Jesus was born in a stable. That doesn't make him a horse.
Wikipedia: "British people, or Britons, are the citizens of the United Kingdom, British Overseas Territories, and Crown dependencies, and their descendants."
If you want to redefine that to something else then you'll need the worldwide community to agree with you. Maybe you can force them to agree with you... You know, bombs etc. sometimes do the trick?
This isn't a bad idea really. OS X crashes on me all the time, whereas my Windows 7 box is rock solid. Having OS X hardware plus Windows might be the best of both worlds!
Countries don't get to reduce their deficit from 7.x to 3.x % of GDP overnight without someone noticing there's something odd going on. For instance, a Dutch member of congress asked questions about this in the Dutch congress when the Euro membership of Greece was being debated but was completely ignored. Fritz Bolkenstein, at the time EU commissioner for the Netherlands, stated publicly that the EU commission knew all too well that the Greek numbers were doctored, but that a political decision was made to let them in regardless.
The question is, did the Greek government cook the books by themselves or not? And given that the politicians making the decision knew that they were doctored, how come all this is suddenly all only Greece's fault?
Supposedly Greece was a democracy, where did the oligarchs come from?
It's not been a democracy for very long, it was a right-wing military dictatorship until the 1970s (think Pinochet). Democracy is still very much being built day by day in Greece. The fact that they were able to resolve this through a referendum and not through civil war is a win.
The point is: Why would you filter at all? Children don't have a school-issued filter on their mouths, ears, pens, etc. This is the perfect example of what's wrong with the United States at this time. State-enforced external morals with no rationale is loathsome. I believe your founders thought the same, which is why they had this thing called the First Amendment.
The problem is highlighted in the article. IT should report to the business. Not to some CIO. When we just worked for the business we were fine as an IT shop. It all went sour when we moved to "big IT" because "shadow IT was wrong".
Wrong. This isn't intended for your mother. This is about ensuring that the entire chain of ISPs, web hosters, etc between you and these well-known sites is IPv6 ready.
The fact is that the "real" internet backbone (the people who provide the connectivity to the people who provide your ISP's connectivity) has been IPv6 for a while now. They are ready. Many others aren't. This will allow everyone to test things. Last IPv6 day a large number of issues were successfully identified and corrected.
Of course, there's the people engineering part of the equation as well. My company has been really lagging in internal IPv6 capabilities. This IPv6 day has got management riled up to the point that they're all suddenly screaming for it. Technology companies hate to be seen as laggards.
Can someone explain what is the point of DNSsec? An https website already has its own certificate
DNS is a naming service, but it was never designed to be a trustworthy naming service. If it was, then DNS spoofing would have been impossible. Another reason why, currently, SSL certificates are needed is IP address spoofing. But if your certificate is embedded in a DNS entry then there is no reason for anyone to need a third-party-signed certificate at all. All you really need is a single source of trust. Right now we have 2: the root nameservers and the root SSL certificate authorities.
So if we fix DNS then we can skip SSL root CAs entirely and just go with DNS. But SSL certs are a lucrative business, which is why Verisign et. al. don't want DNS to be fixed. It would be the end of their best cash cow. But fixing it is necessary for the internet to become a truly trustworthy place of business.
The article, BTW, strikes me as odd. Isn't it Paul Vixie who has been campaigning for DNSSEC for ages now? He isn't even mentioned.
I would be willing to wager that most everyone commenting on this thread would consider that fair use.
Note that fair use doesn't exist in Europe. If it does exist in the US, and it doesn't want to be bound to abandon fair use, then such provisions should have been negotiated in the treaties that the US signed. How exactly do you justify signing a legally binding contract and consequently weaseling out of it? If fair use is so important (and I agree that it is), then why wasn't it introduced in the WTO treaties? Simple: US companies didn't want fair use doctrine spreading around the world. It decreases the control that IP companies have over their "property".
Weren't the trade sanctions against Cuba put there and don't they remain there in part because of Cuban human rights abuses?
No. They exist to placate Cuban immigrants in Florida. Where is the legislation to stop Chinese companies from exercising control over assets formerly held by Tibetans before the takeover by China? Of course this is a silly example, but the very fact that Cuba is named specifically makes this legislation incompatible with international law.
Legal gambling outfits in the US follow strict gambling laws that regulate...
Actually, rules in e.g. some European countries are far stricter. The measures the US took were there only to protect its own gambling industry. If the issue was really the adherence to certain regulations, then non-US companies could simply have followed the same rules when dealing with US based customers, and there would have been no protectionism and therefor no problem. Instead, the gov'ment simply outlawed credit card companies from honoring gambling debts outside of the US.
They are just about the only people who cannot be responsible for this.
You're kidding, right? They literally wrote the standard. If they didn't want the traffic they should have specified the matter in their RFCs.
Webmasters, indeed - what this committee needs is a couple of sysadmins. You can tell them apart by their attention to the consequences of their actions.
In my company Eudora was the email client of choice for a very long time, at least for Windows clients. The server side was Unix. Now Windows has pushed its way gradually to the server market at our company, causing a monoculture of Windows clients. The reason being that a lot of other server side apps work exclusively with Exchange and that the windows apps all integrate well with each other.
We tried Macs, but Apple Mail just doesn't work very well and it doesn't adhere to some of the standards (like the IMAP namespace extension).
Evolution is a bitch to compile on Solaris and its interface sucks. Of course Solaris now comes with evolution installed, but Solaris is always slow on the uptake of new Gnome versions so it's always out of date.
In short, we need an Outlook killer. Thunderbird isn't it. I'm hoping Eudora is. And if it works well on all platforms we can then tackle the windows monoculture.
The only way to take away market share from Exchange on the server space is to make sure the user space is covered. Right now, the combination of Outlook and Exchange just works, the calendaring is really good, etc. Evolution just doesn't work well. Nothing can talk to Exchange like Outlook, and Exchange is what you have to deal with. I'm talking big corporations here, they're all programming for Exchange nowadays. Don't get me wrong, I'd love to see something better, it just isn't there right now - especially when you calculate in AD, voicemail, IP phones, etc.
So bringing Eudora back and making it work on any platform is the way to go. Then you can tackle the server. I hope to god this works. Just like Firefox/Mozilla taking away market space from IE, this is what will make a real break thru for the regular user and will make Free Software a reality in other stuff than browsers.
Well, maybe "no problem" was understating the facts somewhat:). But the kernel is nice and contained: not many changes needed outside of glibc. The question then becomes what kind of hardware support OpenSolaris has right now and in the near or distant future...So yes, it is a huge practical problem and I'd prefer not to face it at all - but we have life boats ready.
Something I was wondering about while reading your post: suppose Linus changes his mind and wants to go GPLv3 after all? In theory, won't he have to do the same thing? Since people have been contributing under "GPLv2 only" he won't be able to change the license to v3 either without going through the same procedure and asking all of them - right? It'd be interesting to see *that* unfold. Somehow I don't think he'd care much about the legal ramifications. The FSF does, which is why they put the phrase in there in the first place.
Linus keeps talking about being just a technical guy and not caring too much about issues of morality. I wish he'd stop saying that, there's enough amorality around already.
I don't think it would be right to use GPL software for voting machines anyway. Government stuff should always be public domain - no license, no nothing. It is public property. Anyone (companies, educational organizations, private persons) should have complete freedom there. This is why universities that have government grants to produce a specific piece of software can't put any license restrictions on the resulting code - at least it's like that in my country.
Note that if you lock down the voting equipment and still publish the specs of what you locked down and how, it could still qualify as GPLv3 compliant - so you *could* use GPLv3 for voting equipment. The reason that doesn't work for Tivo is that they don't publish the hardware specs. Keeping the specs on the voting equipment secret is just security through obscurity anyway, and is bound to fail (and in the case of the US voting systems already _has_ failed).
Every single aspect of the voting process should be open and free so that anyone with the right knowledge can help by detecting and fixing bugs and so that everyone can be witness to the fact (or lack thereof) that the voting procedures are correct.
Let me reverse the order of two sentences (for no particular reason):
GPL version 2 had no restrictions on what hardware was required upon which to run the software.
Quite right, and it presented a loophole: rather than having proprietary software block you from doing what you wanted, the proprietary stuff has moved into the hardware where you don't have any control over it anymore. What use is software freedom if the hardware is locked down? This locking down is against the very nature of the GPL -- so it has to go the way of the dodo. The biggest problem with moving to a completely "free" system is the lack of open hardware drivers (winmodems, video cards,...). Rather than having people re-engineer the whole shebang we need truly open hardware designs - which is something we're already seeing: open phones, open CPUs, etc. The going is slow but we're getting there.
Fundamentally, I think the poster's quote is one of the biggest arguments against the adoption of GPLv3.
That is possible, yes. But the fact is that most GNU software can be used under *any future* GPL version, and this is how to do it:
I hereby declare to be using GPLv3 on any software that is GPLv2 and contains the "any future GPL version" clause. That wasn't too difficult, was it?:)
The only major software I know of that could be a problem is Linux itself, which is only under GPLv2. No problem, we can move to OpenSolaris just as we moved from XFree to Xorg. My estimation is that just as with GPLv2, GPLv3 will creep up on everyone and be a fait accompli before long.
You also pay much more in taxes than Americans do, and get lower quality health care.
Wrong. We do pay slightly more taxes, but the difference is small: international competition makes it impossible to have too much disparity in the cost of labour. Only an American could believe that it's a bad thing to have universal health care. That's why they're way down on the UN's scorecard of countries' quality of living.
We also get better health care, for a lower price. Insuring everyone is cheaper than insuring just part of the population because you can spread the risk more easily. Also, state covered insurance lowers the overhead cost by forcing the insurance companies down on price.
Medicine, unfortunately, is one of those fields where you get what you pay for.
Actually, the cost of medicine in the US is determined by the cost of the doctor's insurance policy. You do get what you pay for: the ability to sue your doctor for malpractice and get rich.
Everyone else just wants good health, and that doesn't cost all that much. Maybe you should emigrate to Europe.
The only was to get round this is not to buy crappy products. Wait till iTunes 6 has a DRM crack available until you buy music over the web (or use iTunes 5), or just use regular CDs with your iTunes/iPod or whatever.
Or even better: just download your music from the internet illegally, it may be crappy quality but at least it will play.
Good luck, and write to your congressman/MP/dictator.
1) As for their tech support, it's only available 24x7 if you're ready to shovel some serious money in their direction. In fairness, though, their website is pretty helpful.
No, you get 24x7 software support automatically on any contract, including smartnet. What you won't get is 2 or 4 hours parts replacement because they need to stock up on parts for those kinds of contracts in order to make delivery promises like that (you know - warehouses containing parts cost $$).
2) NAT hides many particulars about my networks that I'd just as soon not reveal to the world at large. It also offers an oppurtunity to insert a firewall or other bastion host where it might not otherwise be wanted by my not-so-networking-aware bosses.
You're using NAT as a cheap firewall. So do I. That doesn't mean it hasn't slowed down application development. These are separate points. You can open up your firewall, just like you can open your NAT gateway. With a NAT you still don't have your own *real* address on the network. You have to make those cfg changes on the separate NAT device rather than change the host's firewall with a simple button click. You may know how to change a NAT config but most people don't bother...
3) Point is, to use that memory, you have to spend money somehow, often a lot more than just the $43. The act of spending money itself, not to mention figuring out all the possible gotchas, is, in itself, expensive.
There are more things in life that change than IP technologies. Next time you move that 2500 to a different office I'll bet its power supply fails with Cisco refusing to supply a replacement (they've been EOL for how long now?). So you'll automatically buy a 2600 (or whatever it'll be) which will have IPv6. Don't you ever need a new car? Once you do it'll have the new nifty feature in there by default.
4) Personally, I don't think I'm the obstacle to a more wide-open Internet. You can have open access if you want, but I'll stick with NAT and related technologies until I see something that offers a gradual, painless transition process and doesn't cost an arm and a leg.
I don't see a reason why you could be forced into that decision until every dick tom and harry is using IPv6 anyway. Until then ISPs will simply have 2 groups of users. More $$ for network admins, sysadmins and consultants pays my bills:).
It looks like the environment is pretty well-defined, a particular set of software that needs to be present on all lab machines, but all lab machines can be identical. If so, then you can centrally manage your solution.
If you have control over what kind of computers (new ones) go in you can run them as thin clients. Control of the hardware platform makes it very easy to build a network-boot kind of system - using PXE boot and diskless clients.
But it looks like you have a plethora of computers, donated by a large group of donors. In other words: different kinds of computers with different hardware installed, etc. In this case especially you will be able to benefit from what others have done before you. Look for instance at some of the Linux distros like knoppix. These simply work out-of-the-box on a whole slew of computers and are well tested by thousands of volunteers.
You will need to decide many things:
Does a centrally managed lab work for you? (i.e. not too many one-offs)
Do you need local disk? Can you use network disk instead?
do you need NFS (auto)mounted home directories?
Can your machines boot from the network using BIOS settings?
Else, can they boot from the network after booting from some kind of local storage (USB key, local disk)
Slashdot Mirror
Jesus was born in a stable. That doesn't make him a horse.
Wikipedia: "British people, or Britons, are the citizens of the United Kingdom, British Overseas Territories, and Crown dependencies, and their descendants."
If you want to redefine that to something else then you'll need the worldwide community to agree with you. Maybe you can force them to agree with you... You know, bombs etc. sometimes do the trick?
I suspect the 1% nerds are few in number, probably because Slashdot is a relic of the dot com bust and only readers from that era know about it.
I, for one, am glad that most of the script kiddies are gone.
You forgot to start with Scheme.
FTFY.
https://en.m.wikipedia.org/wiki/Texas_Instruments_TI-99/4A
This isn't a bad idea really. OS X crashes on me all the time, whereas my Windows 7 box is rock solid. Having OS X hardware plus Windows might be the best of both worlds!
Countries don't get to reduce their deficit from 7.x to 3.x % of GDP overnight without someone noticing there's something odd going on. For instance, a Dutch member of congress asked questions about this in the Dutch congress when the Euro membership of Greece was being debated but was completely ignored. Fritz Bolkenstein, at the time EU commissioner for the Netherlands, stated publicly that the EU commission knew all too well that the Greek numbers were doctored, but that a political decision was made to let them in regardless.
The question is, did the Greek government cook the books by themselves or not? And given that the politicians making the decision knew that they were doctored, how come all this is suddenly all only Greece's fault?
Supposedly Greece was a democracy, where did the oligarchs come from?
It's not been a democracy for very long, it was a right-wing military dictatorship until the 1970s (think Pinochet). Democracy is still very much being built day by day in Greece. The fact that they were able to resolve this through a referendum and not through civil war is a win.
Nice Dune reference :)
Hey, I resent that!
The point is: Why would you filter at all? Children don't have a school-issued filter on their mouths, ears, pens, etc. This is the perfect example of what's wrong with the United States at this time. State-enforced external morals with no rationale is loathsome. I believe your founders thought the same, which is why they had this thing called the First Amendment.
The problem is highlighted in the article. IT should report to the business. Not to some CIO. When we just worked for the business we were fine as an IT shop. It all went sour when we moved to "big IT" because "shadow IT was wrong".
Wrong. This isn't intended for your mother. This is about ensuring that the entire chain of ISPs, web hosters, etc between you and these well-known sites is IPv6 ready.
The fact is that the "real" internet backbone (the people who provide the connectivity to the people who provide your ISP's connectivity) has been IPv6 for a while now. They are ready. Many others aren't. This will allow everyone to test things. Last IPv6 day a large number of issues were successfully identified and corrected.
Of course, there's the people engineering part of the equation as well. My company has been really lagging in internal IPv6 capabilities. This IPv6 day has got management riled up to the point that they're all suddenly screaming for it. Technology companies hate to be seen as laggards.
Can someone explain what is the point of DNSsec? An https website already has its own certificate
DNS is a naming service, but it was never designed to be a trustworthy naming service. If it was, then DNS spoofing would have been impossible. Another reason why, currently, SSL certificates are needed is IP address spoofing. But if your certificate is embedded in a DNS entry then there is no reason for anyone to need a third-party-signed certificate at all. All you really need is a single source of trust. Right now we have 2: the root nameservers and the root SSL certificate authorities.
So if we fix DNS then we can skip SSL root CAs entirely and just go with DNS. But SSL certs are a lucrative business, which is why Verisign et. al. don't want DNS to be fixed. It would be the end of their best cash cow. But fixing it is necessary for the internet to become a truly trustworthy place of business.
The article, BTW, strikes me as odd. Isn't it Paul Vixie who has been campaigning for DNSSEC for ages now? He isn't even mentioned.
I would be willing to wager that most everyone commenting on this thread would consider that fair use.
Note that fair use doesn't exist in Europe. If it does exist in the US, and it doesn't want to be bound to abandon fair use, then such provisions should have been negotiated in the treaties that the US signed. How exactly do you justify signing a legally binding contract and consequently weaseling out of it? If fair use is so important (and I agree that it is), then why wasn't it introduced in the WTO treaties? Simple: US companies didn't want fair use doctrine spreading around the world. It decreases the control that IP companies have over their "property".
Weren't the trade sanctions against Cuba put there and don't they remain there in part because of Cuban human rights abuses?
No. They exist to placate Cuban immigrants in Florida. Where is the legislation to stop Chinese companies from exercising control over assets formerly held by Tibetans before the takeover by China? Of course this is a silly example, but the very fact that Cuba is named specifically makes this legislation incompatible with international law.
Legal gambling outfits in the US follow strict gambling laws that regulate...
Actually, rules in e.g. some European countries are far stricter. The measures the US took were there only to protect its own gambling industry. If the issue was really the adherence to certain regulations, then non-US companies could simply have followed the same rules when dealing with US based customers, and there would have been no protectionism and therefor no problem. Instead, the gov'ment simply outlawed credit card companies from honoring gambling debts outside of the US.
They are just about the only people who cannot be responsible for this.
You're kidding, right? They literally wrote the standard. If they didn't want the traffic they should have specified the matter in their RFCs.
Webmasters, indeed - what this committee needs is a couple of sysadmins. You can tell them apart by their attention to the consequences of their actions.
In my company Eudora was the email client of choice for a very long time, at least for Windows clients. The server side was Unix. Now Windows has pushed its way gradually to the server market at our company, causing a monoculture of Windows clients. The reason being that a lot of other server side apps work exclusively with Exchange and that the windows apps all integrate well with each other.
We tried Macs, but Apple Mail just doesn't work very well and it doesn't adhere to some of the standards (like the IMAP namespace extension).
Evolution is a bitch to compile on Solaris and its interface sucks. Of course Solaris now comes with evolution installed, but Solaris is always slow on the uptake of new Gnome versions so it's always out of date.
In short, we need an Outlook killer. Thunderbird isn't it. I'm hoping Eudora is. And if it works well on all platforms we can then tackle the windows monoculture.
The only way to take away market share from Exchange on the server space is to make sure the user space is covered. Right now, the combination of Outlook and Exchange just works, the calendaring is really good, etc. Evolution just doesn't work well. Nothing can talk to Exchange like Outlook, and Exchange is what you have to deal with. I'm talking big corporations here, they're all programming for Exchange nowadays. Don't get me wrong, I'd love to see something better, it just isn't there right now - especially when you calculate in AD, voicemail, IP phones, etc.
So bringing Eudora back and making it work on any platform is the way to go. Then you can tackle the server. I hope to god this works. Just like Firefox/Mozilla taking away market space from IE, this is what will make a real break thru for the regular user and will make Free Software a reality in other stuff than browsers.
That's a fucking HUGE project
Well, maybe "no problem" was understating the facts somewhat :). But the kernel is nice and contained: not many changes needed outside of glibc. The question then becomes what kind of hardware support OpenSolaris has right now and in the near or distant future...So yes, it is a huge practical problem and I'd prefer not to face it at all - but we have life boats ready.
Something I was wondering about while reading your post: suppose Linus changes his mind and wants to go GPLv3 after all? In theory, won't he have to do the same thing? Since people have been contributing under "GPLv2 only" he won't be able to change the license to v3 either without going through the same procedure and asking all of them - right? It'd be interesting to see *that* unfold. Somehow I don't think he'd care much about the legal ramifications. The FSF does, which is why they put the phrase in there in the first place.
Linus keeps talking about being just a technical guy and not caring too much about issues of morality. I wish he'd stop saying that, there's enough amorality around already.
I don't think it would be right to use GPL software for voting machines anyway. Government stuff should always be public domain - no license, no nothing. It is public property. Anyone (companies, educational organizations, private persons) should have complete freedom there. This is why universities that have government grants to produce a specific piece of software can't put any license restrictions on the resulting code - at least it's like that in my country.
Note that if you lock down the voting equipment and still publish the specs of what you locked down and how, it could still qualify as GPLv3 compliant - so you *could* use GPLv3 for voting equipment. The reason that doesn't work for Tivo is that they don't publish the hardware specs. Keeping the specs on the voting equipment secret is just security through obscurity anyway, and is bound to fail (and in the case of the US voting systems already _has_ failed).
Every single aspect of the voting process should be open and free so that anyone with the right knowledge can help by detecting and fixing bugs and so that everyone can be witness to the fact (or lack thereof) that the voting procedures are correct.
GPL version 2 had no restrictions on what hardware was required upon which to run the software.
Quite right, and it presented a loophole: rather than having proprietary software block you from doing what you wanted, the proprietary stuff has moved into the hardware where you don't have any control over it anymore. What use is software freedom if the hardware is locked down? This locking down is against the very nature of the GPL -- so it has to go the way of the dodo. The biggest problem with moving to a completely "free" system is the lack of open hardware drivers (winmodems, video cards, ...). Rather than having people re-engineer the whole shebang we need truly open hardware designs - which is something we're already seeing: open phones, open CPUs, etc. The going is slow but we're getting there.
Fundamentally, I think the poster's quote is one of the biggest arguments against the adoption of GPLv3.
That is possible, yes. But the fact is that most GNU software can be used under *any future* GPL version, and this is how to do it: I hereby declare to be using GPLv3 on any software that is GPLv2 and contains the "any future GPL version" clause. That wasn't too difficult, was it? :)
The only major software I know of that could be a problem is Linux itself, which is only under GPLv2. No problem, we can move to OpenSolaris just as we moved from XFree to Xorg. My estimation is that just as with GPLv2, GPLv3 will creep up on everyone and be a fait accompli before long.
You also pay much more in taxes than Americans do, and get lower quality health care.
Wrong. We do pay slightly more taxes, but the difference is small: international competition makes it impossible to have too much disparity in the cost of labour. Only an American could believe that it's a bad thing to have universal health care. That's why they're way down on the UN's scorecard of countries' quality of living.
We also get better health care, for a lower price. Insuring everyone is cheaper than insuring just part of the population because you can spread the risk more easily. Also, state covered insurance lowers the overhead cost by forcing the insurance companies down on price.
Medicine, unfortunately, is one of those fields where you get what you pay for.
Actually, the cost of medicine in the US is determined by the cost of the doctor's insurance policy. You do get what you pay for: the ability to sue your doctor for malpractice and get rich.
Everyone else just wants good health, and that doesn't cost all that much. Maybe you should emigrate to Europe.
The only was to get round this is not to buy crappy products. Wait till iTunes 6 has a DRM crack available until you buy music over the web (or use iTunes 5), or just use regular CDs with your iTunes/iPod or whatever.
Or even better: just download your music from the internet illegally, it may be crappy quality but at least it will play.
Good luck, and write to your congressman/MP/dictator.
No, you get 24x7 software support automatically on any contract, including smartnet. What you won't get is 2 or 4 hours parts replacement because they need to stock up on parts for those kinds of contracts in order to make delivery promises like that (you know - warehouses containing parts cost $$).
2) NAT hides many particulars about my networks that I'd just as soon not reveal to the world at large. It also offers an oppurtunity to insert a firewall or other bastion host where it might not otherwise be wanted by my not-so-networking-aware bosses.
You're using NAT as a cheap firewall. So do I. That doesn't mean it hasn't slowed down application development. These are separate points. You can open up your firewall, just like you can open your NAT gateway. With a NAT you still don't have your own *real* address on the network. You have to make those cfg changes on the separate NAT device rather than change the host's firewall with a simple button click. You may know how to change a NAT config but most people don't bother...
3) Point is, to use that memory, you have to spend money somehow, often a lot more than just the $43. The act of spending money itself, not to mention figuring out all the possible gotchas, is, in itself, expensive.
There are more things in life that change than IP technologies. Next time you move that 2500 to a different office I'll bet its power supply fails with Cisco refusing to supply a replacement (they've been EOL for how long now?). So you'll automatically buy a 2600 (or whatever it'll be) which will have IPv6. Don't you ever need a new car? Once you do it'll have the new nifty feature in there by default.
4) Personally, I don't think I'm the obstacle to a more wide-open Internet. You can have open access if you want, but I'll stick with NAT and related technologies until I see something that offers a gradual, painless transition process and doesn't cost an arm and a leg.
I don't see a reason why you could be forced into that decision until every dick tom and harry is using IPv6 anyway. Until then ISPs will simply have 2 groups of users. More $$ for network admins, sysadmins and consultants pays my bills :).
In Silicon Valley, 2200 won't pay your rent.
If you have control over what kind of computers (new ones) go in you can run them as thin clients. Control of the hardware platform makes it very easy to build a network-boot kind of system - using PXE boot and diskless clients.
But it looks like you have a plethora of computers, donated by a large group of donors. In other words: different kinds of computers with different hardware installed, etc. In this case especially you will be able to benefit from what others have done before you. Look for instance at some of the Linux distros like knoppix. These simply work out-of-the-box on a whole slew of computers and are well tested by thousands of volunteers.
You will need to decide many things:
- Does a centrally managed lab work for you? (i.e. not too many one-offs)
- Do you need local disk? Can you use network disk instead?
- do you need NFS (auto)mounted home directories?
- Can your machines boot from the network using BIOS settings?
- Else, can they boot from the network after booting from some kind of local storage (USB key, local disk)
Good luck!