It's not necessarily a bad thing, and if they can do the job for less more power too them. If they can't, someone at IBM will get their ass kicked, and the next round of contracts will be given to US locals.
These U.S. companies never did learn to think even 10 years down the road. If you start exporting high tech jobs to Asia then who is going to be left to buy your products over here?
I, for one, am sick of the anti-Asian xenophobia that rears its head on Slashdot fairly regularly.
I think people have very good reason to hate the exportattion of their jobs to Asians. The Asian workers who are willing to work for what in America is less than minimum wage only serve to drag down everyone along with them instead of demanding better salaries for themselves. The Asian countries never seem to have went through their industrial revolution and unionization era so they don't seem to have many of the laws that protect us against corporations looking for cheap slave labor.
At the time software wasn't optimised for the p6 arch, and it actually ran slower on the ppro than on the high-end pentiums, despite the ppro costing more.
Pentium Pros had the nagging issue of being around the transition period from 16-bit to 32-bit computing in the mainstream. So, if you ran Linux or NT on them they were speedy actually. If you ran Windows95 on it then it was a waste and somewhat slower than a Pentium of the same clock speed. Kind of funny how we really take our CPUs for granted these days. I never thought I would need more speed than my trust 150 MHz Pentium Pro machine.;-)
Heh, welcome to the club. Debian was a bitch to install on that board for me because it kept locking up. I finally got it running long enough to install the base image off the CD and reboot so I could copy over a custom 2.4.21 kernel I had built with APIC and ACPI support removed. After that everything was dandy (including AMD Viper ATA support to get 50MB/sec on the onboard nforce2 ATA controller). I've had some weird lockups since but they seem to be related to my GeForce FX 5600 and OpenGL. That's what we get for buying bleeding edge hardware. I should've just bought an older non-Nforce system with an Athlon XP 1800+ instead of the 3200+.
SuSE is the only "popular" distro that would work with my RAID controller, therfore support may be there but it certainly isn't full.
Which RAID controller is that? Any Linux distro should support any of the SCSI RAID controllers in the kernel, software RAID, as well as the excellent 3Ware ATA RAID controllers. If you're talking about ATA RAID controllers then just buy one from 3ware. They've consistently been friendly to the open source community by providing good GPL'd drivers for their stuff and I would go out of my way to buy their hardware again. If you're talking about the el-cheapo $50 software RAID controllers from Promise or Highpoint then you're wasting everyone's time. Just use Linux's software RAID.. it'd probably be faster.
When will the RIAA and other organizations realize their outmoded distribution methods and crazed sue little girls and old women tactics will not save their business?
When will you guys quit beating this dead horse? The RIAA isn't going ANYWHERE. If anything they've gotten much stronger in the last 5 years. Personally (and I'm sure I'm joined by many people here in this regard) I had never even heard of the RIAA before they sued Napster. I'm also sick of hearing the "time to find a new business model" people. If you were a shopkeeper and people started breaking in and stealing your stuff every day, wouldn't you want to have the criminals punished who are taking your stuff without paying? Why should you have to find a new business model when your current one is current valid... sell stuff people want in exchange for money. Obviously people WANT music or they wouldn't be downloading it for free... they're just too cheap to pay for it. That's what it comes down to in the end. You know it and I know it.
The idea is that whatever goes on out there will be logged/dumped, but never executed/analyzed, on this machine.
Wrong. Go look up the RPC pre-processing and stream4 vulnerabilities in Snort. I will also add that a very common way to configure a network sensor is to have one administration interface on an internal trusted network and the other passive listen-only interface without the IP on the dirty network. With the snort vulnerabilities your machine could become infected and used to reach your internal network. Unless you've got a very very simple network that only needs one sensor with a monitor and keyboard attached you'll need some admin interface on it to reach it to dump logs and change rulesets.
Am I the only one that balks at this statement? Maybe I am missing something but it does seem that even with rx-only you could be infected, just not by any connection oriented protocols? (Or maybe even still if some really strange bug crops up).
Snort had such a bug once or twice within the last year that allowed a remote attacker to execute code as the user snort runs as (usually people run it as root) just by having the sensor listen to the traffic. Quite spiffy.
Wow, that must be some magic access point. Mine can only handle about 30 people before it's saturated. How did they overcome that limitation of 802.11b?
Re:How does this help us, or Sun
on
Solaris 9 x86 Review
·
· Score: 4, Interesting
On the other hand, Solaris scores a big fat negative with patching. Their patching options seriously suck badly. In the Linux world you have great tools like up2date, urpmi, apt-get, etc. In Solaris you have... PatchPro... a horrible piece of crap java based patch management installer that simply does not work. At least, that's been my opinion of trying to use it with Solaris 8. In the end I always end up going back to just downloading the recommended patch cluster every few months, unzipped it, and running install_cluster to keep up to date on patches. Solaris desperately calls for better patch management without requiring you to install some bloated thing like SMC.
Bullshit. Napster didn't prove they weren't inclined to pay for it, even if people wanted to legitimately purchase music downloads, they couldn't.
I thought I'd try any experiment with iTMS recently and wanted to scratch and itch to find Darling Nikki by the Foo Fighters. So I went into iTunes on my Mac, entered the store, and searched for "Darling Nikki". It only came back with two songs, one was by Prince (who did the original) and the other was by someone called Mucky Pup.
So, I opened up Poisoned instead, typed in "Darling Nikki" into the search box and voila. Foo Fighters, Prince, etc. Guess where I downloaded the song from? The lack of availability of songs that I want *right now* make iTMS absolutely worthless for me compared to P2P file sharing networks. If the record labels were to dump every song in their inventory into iTMS annd cooperate fully so I get just as many quality hits on iTMS as I do with using Poisoned then I would've probably plopped down $.99 to satisfy my urge to hear a little Foo Fighters. The sadder thing is, they don't have any Foo Fighters songs at all on iTMS... and no, I don't mean "Too Fighters" iTunes.
P2P is successful because you can go on it anytime day or night and find virtually any song in existence. Add to the fact that it's free and you have an instant hit. If you can add all the songs in the world to iTMS and then lower the price per track to something reasonable like $0.25 per track I think you'd totally hijack the P2P users. I'd easily pay a quarter for good quality compressed tracks instead of taking a chance on getting broken and crappy ripped songs or fake songs put out by the RIAA to flood the networks. $0.99 is pushing my limits though and is reaching my instant impulse to spend money.
When they raised the candy bar prices in the machine from $0.50 to $0.65 I stopped buying candy bars from the machine and instead now buy them in bulk from Sam's club, bring them in to work and we have a candy/pop club. Now the vending machine sits unused by only a few poor fools who are willing to be raped by the price gouging vending company. I've heard rumors that they're going to have to raise the prices of candy bars up to $0.80 soon to cover the lack of sales. What do you think that'll do? I know I'll be rushing out to buy more candy just like when they went from $0.60 cans of 12oz soda to $1.25 plastic bottles of 20oz soda in the machines. I'm sick of being ripped off.
Maybe nowhere. The analysis methodology used could be spoofed by SCO by them running a program on their respective servers that sends out SYN-ACK and SYN-RST to random IP addresses.
OK, come on now for crying out loud. Next thing you're going to say is you can prove the existence of Santa Claus. Sometimes the obvious answer is the truth. Your parents give you gifts as Santa and SCO was really under a DDOS attack.
Sorry for the spoiler kiddies, but you were bound to learn one of these days.
Assuming 1 TB of slow IDE is about $250 right now (I haven'r priced in a while)
Is this post from the mysterious future that the Slashdot front page is always talking about?:-) 300 gigs of slow Maxtor 5400 RPM IDE hard disk is around $279. A terabyte of IDE disks is around $1000, or $1 per gigabyte. A 4 port 3ware IDE raid controller is about $400. Still, not terribly expensive for 1 TB of disk space.. 750 gigs with RAID-5 and 4x250 GB disks.
Anyway, what we really need to get the public interested in spaceflight again is a SSTO nuclear-powered rocket that takes off and lands vertically.
Yea, that's a great idea, but if you could come up with a different word than "nuclear" that would be wonderful. To the environmental-whacko crowd, "nuclear" means kids running away from falling radiation debris. Ridiculous.
they always seem to become an entity unto themselves that just *seems* to be under control, when it reality no two or three guys have enough access and enough experience with the thing to know exactly what's there.
Turns out after AT&T deleted an ex-employee's porn, mp3, and warez stash he was hiding in his own personal table they were able to optimize the database down to about 3GB of customer billing data. You just can't find good help these days.
The UN is a waste. How are you going to get a group of people representing every ethnicity, religon and country on earth to agree on anything?
Conquer them. Yes, I'm being serious. The U.S.A. should embark on a plan of world domination through subversive cultural methods (this is well underway since the mid 20th century) and economic control.
Don't things like XMLTV use web sites that sometimes block IP's from using them?
I've been updating nightly from Zap2it's website via XMLTV nightly and they haven't blocked me yet. Personally I wouldn't mind if they offered a pay service for $5/month for guide information. I'd buy it for my MythTV box just to get the information in a format that doesn't require me to hack around updating XMLTV every month or so when zap2it changes their website around. This gets to be a pain in the ass.
The important thing for me is that the WAF is high (wife acceptance factor).
I guess my WAF is pretty high. I wanted to take mine down to rebuild it with the PVR-250 cards and erase whatever it has recorded (about 500 shows) but my wife freaked out and offered to buy me a DVD writer for Christmas if I'd back up her shows. I reluctantly accepted her offer.:-/ I guess it got accepted pretty good for what I had originally setup as a toy project to catch Seinfeld episodes I had missed in first-run.
IANAL, and I'm no kernel expert, but AFAIK -- kernel module sources do not have to be released (and if I'm wrong, IMO, they shouldn't have to be).
That's why I linked to the article where Linus Torvalds sets the record straight on binary modules. I would imagine he qualifies as a kernel expert.:-)
AOL acquired Nullsoft? Yay. Now I can look forward to my music client being plastered with icon smilies and have it crash whenever I play a malformed MP3.
AOL acquired Nullsoft a long time ago. Well before their version 3 release... which coincidently is said to be their worst release ever. Hmmm. Maybe you have a point.
If Tivo has commercially-sensitive proprietary kernel modules, they are not obliged to distribute them, and AFAIK, the Tivo filesystem code falls into this category
That's not what Linus said in this earlier Slashdot story from Monday. It seems pretty unbelievable that TiVo developed their "proprietary kernel modules" for whatever hardware they're using without any knowledge of the kernel internals or intention to link the resulting binary module to the kernel. IANAL of course, but from reading through Linus' postings it seems like he thinks this kind of situation would require TiVo to release the source code to any binary kernel modules as well since they are derived from the GPL'd Linux kernel. I'm not familiar with TiVos in particular, but does anyone have any proof that they've made no derivative works from userland code or the Linux kernel without releasing source code to the modifications?
Personally I think TiVo should have to distribute the source code to their product so that people can choose whether or not they want to buy the service or would prefer to just write their own interface to guide information using XmlTV. One of the main reasons I wouldn't buy a TiVo is I don't want to be tied to one company's guide information. If they fold I am screwed and my TiVo would be useless. It almost happened to ReplayTV users.
They are however required to provided the source of the GPLed stuff on the image seperately, on request, but this is probably not very interesting anyway
This seems to be an odd provision of the GPL. Since the components they use are undoubtable widely available in source form at other sites, why should they be required to provide a mirror of it? Pointing you to a mirror site of the source packages should be sufficient for most reasonable people.
God knows Red Hat is the only Linux distributor. Too bad nobody else can pick up their code and support it. Oh wait...
I know you're trying to be funny, but when the software you want to run is things like Oracle that require a certain version of a certain distribution to retain your Oracle support contract then you can't go straying far from the flock. Sure, you might get Oracle installed on Debian or Slackware, but if you have ANY problems they will not support you. The only supported distributions are Red Hat and UnitedLinux (SuSE I guess).
Slashdot Mirror
These U.S. companies never did learn to think even 10 years down the road. If you start exporting high tech jobs to Asia then who is going to be left to buy your products over here?
I, for one, am sick of the anti-Asian xenophobia that rears its head on Slashdot fairly regularly.
I think people have very good reason to hate the exportattion of their jobs to Asians. The Asian workers who are willing to work for what in America is less than minimum wage only serve to drag down everyone along with them instead of demanding better salaries for themselves. The Asian countries never seem to have went through their industrial revolution and unionization era so they don't seem to have many of the laws that protect us against corporations looking for cheap slave labor.
Pentium Pros had the nagging issue of being around the transition period from 16-bit to 32-bit computing in the mainstream. So, if you ran Linux or NT on them they were speedy actually. If you ran Windows95 on it then it was a waste and somewhat slower than a Pentium of the same clock speed. Kind of funny how we really take our CPUs for granted these days. I never thought I would need more speed than my trust 150 MHz Pentium Pro machine. ;-)
Doesn't matter, but it was probably more likely a Pentium Pro 150. Much more common.
Heh, welcome to the club. Debian was a bitch to install on that board for me because it kept locking up. I finally got it running long enough to install the base image off the CD and reboot so I could copy over a custom 2.4.21 kernel I had built with APIC and ACPI support removed. After that everything was dandy (including AMD Viper ATA support to get 50MB/sec on the onboard nforce2 ATA controller). I've had some weird lockups since but they seem to be related to my GeForce FX 5600 and OpenGL. That's what we get for buying bleeding edge hardware. I should've just bought an older non-Nforce system with an Athlon XP 1800+ instead of the 3200+.
Which RAID controller is that? Any Linux distro should support any of the SCSI RAID controllers in the kernel, software RAID, as well as the excellent 3Ware ATA RAID controllers. If you're talking about ATA RAID controllers then just buy one from 3ware. They've consistently been friendly to the open source community by providing good GPL'd drivers for their stuff and I would go out of my way to buy their hardware again. If you're talking about the el-cheapo $50 software RAID controllers from Promise or Highpoint then you're wasting everyone's time. Just use Linux's software RAID.. it'd probably be faster.
When will you guys quit beating this dead horse? The RIAA isn't going ANYWHERE. If anything they've gotten much stronger in the last 5 years. Personally (and I'm sure I'm joined by many people here in this regard) I had never even heard of the RIAA before they sued Napster. I'm also sick of hearing the "time to find a new business model" people. If you were a shopkeeper and people started breaking in and stealing your stuff every day, wouldn't you want to have the criminals punished who are taking your stuff without paying? Why should you have to find a new business model when your current one is current valid... sell stuff people want in exchange for money. Obviously people WANT music or they wouldn't be downloading it for free... they're just too cheap to pay for it. That's what it comes down to in the end. You know it and I know it.
Wrong. Go look up the RPC pre-processing and stream4 vulnerabilities in Snort. I will also add that a very common way to configure a network sensor is to have one administration interface on an internal trusted network and the other passive listen-only interface without the IP on the dirty network. With the snort vulnerabilities your machine could become infected and used to reach your internal network. Unless you've got a very very simple network that only needs one sensor with a monitor and keyboard attached you'll need some admin interface on it to reach it to dump logs and change rulesets.
Snort had such a bug once or twice within the last year that allowed a remote attacker to execute code as the user snort runs as (usually people run it as root) just by having the sensor listen to the traffic. Quite spiffy.
Wow, that must be some magic access point. Mine can only handle about 30 people before it's saturated. How did they overcome that limitation of 802.11b?
On the other hand, Solaris scores a big fat negative with patching. Their patching options seriously suck badly. In the Linux world you have great tools like up2date, urpmi, apt-get, etc. In Solaris you have... PatchPro... a horrible piece of crap java based patch management installer that simply does not work. At least, that's been my opinion of trying to use it with Solaris 8. In the end I always end up going back to just downloading the recommended patch cluster every few months, unzipped it, and running install_cluster to keep up to date on patches. Solaris desperately calls for better patch management without requiring you to install some bloated thing like SMC.
I thought I'd try any experiment with iTMS recently and wanted to scratch and itch to find Darling Nikki by the Foo Fighters. So I went into iTunes on my Mac, entered the store, and searched for "Darling Nikki". It only came back with two songs, one was by Prince (who did the original) and the other was by someone called Mucky Pup.
So, I opened up Poisoned instead, typed in "Darling Nikki" into the search box and voila. Foo Fighters, Prince, etc. Guess where I downloaded the song from? The lack of availability of songs that I want *right now* make iTMS absolutely worthless for me compared to P2P file sharing networks. If the record labels were to dump every song in their inventory into iTMS annd cooperate fully so I get just as many quality hits on iTMS as I do with using Poisoned then I would've probably plopped down $.99 to satisfy my urge to hear a little Foo Fighters. The sadder thing is, they don't have any Foo Fighters songs at all on iTMS... and no, I don't mean "Too Fighters" iTunes.
P2P is successful because you can go on it anytime day or night and find virtually any song in existence. Add to the fact that it's free and you have an instant hit. If you can add all the songs in the world to iTMS and then lower the price per track to something reasonable like $0.25 per track I think you'd totally hijack the P2P users. I'd easily pay a quarter for good quality compressed tracks instead of taking a chance on getting broken and crappy ripped songs or fake songs put out by the RIAA to flood the networks. $0.99 is pushing my limits though and is reaching my instant impulse to spend money.
When they raised the candy bar prices in the machine from $0.50 to $0.65 I stopped buying candy bars from the machine and instead now buy them in bulk from Sam's club, bring them in to work and we have a candy/pop club. Now the vending machine sits unused by only a few poor fools who are willing to be raped by the price gouging vending company. I've heard rumors that they're going to have to raise the prices of candy bars up to $0.80 soon to cover the lack of sales. What do you think that'll do? I know I'll be rushing out to buy more candy just like when they went from $0.60 cans of 12oz soda to $1.25 plastic bottles of 20oz soda in the machines. I'm sick of being ripped off.
OK, come on now for crying out loud. Next thing you're going to say is you can prove the existence of Santa Claus. Sometimes the obvious answer is the truth. Your parents give you gifts as Santa and SCO was really under a DDOS attack.
Sorry for the spoiler kiddies, but you were bound to learn one of these days.
Is this post from the mysterious future that the Slashdot front page is always talking about? :-) 300 gigs of slow Maxtor 5400 RPM IDE hard disk is around $279. A terabyte of IDE disks is around $1000, or $1 per gigabyte. A 4 port 3ware IDE raid controller is about $400. Still, not terribly expensive for 1 TB of disk space.. 750 gigs with RAID-5 and 4x250 GB disks.
Yea, that's a great idea, but if you could come up with a different word than "nuclear" that would be wonderful. To the environmental-whacko crowd, "nuclear" means kids running away from falling radiation debris. Ridiculous.
Turns out after AT&T deleted an ex-employee's porn, mp3, and warez stash he was hiding in his own personal table they were able to optimize the database down to about 3GB of customer billing data. You just can't find good help these days.
Conquer them. Yes, I'm being serious. The U.S.A. should embark on a plan of world domination through subversive cultural methods (this is well underway since the mid 20th century) and economic control.
I've been updating nightly from Zap2it's website via XMLTV nightly and they haven't blocked me yet. Personally I wouldn't mind if they offered a pay service for $5/month for guide information. I'd buy it for my MythTV box just to get the information in a format that doesn't require me to hack around updating XMLTV every month or so when zap2it changes their website around. This gets to be a pain in the ass.
What's wrong with analog? Everything I record is on broadcast TV channels anyway so I couldn't care less.
I guess my WAF is pretty high. I wanted to take mine down to rebuild it with the PVR-250 cards and erase whatever it has recorded (about 500 shows) but my wife freaked out and offered to buy me a DVD writer for Christmas if I'd back up her shows. I reluctantly accepted her offer. :-/ I guess it got accepted pretty good for what I had originally setup as a toy project to catch Seinfeld episodes I had missed in first-run.
That's why I linked to the article where Linus Torvalds sets the record straight on binary modules. I would imagine he qualifies as a kernel expert. :-)
AOL acquired Nullsoft a long time ago. Well before their version 3 release... which coincidently is said to be their worst release ever. Hmmm. Maybe you have a point.
That's not what Linus said in this earlier Slashdot story from Monday. It seems pretty unbelievable that TiVo developed their "proprietary kernel modules" for whatever hardware they're using without any knowledge of the kernel internals or intention to link the resulting binary module to the kernel. IANAL of course, but from reading through Linus' postings it seems like he thinks this kind of situation would require TiVo to release the source code to any binary kernel modules as well since they are derived from the GPL'd Linux kernel. I'm not familiar with TiVos in particular, but does anyone have any proof that they've made no derivative works from userland code or the Linux kernel without releasing source code to the modifications?
Personally I think TiVo should have to distribute the source code to their product so that people can choose whether or not they want to buy the service or would prefer to just write their own interface to guide information using XmlTV. One of the main reasons I wouldn't buy a TiVo is I don't want to be tied to one company's guide information. If they fold I am screwed and my TiVo would be useless. It almost happened to ReplayTV users.
This seems to be an odd provision of the GPL. Since the components they use are undoubtable widely available in source form at other sites, why should they be required to provide a mirror of it? Pointing you to a mirror site of the source packages should be sufficient for most reasonable people.
At the rate the Debian distributions come out, Sid will never become the stable distribution within my lifetime.
Yes, it's a joke if you're a Debian geek. If not you probably won't understand.
I know you're trying to be funny, but when the software you want to run is things like Oracle that require a certain version of a certain distribution to retain your Oracle support contract then you can't go straying far from the flock. Sure, you might get Oracle installed on Debian or Slackware, but if you have ANY problems they will not support you. The only supported distributions are Red Hat and UnitedLinux (SuSE I guess).