Yup, same here. We have over 300 servers and a couple of us in our dept. insisted that we pay Red Hat instead of going CentOS only, which we easily could've done. Never use the support but also know that Linux would be in a different position today if it weren't for Red Hat's work.
Redhat is actively recommending you don't use their virtualization product yet for production
They are?! That's not the signals I'm seeing from Red Hat. The Virtualization capability of RHEL5 is plastered all over their website and our sales/engineering reps have yet to say hold back on rolling it out.
Eh, the gist of the article was the work conditions for IT jobs were not appealing to many women and that women are leaving the field as a result, even though some companies and organizations WANT MORE women in those jobs and are trying to recruit more of them. It's not just women having kids, its the idea that you're never really away from the immediate, pressing needs of your employer and their demands, and that some women have said enough. So the female side of the IT work force might not grow, it could shrink even more. Hell, even some of us men think IT work conditions suck.
Close, but you are wrong on one point. The fact that Sun systems *can* hold tons of RAM is the only reason that Solaris/SPARC is still in the EDA game. Once Opteron/RHEL3 based systems that address 32+ GB of RAM are shipping in volume, I can't see a single chip design house opting for a Sun proprietary system except for legacy support.
Re:McDonalds and SCO
on
SCOoby Snacks
·
· Score: 2, Funny
However, those students doing well in theory classes would always do well in the lab (even if it took them a little longer, and a bit more work).
Which costs employers extra money in the real world, and what the professors are trying to address. I like their idea of balancing the theory more than it currently is. I wish my professors had done something similiar when I was in school. I know I would've benefitted from it.
Dell offered the best hardware and support for the price and also they do price matching so he got quite a few things cheaper than expected.
Maybe the best price/performance hardware support, but their Linux software support leaves a lot to be desired (even at the Gold contract level). We bought several Dell PE2650 servers running Linux and I'm finding that the Dell support techs just don't have enough real world experience with Linux to make Dell into a big Unix player (yet). I'm told Dell is working on correcting that as I type this, but until they do Dell won't be as much of an option for those of us who run Unix shops and know what quality support comes from Sun. Anyone can read a manual--including me--when I call Dell (or Sun or HP), I want to talk to someone who knows more than that.
Surely Sun can't exactly sell the hardware for any cheaper than it can already be bought for, so what's the advantage of choosing them over a company like Dell?
Why not? Think Dell does anything Sun can't do in designing an x86 system? I don't. Sun engineers design the server, then Sun contracts with some of the same manufacturers other x86 vendors use to have them built. It's not as difficult as it may seem. One of the great assests of building x86 systems is the off the shelf nature of the components. That reduces the learning curve considerably when compared to designing everything yourself for a system that isn't as widely used, i.e. sparc.
I welcome Sun's effort to ship better Linux servers. When you consider how much Sun knows about Unix, it's great to have that expertise spilling into the Linux world.
Guess you weren't around when she joined the phoebe (RHL9 beta) mailing list...she wasn't a subscriber for very long, but man did she start up one hell of a flame fest in her short time there.
Last I heard, the redhat cds contained proprietary software. They do contain plenty of GPL'd stuff, but redhat adds a bunch of non-GPL'd things in.
This is wrong. The Red Hat CDs have trademarked logos in them, not proprietary software. Those trademarks do not prevent you (as a Linux enthusiast and community member, not as a for profit business) from freely distributing the ISOs and software in exactly the same form you downloaded them. Red Hat provides an explanation of this policy on their website.
Where I work, I'm seeing Athlon based servers outperform UltraSPARC III servers (v480s to be exact) by a factor of 2+. Physical compile time for new chip designs has been cut by more than half since the Athlon boxes were brought online, and Athlon stability hasn't been an issue at all.
Sun has a major problem on their hands in the EDA world...and their actions and new product introductions aren't convincing me otherwise.
So what if AIX has advanced features that Linux lacks? AIX market share is nothing compared to Solaris and HPUX, and most CAD/EDA Unix shops running Solaris/HPUX have found that Linux on Intel handily outperforms Solaris/Sparc & HPUX/PA-RISC on pretty much anything that doesn't need 64-bits or more that 3GB RAM. So if Linux already does well on the low end and will almost certainly do well in the high end, why shouldn't IBM get in early and position themselves as a leading Linux vendor? It's not like AIX is going to suddenly turn around and outpace Linux usage.
I applaud IBM for realizing AIX had it's chance and won't be a dominant player inteh OS field. They'll roll that AIX expertise and technology into Linux and the whole Unix world, IBM included, will be better off for it.
You ought to check it out. Cory is an interesting guy. I saw him talk about the Broadcast flag and other DRM schemes during an EFF-Austin kick off meeting and it was well worth my time. He's technically adept and knows his subject well, plus he's funny and makes clever jokes all the time. I wish the meeting had be longer.
Read any of Pat Cadigan's stuff? Her cyberpunk classic "Synners" is out again in trade paperback, and you can always find her more recent stuff like "Dervish is Digital", etc. in book stores as well. Rudy Rucker is another writer I like, and his last SF book "Spaceland" was a fun read.
Personally, I'm looking forward to Cory Doctorow's "Down & Out in the Magic Kingdom", but I don't think that'll be out until Jan or Feb. I'm going to read "Prey" and a couple others in the meantime, reviews be dammed.
But more important in the comparison than Stevenson's having more than one plot-skeleton in him is his cleverness with language. Unlike Chrichton, Stevenson is an inventive user of language who trusts his audience instead of reaching for the Big Book of simple language.
I think you're partly correct on that. NS's work has more interesting language, but often a bit too much. IMO, NS is in bad need of an editor. A good 20-25% of "Cryptonomicon" could've been cut without any harm to the plots or the technology HOWTO asides. It'd also do NS some good to work on better endings. If he would spend half the time he spends setting up and describing his detailed plots to his finales, his books wouldn't fizzle out so much in the end. The 10 page wrap up of "The Diamond Age" is a prime example of that.
In the long run, an "easy to use by novices" version of Linux will have been produced. There are many developing countries which will embrace such a product and they will likely be running it on x86 chips. This will benefit Intel. Remember the testimony from the MS anti-trust trial? Steve McGeady shed some light on a few very touchy things that MS did to Intel, namely forcing them shutdown the Intel Architecture Labs and going ballistic over Intel's support of Java. It is clearly in Intel's best interest to have equal footing in the Wintel monopoly and right now they do not. Of course Intel would like to remain the preferred arch of MS, but Linux is clearly a better deal for them. Linux helps Intel replace big iron Unix systems and provide lower cost desktop computers in developing countries, all without having to worry about "Linux" getting angry about it.
I mostly agree, except I can't help but think Intel would rather Linux win out in the long run. After all, MS has applied lots of pressure in the past to suppress technology Intel was developing that could've hurt MS. Getting out from under that would be in Intel's best interest.
Not sure which fab you worked in, but the pure water coming into the spin-rinse-dryers and wet benches isn't full of HF, only the used water leaving the process step. The incoming water is the one that folks might think it would be good to drink and what has to be pure.
You are. I worked in a chip fab for 3 years and is common knowledge that drinking the water isn't a good idea because of its purity. The water in question is *far* more pure than any fresh water supply. The less ions in the water, the less damage to the chips.
A sip here and there probably wouldn't kill you, but drinking DI (deionized) water is certainly not a subsitute for degowning and drinking from the water fountains outside the clean rooms!:)
The replacement they were referring to in that article is super critical carbon dioxide. It is a viable solution to the environmental problem for chip production and already used for "greener" dry cleaning, but definately won't be ramping up in fabs anytime soon. Chip manufacturers are very slow and recluctant to change processes.
The water is used for rinsing wafers, which happens many, many times in a typical chip process. The water is highly filtered and deionized before the wafers are washed, then is cleansed to remove the acids and solvents that are picked up during rinse cycles. So it is reusable, but only after minerals are added back to it. You cannot drink fab quality water because it a large concentration gradient would form and minerals from the other fluids in your body would be depleted by the migration into the ultra pure water.
Slashdot Mirror
Yup, same here. We have over 300 servers and a couple of us in our dept. insisted that we pay Red Hat instead of going CentOS only, which we easily could've done. Never use the support but also know that Linux would be in a different position today if it weren't for Red Hat's work.
GFS is so dead and beyond the grave that only zombies use it.
Is that a typo for CFS, or are you really saying that about GFS, global file system?
They are?! That's not the signals I'm seeing from Red Hat. The Virtualization capability of RHEL5 is plastered all over their website and our sales/engineering reps have yet to say hold back on rolling it out.
Eh, the gist of the article was the work conditions for IT jobs were not appealing to many women and that women are leaving the field as a result, even though some companies and organizations WANT MORE women in those jobs and are trying to recruit more of them. It's not just women having kids, its the idea that you're never really away from the immediate, pressing needs of your employer and their demands, and that some women have said enough. So the female side of the IT work force might not grow, it could shrink even more. Hell, even some of us men think IT work conditions suck.
I thought they were going to give their software away and make money on support.
It was always there.
Wonderful, isn't it?
Close, but you are wrong on one point. The fact that Sun systems *can* hold tons of RAM is the only reason that Solaris/SPARC is still in the EDA game. Once Opteron/RHEL3 based systems that address 32+ GB of RAM are shipping in volume, I can't see a single chip design house opting for a Sun proprietary system except for legacy support.
I thought the Red Hat suit took care of that...
Like they say, no matter when you moved to Austin, it was better before you got here.
Which costs employers extra money in the real world, and what the professors are trying to address. I like their idea of balancing the theory more than it currently is. I wish my professors had done something similiar when I was in school. I know I would've benefitted from it.
Dell offered the best hardware and support for the price and also they do price matching so he got quite a few things cheaper than expected.
Maybe the best price/performance hardware support, but their Linux software support leaves a lot to be desired (even at the Gold contract level). We bought several Dell PE2650 servers running Linux and I'm finding that the Dell support techs just don't have enough real world experience with Linux to make Dell into a big Unix player (yet). I'm told Dell is working on correcting that as I type this, but until they do Dell won't be as much of an option for those of us who run Unix shops and know what quality support comes from Sun. Anyone can read a manual--including me--when I call Dell (or Sun or HP), I want to talk to someone who knows more than that.
Surely Sun can't exactly sell the hardware for any cheaper than it can already be bought for, so what's the advantage of choosing them over a company like Dell?
Why not? Think Dell does anything Sun can't do in designing an x86 system? I don't. Sun engineers design the server, then Sun contracts with some of the same manufacturers other x86 vendors use to have them built. It's not as difficult as it may seem. One of the great assests of building x86 systems is the off the shelf nature of the components. That reduces the learning curve considerably when compared to designing everything yourself for a system that isn't as widely used, i.e. sparc.
I welcome Sun's effort to ship better Linux servers. When you consider how much Sun knows about Unix, it's great to have that expertise spilling into the Linux world.
Guess you weren't around when she joined the phoebe (RHL9 beta) mailing list...she wasn't a subscriber for very long, but man did she start up one hell of a flame fest in her short time there.
This is wrong. The Red Hat CDs have trademarked logos in them, not proprietary software. Those trademarks do not prevent you (as a Linux enthusiast and community member, not as a for profit business) from freely distributing the ISOs and software in exactly the same form you downloaded them. Red Hat provides an explanation of this policy on their website.
Where I work, I'm seeing Athlon based servers outperform UltraSPARC III servers (v480s to be exact) by a factor of 2+. Physical compile time for new chip designs has been cut by more than half since the Athlon boxes were brought online, and Athlon stability hasn't been an issue at all.
Sun has a major problem on their hands in the EDA world...and their actions and new product introductions aren't convincing me otherwise.
You are?!
So what if AIX has advanced features that Linux lacks? AIX market share is nothing compared to Solaris and HPUX, and most CAD/EDA Unix shops running Solaris/HPUX have found that Linux on Intel handily outperforms Solaris/Sparc & HPUX/PA-RISC on pretty much anything that doesn't need 64-bits or more that 3GB RAM. So if Linux already does well on the low end and will almost certainly do well in the high end, why shouldn't IBM get in early and position themselves as a leading Linux vendor? It's not like AIX is going to suddenly turn around and outpace Linux usage.
I applaud IBM for realizing AIX had it's chance and won't be a dominant player inteh OS field. They'll roll that AIX expertise and technology into Linux and the whole Unix world, IBM included, will be better off for it.
You ought to check it out. Cory is an interesting guy. I saw him talk about the Broadcast flag and other DRM schemes during an EFF-Austin kick off meeting and it was well worth my time. He's technically adept and knows his subject well, plus he's funny and makes clever jokes all the time. I wish the meeting had be longer.
Read any of Pat Cadigan's stuff? Her cyberpunk classic "Synners" is out again in trade paperback, and you can always find her more recent stuff like "Dervish is Digital", etc. in book stores as well. Rudy Rucker is another writer I like, and his last SF book "Spaceland" was a fun read.
Personally, I'm looking forward to Cory Doctorow's "Down & Out in the Magic Kingdom", but I don't think that'll be out until Jan or Feb. I'm going to read "Prey" and a couple others in the meantime, reviews be dammed.
But more important in the comparison than Stevenson's having more than one plot-skeleton in him is his cleverness with language. Unlike Chrichton, Stevenson is an inventive user of language who trusts his audience instead of reaching for the Big Book of simple language.
I think you're partly correct on that. NS's work has more interesting language, but often a bit too much. IMO, NS is in bad need of an editor. A good 20-25% of "Cryptonomicon" could've been cut without any harm to the plots or the technology HOWTO asides. It'd also do NS some good to work on better endings. If he would spend half the time he spends setting up and describing his detailed plots to his finales, his books wouldn't fizzle out so much in the end. The 10 page wrap up of "The Diamond Age" is a prime example of that.
In the long run, an "easy to use by novices" version of Linux will have been produced. There are many developing countries which will embrace such a product and they will likely be running it on x86 chips. This will benefit Intel. Remember the testimony from the MS anti-trust trial? Steve McGeady shed some light on a few very touchy things that MS did to Intel, namely forcing them shutdown the Intel Architecture Labs and going ballistic over Intel's support of Java. It is clearly in Intel's best interest to have equal footing in the Wintel monopoly and right now they do not. Of course Intel would like to remain the preferred arch of MS, but Linux is clearly a better deal for them. Linux helps Intel replace big iron Unix systems and provide lower cost desktop computers in developing countries, all without having to worry about "Linux" getting angry about it.
That's my opinion. YMMV.
I mostly agree, except I can't help but think Intel would rather Linux win out in the long run. After all, MS has applied lots of pressure in the past to suppress technology Intel was developing that could've hurt MS. Getting out from under that would be in Intel's best interest.
Not sure which fab you worked in, but the pure water coming into the spin-rinse-dryers and wet benches isn't full of HF, only the used water leaving the process step. The incoming water is the one that folks might think it would be good to drink and what has to be pure.
You are. I worked in a chip fab for 3 years and is common knowledge that drinking the water isn't a good idea because of its purity. The water in question is *far* more pure than any fresh water supply. The less ions in the water, the less damage to the chips.
:)
A sip here and there probably wouldn't kill you, but drinking DI (deionized) water is certainly not a subsitute for degowning and drinking from the water fountains outside the clean rooms!
The replacement they were referring to in that article is super critical carbon dioxide. It is a viable solution to the environmental problem for chip production and already used for "greener" dry cleaning, but definately won't be ramping up in fabs anytime soon. Chip manufacturers are very slow and recluctant to change processes.
The water is used for rinsing wafers, which happens many, many times in a typical chip process. The water is highly filtered and deionized before the wafers are washed, then is cleansed to remove the acids and solvents that are picked up during rinse cycles. So it is reusable, but only after minerals are added back to it. You cannot drink fab quality water because it a large concentration gradient would form and minerals from the other fluids in your body would be depleted by the migration into the ultra pure water.