As long as your windows software is compatible with Windows Terminal Server, you can use WinConnect Server XP.
This lets you run terminal services with a lower initial cost, and there is a free download that lets you try it out for free (there is a timeout on the sessions).
This IMHO opens the door for a linux desktop migration than anything else out there right now.
There is already so much hardware out there, why buy new? Get a real PIII instead and recycle. (Make sure you avoid the bad old chipsets used in AMD & Cyrix systems from the era. Try it with memtest86+. Any system with the L2 on the CPU and PC100/PC133 should have a memory bandwith of ~250MB/second. Anything less and you'll regret it.
Upgrade the memory and put in a new PATA hard drive in and you're set for the $150 or less. And a nice side affect if you're at all interested in the environment is this doesn't require more chemical production in manufacturing plants (if you don't count the additional memory & new hard drive of course).
With enough memory (512 - 1GB) and a speedy hard drive you'd be surprised how well the old computers can work.
Also if you're a developer, programming on slower computers (and even slower dual cpu systems like my dual pII 450) gives you an encentive to make faster, lower resource using code.
If it wasn't for Linspire, there wouldn't be any cheap hardware available in large stores like Frys. Also by buying linspire computers you are supporting linux developers. What's wrong with that?
Their latest ones are quiet low power (think Watts) AMD Geode based machines. Who else is fighting against the MS tax in the large stores?
...someone tell the parent poster he just described IA64.
I believe it is the instruction set that is a major hutle, not executing the same instructions on two cores and switching cores based on correct branch outcomes.
Also there are certain workloads that other process can't beat when compared with IA64 IIRC.
Out of curiosity, where do "bitchload" and "assload" fit in? I have deabted with several friends before as to whether or not either or both are larger than a shitload. Maybe its just a unit thing and its like comparing a metric tonne with a ton.
bitchload = fuckload assload = shitload
one stop to the bowl should be larger than one round of fucking so: shitload > fuckload
Shit should be denser than cum so (assuming the shit isn't liquidic): shitton > fuckton
Yeah, 2.4 was the start of the end for kernel.org supplying production ready kernels. My first kernel was 2.0.35. I was a late adopter of 2.2, and refused to try 2.4 until ext3 was ported and even then only started at 2.4.7 + ext3 patch.
Though the 2.6 development cycle does avoid the huge distro forks that were required for 2.4.x (read RH and to a lesser degree SuSe).
It would be nice to see more centralization in kernel maintenance after a release. Currently there is the 2.6.x.y that only tracks the latest release, but what about all of those distros that have 2.6.5 (SLES9), 2.6.8 (Debian3.1) or 2.6.9 (RHEL/CENTOS) based kernels? It has been made very clear that the distributions are responsible for the reliability of their kernel releases. But why shouldn't there be an upstream for the distros to work with?
Release 2.6.8 and all distros that have that version should collaborate and maintain that release. Patches that reach consensus among the various distributions using that version would merge it upstream and then you'd have 2.6.8.x released that they all use as a base.
I'm sure others have thought of this before. Any idea why it hasn't been done? Hopefully not because of personality conflicts...
Also debian's 2.6.15-8 has worked fine on my home desktop/development machine which is dual pII 450 with a 440BX chipset.
I wonder what the problem was. Did you analyze your oops or did it just hang? Were you running X? Did you try turning on nmi_watchdog?
Are you using a 1KB block size in your ext3 partitions? If so you'll be limited to 16GB file sizes. If you use 4KB blocks, the limit is measured in TB.
Are you sure it is file size limit in the filesystem?
As someone who has been a contributor to several projects and maintainer of one (so far vTiger 4.2.x) I have some experience with this.
As a user I would read the posts from people that knew more and learn from them and then respond to others as my skill level allowed. As I learned more some messages just weren't worth the trouble of a response, but others took up the slack.
I think OSS works best when there is an unenforced support hierarchy. The people who know the most don't spend a lot of time with the n00bs, but other users with a bit more experience do interact with them. If it turns out the problem requires more expertise, a developer or more experienced user/admin can step in to help identify the problem and/or provide a fix.
Whenever I step out of this support model, I end up getting frustrated with users who can't or won't do what I ask them to do to help them fix the problem they are having. At that point, it is better to ignore that user than to post an insulting comment. Usually someone else will step in at that point and try help the user and sometimes they succeed.
It is these helpful users that you need to nurture. Compliments and encouragement keeps them around and makes your life easier. Also many times they turn out to be helpful in other ways like sending in patches, etc.
So if you get frustrated with n00bs easily, don't interact with them. Let the other users that will inevitably join an OSS community do that for you and interact with them. They have enough experience to help others, but may need help from time to time.
Slashdot Mirror
Can Firefox get you laid? No.
Which can also be done with any sufficiently object oriented language, even Perl.
And that didn't require ActiveX.
Every brother in the world will hate you.
As long as your windows software is compatible with Windows Terminal Server, you can use WinConnect Server XP.
This lets you run terminal services with a lower initial cost, and there is a free download that lets you try it out for free (there is a timeout on the sessions).
This IMHO opens the door for a linux desktop migration than anything else out there right now.
Journaling beyond metadata? Wouldn't that aid file recovery if the writing software screws up a write?
Not if the last write operation was truncate (which sets the file size to zero just before writing the contents of the file over again)
The fact is that most apps write the entire contents of their files on every save, and it is a multi-step process. truncate, write 4k, write 4k, etc.
You'd need transaction support like what reiser4 offers. But your apps would still need to use the API.
(apparently some punks had vandalized it the night before)
And you now know why they vandalized...
There is already so much hardware out there, why buy new? Get a real PIII instead and recycle. (Make sure you avoid the bad old chipsets used in AMD & Cyrix systems from the era. Try it with memtest86+. Any system with the L2 on the CPU and PC100/PC133 should have a memory bandwith of ~250MB/second. Anything less and you'll regret it.
Upgrade the memory and put in a new PATA hard drive in and you're set for the $150 or less. And a nice side affect if you're at all interested in the environment is this doesn't require more chemical production in manufacturing plants (if you don't count the additional memory & new hard drive of course).
With enough memory (512 - 1GB) and a speedy hard drive you'd be surprised how well the old computers can work.
Also if you're a developer, programming on slower computers (and even slower dual cpu systems like my dual pII 450) gives you an encentive to make faster, lower resource using code.
So don't buy from the large stores? You could ALWAYS purchase an OS-less computer at a local non-corporate computer store.
The point is that Linspire is moving people away from Microsoft that otherwise may not know about an alternative.
The attack on MS needs to be done on all fronts, including the retail side that Linspire is targetting.
If it wasn't for Linspire, there wouldn't be any cheap hardware available in large stores like Frys. Also by buying linspire computers you are supporting linux developers. What's wrong with that?
Their latest ones are quiet low power (think Watts) AMD Geode based machines. Who else is fighting against the MS tax in the large stores?
Are those files in a package from Ubuntu?
Plus, there is some good competition out there for dell in the low end computer market. emachines, for example.
Please don't compare emachines to anything claiming quality. emachines will always loose.
...someone tell the parent poster he just described IA64.
I believe it is the instruction set that is a major hutle, not executing the same instructions on two cores and switching cores based on correct branch outcomes.
Also there are certain workloads that other process can't beat when compared with IA64 IIRC.
Who do you mean by they (OOo?), and what is not open? (If they == apple then I understand.)
Fuckton > Shitton > Fuckload > Shitload > Grip
Out of curiosity, where do "bitchload" and "assload" fit in? I have deabted with several friends before as to whether or not either or both are larger than a shitload. Maybe its just a unit thing and its like comparing a metric tonne with a ton.
bitchload = fuckload
assload = shitload
one stop to the bowl should be larger than one round of fucking so: shitload > fuckload
Shit should be denser than cum so (assuming the shit isn't liquidic): shitton > fuckton
Yeah, 2.4 was the start of the end for kernel.org supplying production ready kernels. My first kernel was 2.0.35. I was a late adopter of 2.2, and refused to try 2.4 until ext3 was ported and even then only started at 2.4.7 + ext3 patch.
Though the 2.6 development cycle does avoid the huge distro forks that were required for 2.4.x (read RH and to a lesser degree SuSe).
It would be nice to see more centralization in kernel maintenance after a release. Currently there is the 2.6.x.y that only tracks the latest release, but what about all of those distros that have 2.6.5 (SLES9), 2.6.8 (Debian3.1) or 2.6.9 (RHEL/CENTOS) based kernels? It has been made very clear that the distributions are responsible for the reliability of their kernel releases. But why shouldn't there be an upstream for the distros to work with?
Release 2.6.8 and all distros that have that version should collaborate and maintain that release. Patches that reach consensus among the various distributions using that version would merge it upstream and then you'd have 2.6.8.x released that they all use as a base.
I'm sure others have thought of this before. Any idea why it hasn't been done? Hopefully not because of personality conflicts...
Also debian's 2.6.15-8 has worked fine on my home desktop/development machine which is dual pII 450 with a 440BX chipset.
I wonder what the problem was. Did you analyze your oops or did it just hang? Were you running X? Did you try turning on nmi_watchdog?
Are you using a 1KB block size in your ext3 partitions? If so you'll be limited to 16GB file sizes. If you use 4KB blocks, the limit is measured in TB.
Are you sure it is file size limit in the filesystem?
No, it's:
/dev/hda12
dd
Hint, that will write zeros to the entire contents of the 12th partition on the first PATA hard drive. Don't run it.
As someone who has been a contributor to several projects and maintainer of one (so far vTiger 4.2.x) I have some experience with this.
As a user I would read the posts from people that knew more and learn from them and then respond to others as my skill level allowed. As I learned more some messages just weren't worth the trouble of a response, but others took up the slack.
I think OSS works best when there is an unenforced support hierarchy. The people who know the most don't spend a lot of time with the n00bs, but other users with a bit more experience do interact with them. If it turns out the problem requires more expertise, a developer or more experienced user/admin can step in to help identify the problem and/or provide a fix.
Whenever I step out of this support model, I end up getting frustrated with users who can't or won't do what I ask them to do to help them fix the problem they are having. At that point, it is better to ignore that user than to post an insulting comment. Usually someone else will step in at that point and try help the user and sometimes they succeed.
It is these helpful users that you need to nurture. Compliments and encouragement keeps them around and makes your life easier. Also many times they turn out to be helpful in other ways like sending in patches, etc.
So if you get frustrated with n00bs easily, don't interact with them. Let the other users that will inevitably join an OSS community do that for you and interact with them. They have enough experience to help others, but may need help from time to time.
Believe me. They're not a vixen anymore when their head is in the bowl.
Only if they're *all* shirtless.
I haven't seen the first episode, but I can guess.
Knight Rider!
Am I the only one who burst out laughing so loud you heard yourself echoing off of the walls?
The Mac menu remains infinitely tall, and you just whack the cursor up against it.
I think there are valid arguments that this motion is worse from a RSI perspective.
I've said "Who're you" before and had some women get offended because of the second meaning it has.
This search might be helpful.