labor to assemble it all isn't much to me. IOW, I agree - pick decent parts and get *exactly* what you want. I usually pick the previous generation CPU and get the biggest mobo I can for that from trusted brands. Then I stuff the mobo with the most it can handle, which is a *lot* nowdays. Of course, I get it all below retail from local OEM's, cash paid in person. It seems the "big boys" have all kinds of custom stuff done that makes life hard, but the "white box" stuff is easier/more flexible.
To further reinforce your point, I built this machine 2 yrs ago for $1200 USD. The equivalent Dell Poweredge was $2200 USD. This is for SMP with big RAM, and yes it plays very nice with linux.
Weird; I've always felt that AMD was the value leader, if you don't need Genuine Intel for some reason. I may be wrong, but could the difficulty of clocking AMD vs Intel be because AMD is already optimized so much nowdays? I don't have any recent experience with it, but that's what I always thought. Could someone explain to a geek who hasn't used them since the K6?
"I remember visiting the Univac every Saturday morning... Sorting cards occasionally..." Wow, thanks for the memories. When I was 6 yrs old, my Dad was professor of engineering at Illinois and had an account on the computer there. We used to go there on weekends; they would run their punch card programs and start a game of poker (complete with cigars/beer) while the computer crunched calculus. Every year they bitched about how expensive the computer was, often hundreds of dollars for a semester of access! During finals week, the machine was slow enough to go to the football game and get a haircut. We would come back and sign in with the secretary and security guards. The guys would collect their results in boxes of 17-inch green-bar paper.
It blew my mind that you could represent things with punched holes, like they showed me. Years later, the Atari and the PC came out. That old 7090 was the one that blew my mind, but the Atari was the first one that I could actually use myself.
BTW, I'm the rugrat who dumped the pencil sharpener into the card reader... sorry 'bout that.
Makes me wonder whatever happened to some of those old games. In another post I said I had an atari 800, but my Dad used a Kaypro 2. It had a neat game called "Where in time is Carmen SanDiego". Also GEM desktop. All text based, you had to answer the questions about what to do... a green screen and 5-1/4 floppies.
Does anyone know who owns the rights to this game? Were there any ports? I'd love to find an actual working Kaypro and s/w for it (CP/M-80, anyone?) or at least a *legal* way to do that game on Linux. Ideas?
Wow, thanks for the memories! We got an Atari 800 with the cassette tape, ROM cartridges, and a pair of joysticks IIRC back in 1983. That was when I *really* got interested in computers, even tho I had used them before. I wasted many happy hours playing with the ROM BASIC and trying to beat it at chess. Dad had a Kaypro 2 for work with CP/M and a super-hot Z-80 chip. He used it mostly for work, but it had a Blackjack game, among others. All in a neat aluminum case with a 7-inch green CRT. And I *still* use the Panasonic KX-P 1090 dot-matrix from back then.
OK, you have some good points. I think that I was both right and wrong in a few ways. DISCLAIMER: I think both Windows *and* Linux need to get some sane security policies.:D
"Thus your machine is reasonably atypical even for a managed linux box, let alone one being used as a single-user desktop for an ignorant end user like the average Windows machine." I don't necessarily have any point of reference for being "atypical"; I just RTFM a lot and do what seems to make sense. I'll agree that typical Windows users are ignorant, but not always dumb. One of them is sleeping a few steps away from me, with a PhD in engineering, and expects things to "just work" when you click "Install". IMHO, this is a social failure (Social engineering hacks, anyone?) more than anything, a failure to meet expectations or teach (Marketing failure anyone? Anyone?) reasonable expectations/remedies. Which leads to my next point: I'm probably in the other 1% because a worm won't convince me to run it. Pro/actively managing your machine means using only trusted, known sources, IMHO. Hence, GPG keys for everything, whitelists, and source code for system-level stuff. Of course,/home/~ is backed up, and.bashrc is only relevant if you insist on bash. Other shells and utils are available on read-only media.
"We aren't trying to compare against the Debian machines, we're trying to compare against the typical Windows box - directly connected to the internet, unmanaged and..." True. I was making a bad comparison. All I can say is that the typical Linux installation defaults aren't much better than WinXP, IMHO. I was simply trying to point out that Windows can be attacked at the end-user level, whereas a Linux attack might be more successful at the source/distro level.
"Anything a normal user can do, a worm can do. Everything a worm needs to do, a normal user can do. Every tool (and usually far, far more) a worm needs to do its work, is installed on the average linux box." True. However, that worm first needs to have normal user priveleges, which it won't gain without becoming root *first* (man useradd).
Yes Linux probably *will* get nailed by a worm in a few years. There's plenty of flaws in the user-space code. However, I think it will happen because so many people have jumped on the bandwagon in the last few years, with broadband connections. Sort of like Win9.x. Then wash, rinse, repeat - with another OS.
My new conclusion: Both Windows *and* Linux are open to attack, but in completely different ways. Windows can be attacked directly at the end-user, as demonstrated by SoBig, Blaster, etc. Linux would be easier to attack at the source/distro level, if it isn't caught by peer review. "Many eyes make all bugs shallow". IMHO, attacking Linux in the same way as Windows is sheer folly, unless you *enjoy* writing indirect inode meta-data and hoping the user reboots.
"What else do you need ?" You need an account with permission to run said binaries, at least on my machines. This assumes that I modify the "default deny" policy and make an exception for you. Of course, that policy was implemented before it was *ever* on a network
If you somehow manage to penetrate *without* an account, you'll still have to deal with system accounts having a home directory of/dev/null, and some creative usage of things like chattr [1], chmod, and tripwire. Oh, and check out "man last[1]".
Bummer it *is* fully patched, and nmap only finds the printer - which isn't listening. Having a slightly paranoid owner doesn't hurt this machine.
My conclusion: whoever attacked the Debian and GNU machines had a damn good chance of succeeding. IMHO the single best way to spread malware in linux would be to compromise a distro or source project. I can't see malware affecting end users in a large way otherwise - there's too many variables.
Actually I kinda hate to say this, but I *was* feeling sorry for my Windows-using friends and family. I mean, downtime and screwups totally suck for everyone, no matter *what* system they use.
Since I use linux, I downloaded all the updates offered for Win98SE, and the SP's and patches for XP Home.
I have a separate machine for testing Windows stuff, which I installed from new and use it for testing the patches. It has never been on a network of any kind. I run Norton AV on the thing just to be sure.
If it all checks out OK, I burn them onto separate CD's and give them to people, telling them to "Install these updates before you go on a network *at all*."
Mainly, I do all this just because it sux0rs to spend your holidays doing unpaid tech support for everyone you know...
"DMCA don't mean shit, here, though. It's all about POSIX (and the C-standards for ctype.h). POSIX lists the posix-compatable system calls, return types, and constant declarations."
Very well said, I totally agree. Being that the Open Group owns the UNIX trademark and sets the POSIX standards, I have to wonder if SCO might be stepping on their toes a bit with this. Not good. It'll be interesting to see if the Open Group gets pulled into this. AFAIK, Novell owns the copyrights and maybe a few patents, which it purchased from USL after the BSD case. The trademark (and OSF/1 and POSIX oversight) went to Open Group, and the codebase went to the "old" SCO.
Except in the west, the cost of goods *hasn't* gone down, as far as I can see. How long has it been since NAFTA? I'm non-union, and I *still* opposed it.
I totally agree, but I have to differ with the AC's post about economic lag and the causes of it. Mainly, I believe that the lag is caused by bureaucracy (taxes, fees, regulations, profit-taking, overpriced management - it went to their heads) in both the private sector and in gov't. Further, I don't think it will change until forced to do so by social and economic pressure. For a real-world example, Levi Strauss has *no* manufacturing in the US anymore, yet I notice the price in the US certainly hasn't dropped. And yes, I spend too much time worrying about my [insert payment schedule here] bills. You can guess the rest.
The "negative" of what you last printed remains on the drum. When you say that it just "spits out powder" you are correct - but it gets spit out into english. (or whatever kind of info.) I have only a basic knowledge of how they work, but I've seen it before when changing my own toner, and I have no reason to doubt this.
Taking it a step further... some family members work for DoD contractors. They have a system where used toner cartridges are accounted for before incinerating them because a bit of skill can retrieve the last few pages from them. Same for media such as CD's and HDD's. The machines these parts come from are locked in a bank vault with *no* networking, no portable devices allowed, etc.
I can vouch for the effectiveness of dumpster diving; I snarfed the entire budget info for the science dept. in college once. Interesting reading, too.
There's an article at CNN/Money that says "The jobs most at risk require fewer skills, are automated, or are highly portable. (emphasis mine).
In my case: I'm a skilled US steelworker, trained at own expense (welder/fabrication) and I've seen my career degraded by management continually pushing the desired skill level down to nil over the last 15 years. Enter foreign competition during the same time. Recently (last year) I started going back to school for comp sci.
I now believe that [begin sarcasm] it would be OK to flip burgers and mop floors for 80 hrs a week if only Uncle Sam didn't call it "middle class".[end sarcasm]
Seriously, I don't think this kind of crap will end until the economy implodes under the weight at the top. Until then, there will be fewer and fewer "middle class" people who can afford the products/services so hyped -
gtg now before I get violently pissed, even. And BTW, I'm a non-union republican feeling your pain ATM.
Vegas would be awesome! Can you imagine the local hooker's union giving them 1/2 day seminars on how to deal with their new geeky / nerdly friends!!?? I can't wait...
I got AKPM's -mm1 patch for -test11 earlier, and was reading today's LKML while waiting for a few *big* files elsewhere (Intel manuals). Checked back here, and found out about 2.6.
Well, huzzah to the kernel team, I've enjoyed their work for enough years. Not much champagne available here, but a heartfelt and lukewarm Milwaukee piss (offered).
I've been using 2.5.x and -test kernels off and on here, and its definitely a step in the right direction even for my humble desktop, IMHO. If I was to be bold I'd even say that 2.6 is a positive change (for users) in the same way that 2.0 was. Just based on the scheduling and device support, SMP (I use it), bigmem, etc.
And no, I'm not really worried about the SCO/IBM thing - the outcome won't change my opinions or Linux usage patterns an iota.
Hrmmm, that's the same reason why I switched several years ago, so I agree totally with you on that point. The convergence will happen, I'm sure. Thanks for reminding me "why" I did it, because people always ask.
Here's an odd thing I've noticed: I end up doing tech support for friends, family, and co-workers. Sometimes they ask what version of Windows I use, and I tell them that I don't use Windows at all. Then they say "But I thought you liked computers!"
Slashdot Mirror
Maybe because IBM has a few billion invested in Linux since y2k, and a pending lawsuit? Do you think thry would just gamble on that let alone publish the results?
IOW, I agree - pick decent parts and get *exactly* what you want. I usually pick the previous generation CPU and get the biggest mobo I can for that from trusted brands. Then I stuff the mobo with the most it can handle, which is a *lot* nowdays. Of course, I get it all below retail from local OEM's, cash paid in person.
It seems the "big boys" have all kinds of custom stuff done that makes life hard, but the "white box" stuff is easier/more flexible.
To further reinforce your point, I built this machine 2 yrs ago for $1200 USD. The equivalent Dell Poweredge was $2200 USD. This is for SMP with big RAM, and yes it plays very nice with linux.
'Nuff said.
Weird; I've always felt that AMD was the value leader, if you don't need Genuine Intel for some reason. I may be wrong, but could the difficulty of clocking AMD vs Intel be because AMD is already optimized so much nowdays? I don't have any recent experience with it, but that's what I always thought. Could someone explain to a geek who hasn't used them since the K6?
Wow, thanks for the memories. When I was 6 yrs old, my Dad was professor of engineering at Illinois and had an account on the computer there. We used to go there on weekends; they would run their punch card programs and start a game of poker (complete with cigars/beer) while the computer crunched calculus. Every year they bitched about how expensive the computer was, often hundreds of dollars for a semester of access!
During finals week, the machine was slow enough to go to the football game and get a haircut. We would come back and sign in with the secretary and security guards. The guys would collect their results in boxes of 17-inch green-bar paper.
It blew my mind that you could represent things with punched holes, like they showed me. Years later, the Atari and the PC came out. That old 7090 was the one that blew my mind, but the Atari was the first one that I could actually use myself.
BTW, I'm the rugrat who dumped the pencil sharpener into the card reader... sorry 'bout that.
Does anyone know who owns the rights to this game? Were there any ports? I'd love to find an actual working Kaypro and s/w for it (CP/M-80, anyone?) or at least a *legal* way to do that game on Linux. Ideas?
Yep those were fer damn sure the DAYS!
Wow, thanks for the memories! We got an Atari 800 with the cassette tape, ROM cartridges, and a pair of joysticks IIRC back in 1983. That was when I *really* got interested in computers, even tho I had used them before. I wasted many happy hours playing with the ROM BASIC and trying to beat it at chess. Dad had a Kaypro 2 for work with CP/M and a super-hot Z-80 chip. He used it mostly for work, but it had a Blackjack game, among others. All in a neat aluminum case with a 7-inch green CRT. And I *still* use the Panasonic KX-P 1090 dot-matrix from back then.
one of these. Got it for christmas 2 yrs ago, sees daily use, and never a problem.
at least now I could have my silicon pre-strained, instead of having all those Viagara spams do it...
Yeah its "OldHat" (pun intended), but last I checked, my dialup ISP was a profitable family biz. OK, not huge $$$ or anything, but decent.
"Thus your machine is reasonably atypical even for a managed linux box, let alone one being used as a single-user desktop for an ignorant end user like the average Windows machine." /home/~ is backed up, and .bashrc is only relevant if you insist on bash. Other shells and utils are available on read-only media.
I don't necessarily have any point of reference for being "atypical"; I just RTFM a lot and do what seems to make sense. I'll agree that typical Windows users are ignorant, but not always dumb. One of them is sleeping a few steps away from me, with a PhD in engineering, and expects things to "just work" when you click "Install". IMHO, this is a social failure (Social engineering hacks, anyone?) more than anything, a failure to meet expectations or teach (Marketing failure anyone? Anyone?) reasonable expectations/remedies. Which leads to my next point:
I'm probably in the other 1% because a worm won't convince me to run it. Pro/actively managing your machine means using only trusted, known sources, IMHO. Hence, GPG keys for everything, whitelists, and source code for system-level stuff. Of course,
"We aren't trying to compare against the Debian machines, we're trying to compare against the typical Windows box - directly connected to the internet, unmanaged and ..."
True. I was making a bad comparison. All I can say is that the typical Linux installation defaults aren't much better than WinXP, IMHO. I was simply trying to point out that Windows can be attacked at the end-user level, whereas a Linux attack might be more successful at the source/distro level.
"Anything a normal user can do, a worm can do. Everything a worm needs to do, a normal user can do. Every tool (and usually far, far more) a worm needs to do its work, is installed on the average linux box."
True. However, that worm first needs to have normal user priveleges, which it won't gain without becoming root *first* (man useradd).
Yes Linux probably *will* get nailed by a worm in a few years. There's plenty of flaws in the user-space code. However, I think it will happen because so many people have jumped on the bandwagon in the last few years, with broadband connections. Sort of like Win9.x. Then wash, rinse, repeat - with another OS.
My new conclusion: Both Windows *and* Linux are open to attack, but in completely different ways. Windows can be attacked directly at the end-user, as demonstrated by SoBig, Blaster, etc. Linux would be easier to attack at the source/distro level, if it isn't caught by peer review. "Many eyes make all bugs shallow". IMHO, attacking Linux in the same way as Windows is sheer folly, unless you *enjoy* writing indirect inode meta-data and hoping the user reboots.
You need an account with permission to run said binaries, at least on my machines. This assumes that I modify the "default deny" policy and make an exception for you. Of course, that policy was implemented before it was *ever* on a network
If you somehow manage to penetrate *without* an account, you'll still have to deal with system accounts having a home directory of /dev/null, and some creative usage of things like chattr [1], chmod, and tripwire. Oh, and check out "man last[1]".
Bummer it *is* fully patched, and nmap only finds the printer - which isn't listening. Having a slightly paranoid owner doesn't hurt this machine.
My conclusion: whoever attacked the Debian and GNU machines had a damn good chance of succeeding. IMHO the single best way to spread malware in linux would be to compromise a distro or source project. I can't see malware affecting end users in a large way otherwise - there's too many variables.
Since I use linux, I downloaded all the updates offered for Win98SE, and the SP's and patches for XP Home.
I have a separate machine for testing Windows stuff, which I installed from new and use it for testing the patches. It has never been on a network of any kind. I run Norton AV on the thing just to be sure.
If it all checks out OK, I burn them onto separate CD's and give them to people, telling them to "Install these updates before you go on a network *at all*."
Mainly, I do all this just because it sux0rs to spend your holidays doing unpaid tech support for everyone you know...
Here are some of the lesser known, but still useful Victorian robots...
Very well said, I totally agree. Being that the Open Group owns the UNIX trademark and sets the POSIX standards, I have to wonder if SCO might be stepping on their toes a bit with this. Not good. It'll be interesting to see if the Open Group gets pulled into this. AFAIK, Novell owns the copyrights and maybe a few patents, which it purchased from USL after the BSD case. The trademark (and OSF/1 and POSIX oversight) went to Open Group, and the codebase went to the "old" SCO.
Except in the west, the cost of goods *hasn't* gone down, as far as I can see. How long has it been since NAFTA? I'm non-union, and I *still* opposed it.
I totally agree, but I have to differ with the AC's post about economic lag and the causes of it. Mainly, I believe that the lag is caused by bureaucracy (taxes, fees, regulations, profit-taking, overpriced management - it went to their heads) in both the private sector and in gov't. Further, I don't think it will change until forced to do so by social and economic pressure. For a real-world example, Levi Strauss has *no* manufacturing in the US anymore, yet I notice the price in the US certainly hasn't dropped. And yes, I spend too much time worrying about my [insert payment schedule here] bills. You can guess the rest.
The "negative" of what you last printed remains on the drum. When you say that it just "spits out powder" you are correct - but it gets spit out into english. (or whatever kind of info.) I have only a basic knowledge of how they work, but I've seen it before when changing my own toner, and I have no reason to doubt this.
I can vouch for the effectiveness of dumpster diving; I snarfed the entire budget info for the science dept. in college once. Interesting reading, too.
In my case: I'm a skilled US steelworker, trained at own expense (welder/fabrication) and I've seen my career degraded by management continually pushing the desired skill level down to nil over the last 15 years. Enter foreign competition during the same time. Recently (last year) I started going back to school for comp sci.
I now believe that [begin sarcasm] it would be OK to flip burgers and mop floors for 80 hrs a week if only Uncle Sam didn't call it "middle class".[end sarcasm]
Seriously, I don't think this kind of crap will end until the economy implodes under the weight at the top. Until then, there will be fewer and fewer "middle class" people who can afford the products/services so hyped -
gtg now before I get violently pissed, even. And BTW, I'm a non-union republican feeling your pain ATM.
Vegas would be awesome! Can you imagine the local hooker's union giving them 1/2 day seminars on how to deal with their new geeky / nerdly friends!!?? I can't wait...
".... switching to the 2-ply tinfoil for my hat." I'm experimenting with concentric stainless salad bowls. Now say that 3 times fast!
This is more like going from Win85 to 2k server.
Agreed. In fact, HELL YES! On my SMP desktop, I had to decelerate the mouse movement by ~25% because it was so much quicker.
Well, huzzah to the kernel team, I've enjoyed their work for enough years. Not much champagne available here, but a heartfelt and lukewarm Milwaukee piss (offered).
I've been using 2.5.x and -test kernels off and on here, and its definitely a step in the right direction even for my humble desktop, IMHO. If I was to be bold I'd even say that 2.6 is a positive change (for users) in the same way that 2.0 was. Just based on the scheduling and device support, SMP (I use it), bigmem, etc.
And no, I'm not really worried about the SCO/IBM thing - the outcome won't change my opinions or Linux usage patterns an iota.
Here's an odd thing I've noticed: I end up doing tech support for friends, family, and co-workers. Sometimes they ask what version of Windows I use, and I tell them that I don't use Windows at all. Then they say "But I thought you liked computers!"
I've never had a good answer for that one.