"Do you trust someone with a profit motive to screw with your connection, or someone with a political motive? Profit motive. Because I can take my business elsewhere."
The overselling of bandwidth, however, is necessary to make remotely efficient use of the "infrastructure."
No it isn't.
Proper business management is.
If they promise me "X", by contract, they better get me "X", not "X, unless you happen to try to get X, in which case we'll give you x". It is they the ones that promise. What about making promises they can back instead of bogus ones?
"If profit is motivation, then surely one company would realize that "hey, if everyone else is being a dick, I can make a ton of profit by not being a dick and simply getting all the frustrated customers." The legion of doom argument that supposes all corporations sit and plot the downfall of the working class together always seems to amaze me, because it goes against the very fundamentals of capitalism. It's no different than game theory."
In fact it *is* game theory, a variant of the prisioner's dilemma. In this case your "clever company" would indeed get a great benefit, then the other competitors would go by the same price scheme and at the end they *all* would end up losing a ton o money (well, squeezing a lot less money from the people).
Certainly, when you have a lot of competitors you are almost guaranteed than at least one will go for the short time profits. That's how competition works and that's why it's good having flourishing competition for a healthy market. When you have just a short bunch of big competitors (say, three, four) with a high entry barrier it's much easier for them to "play by their rules" and at the same time watch out each other in short range to be sure they comply to "their rules". This way they all are able to go with higher margins.
Which of those market models do you think fits better the telcos? The one with high competition or the oligopoly with high entry barriers, uh?
"There are companies out there that just build data centers."
The fact that they "just build data centers" is no guarantee that they will build them better than the company that build datacenters on top of something else. Once you are big enough in an industry, there's no guaranteed benefit in going out for a provider. In fact you are *guaranteed* in the limit that it will be cheaper to do it yourself (your costs will be just the same than those of the provider but you will save on its benefits).
So the question is not why they build their own datacenter but if they'll be up to the task.
"... and if I was trying to host an illegal file sharing site on one of their clouds they would shut it down for exactly the same reasons as they did WikiLeaks."
Except for the "illegal" part.
"It is stuff that is illegal to have."
Don't think so. And provided that no USA fiscal office has started procedures against Wikileaks for the published contents, so do think USA legal body. It is stuff that it may be illegal to share if you are managing it. Perfectly legal to have and publish if you have managed to get to somehow. That's why, in case of any doubt, those pesky UK terrorists made a point introducing the First Amendment to their Constitution. Cfr. the legal situation of Bernstein & Woodward versus "Deep Throat" regarding the Watergate scandal.
"My gut feeling agrees with the AC that over-use of bandwidth may be the case."
I hear about "bandwidth over-use" once and again, but how the hell could anyone over use a resource provided on-demand? In this case, the network provider have all the ability, even on the cheap,to control their customers' resource usage. If they contracted 1TB/month per server, how could they be able to use one single bit more than this? The provider could just say "see? 1TB; you are off till next month the first" a simple firewall rule after that and they are done.
Or might it be the case that even when the provider in fact offered them 1TB/month to its customer they now say "OMG, they are in fact using what we promised them by contract, we can't tolerate that!"?
"Sorry, but you are a bunch of amateurs - that's all there is to it."
Because they depend on third parties? Surely so you do and so do everybody in business.
Provided (a big "if") their contract clearly stated they were offered 100TB per server, it would be the provider unable to hold to their contract the amateurish one (if not blatantly illegal).
I would look for legal advice since it seems they should be able to claim for contract breaching and damages for the lost business revenue on top of that.
"And I say that lowering the levels of CO2 will cause fairies to explode into flower scented farts. The burden of proof is on you to prove that it won't happen."
OK, I'll do. Lowering the levels of CO2 from current 380ppm to 280ppm will put them as they were by the year 1800 and it's known fairies didn't explode back then.
"I am a bit concerned that what appears to be an entire agency moving its operations toward complete dependency on a single commercial entity."
The point is that they probably *already* are completly dependent on a single commercial entity. Other poster said USDA is basically a Microsoft shop to a point they can't realistically go away from it (wonderful Microsoft lock-in) much less in these recession years. So going to Microsoft's cloud just makes sense: it will be cheaper and it will probably go a bit better managed (or at least, there's the hope that they will be able to sue if not).
The pity is that they went this short-sighted path instead of realizing what's the real culprit (Microsoft lock-in) and understand that the sooner they break the lock-in, the better and cheaper in the long run.
"There's plenty of proprietary source implementations of TCP/IP, you know."
And they work in a compatible way that allowed to its success due to the existance of an open source one to be tested against, you know.
"Which to me shows that what matters is an open API and public domain algorithms, not open source."
You are true, but it's a naive point of view. There's no way for anything more complex than a whistle that you can have an unambigous protocol definition. In order for the API or algorithms to gain wide adoption you need an open sourced implementation (BSD licensed if you want or expect close source distributing companies to adopt it).
"Open Source business model works well when your company model is a consulting based business."
Yes. But let's make "consulting" a wider word and it will become even truer. If your business is about contracting (I'll pay you to *do* this) instead of licensing (I'll pay you to be *allowed* to do this) then it makes quite a lot of sense.
It's not only consulting but developing on-demand too (and everything where "doing" is involved instead of "licensing").
"Suns model was product based business."
The problem is not what Sun's model *was* but that it wasn't able to adapt.
"IBM doesn't do many products any more"
So they *had* a business model and they *adapted* it to a changing environment.
"I've not worked as a programmer for, hmm, maybe 15 years and all of this was known way back even before I "retired" from that line of work."
Yes: there's an obvious problem with programming and it's that "we" as a guild are not building upon past experience. For the most part, the current generation of programmers are making the same kind of mistakes that where common -and learnt how to avoid, even 20 years ago. Can you imagine, say, aviation if you had to engineer an Airbus 380 all the way from Otto Lilienthal to-date instead of building on past knowledge? That's what I feel too many times when dealing with new development projects.
Now the question is: what could we do so new programmers start where the previous generation left instead of having to build their own knowledge almost all the way along again?
No, it isn't. It goes "don't do that... but don't fall in the other extreme".
That's on line with his central idea that programming is "an art, one that requires a skilled hand and a creative mind to achieve a happy medium between problematic extremes".
"My employer uses F/OSS extensively - and as the sysadmin, I've started to notice a pattern."
I'd think you just noticed half the pattern.
"F/OSS products which scratch an individual or a small group of peoples' itch generally get developed to a certain point and then stagnate. If you're lucky, that point is acceptable to you."
If you are not so lucky, it's open source, go scratch your own itch and become part of the "group of people" that develop the product under your company's bankroll to its "certain point" and then let it stagnate. This way it'll grow tailor-made to your company's needs while still probably being cheaper and less risky than going the closed source path.
"The products that do really well - the "best of F/OSS", if you like - are almost invariably the sort which scratches a very common itch."
So there's a lot of people that do what I pointed out on my previous paragraph so you are less likely to be the one to step ahead into the "group of people" that makes it happen.
"They're usually bankrolled by a number of companies (the Linux kernel falls under this category) or become self-funding when the project leader sets up a company to sell a commercial version with support and possibly extra features."
For some of them it might be the case that your company is the one most interested on the project reaching a high level "scratching point", so it becomes a revenue opportunity for it.
"What I like about rolling releases is you get to deal with application incompatabilities one at a time as they come up"
Only if you can dedicate time each and every day to support efforts and you can afford not knowing when your system will be down for maintenance. Most people neither can nor want to afford that.
"rather than having to spend a week or few all at once when upgrading a distro."
False disyuntive. My experience with Debian (Woody, Sarge, Etch, Lenny) is that it takes me about four hours to calmly upgrade from one to the next, so it's either taking care of problems each and every day, not knowing when something will break, or about four hours about each two years.
"I think it's also probably better for security, as you get the latest patches for the software."
Almost no developer follows sounded software promotion practices, so you end up with security patches *and* new bugs at the same time. Not to talk about integration-related problems.
"If our allies suddenly can't trust our foreign service to keep classified information, well, classified, how likely do you think it is they will continue to trust our diplomats?"
Well, the obvious point here is that your foreign service is in fact unable to keep classified information, well, classified so why the rest of the world should trust your diplomats?
And, please, pay attention that your diplomats are not unable to keep secret things secret because of Assange but because of themselves, so why fingerpointing Assange instead of the real culprits, the diplomats, that's it?
"You can only take pride in being American if it took you some work to become one."
*Or* behave as one.
I can accept someone USA-born taking pride of his nationality if he really lives for what he percieves to be the "true nature" of being American (you know, the land of the free, the home of the brave and all that stuff) and it's up to the challenge of making sure those ideals reach intact or even embettered for those that will come after him.
"the vast majority of the material that's been released shows no evidence of any government abuses"
There's more on the "abuse" word than "plain illegal", specially with respect to politics. If there's really no government abuses, why all the airing? Can it be because at least some people found the data supportive of at least questionable practices? And if it's indeed questionable practices at least for somebody, how can't it be considered whistleblowering?
"Is shipping the data to a country where it is known that you run an increased risk of agencies and interested 3rd parties taking a peek at the data 'due diligence'?"
Slashdot Mirror
"Do you trust someone with a profit motive to screw with your connection, or someone with a political motive?
Profit motive. Because I can take my business elsewhere."
Unless you can't. Which happens to be the case.
The overselling of bandwidth, however, is necessary to make remotely efficient use of the "infrastructure."
No it isn't.
Proper business management is.
If they promise me "X", by contract, they better get me "X", not "X, unless you happen to try to get X, in which case we'll give you x". It is they the ones that promise. What about making promises they can back instead of bogus ones?
"If profit is motivation, then surely one company would realize that "hey, if everyone else is being a dick, I can make a ton of profit by not being a dick and simply getting all the frustrated customers." The legion of doom argument that supposes all corporations sit and plot the downfall of the working class together always seems to amaze me, because it goes against the very fundamentals of capitalism. It's no different than game theory."
In fact it *is* game theory, a variant of the prisioner's dilemma. In this case your "clever company" would indeed get a great benefit, then the other competitors would go by the same price scheme and at the end they *all* would end up losing a ton o money (well, squeezing a lot less money from the people).
Certainly, when you have a lot of competitors you are almost guaranteed than at least one will go for the short time profits. That's how competition works and that's why it's good having flourishing competition for a healthy market. When you have just a short bunch of big competitors (say, three, four) with a high entry barrier it's much easier for them to "play by their rules" and at the same time watch out each other in short range to be sure they comply to "their rules". This way they all are able to go with higher margins.
Which of those market models do you think fits better the telcos? The one with high competition or the oligopoly with high entry barriers, uh?
"You still haven't explained why the ball "belonging to dog" is written "dog's ball" but the ball "belonging to it" isn't written "it's ball"."
Yeah, yeah, whatever...
Now for the really interesting question: Who's on first?
"The point is that artificial, technological restrictions are stupid, and it's a mystery as to why people choose to submit to them."
Because when you have no choice you don't have to think about the choice.
People *like* being stupid for the most part (and by "people" I mean "everybody"; you and I included).
"There are companies out there that just build data centers."
The fact that they "just build data centers" is no guarantee that they will build them better than the company that build datacenters on top of something else. Once you are big enough in an industry, there's no guaranteed benefit in going out for a provider. In fact you are *guaranteed* in the limit that it will be cheaper to do it yourself (your costs will be just the same than those of the provider but you will save on its benefits).
So the question is not why they build their own datacenter but if they'll be up to the task.
"... and if I was trying to host an illegal file sharing site on one of their clouds they would shut it down for exactly the same reasons as they did WikiLeaks."
Except for the "illegal" part.
"It is stuff that is illegal to have."
Don't think so. And provided that no USA fiscal office has started procedures against Wikileaks for the published contents, so do think USA legal body. It is stuff that it may be illegal to share if you are managing it. Perfectly legal to have and publish if you have managed to get to somehow. That's why, in case of any doubt, those pesky UK terrorists made a point introducing the First Amendment to their Constitution. Cfr. the legal situation of Bernstein & Woodward versus "Deep Throat" regarding the Watergate scandal.
"My gut feeling agrees with the AC that over-use of bandwidth may be the case."
I hear about "bandwidth over-use" once and again, but how the hell could anyone over use a resource provided on-demand? In this case, the network provider have all the ability, even on the cheap ,to control their customers' resource usage. If they contracted 1TB/month per server, how could they be able to use one single bit more than this? The provider could just say "see? 1TB; you are off till next month the first" a simple firewall rule after that and they are done.
Or might it be the case that even when the provider in fact offered them 1TB/month to its customer they now say "OMG, they are in fact using what we promised them by contract, we can't tolerate that!"?
"Sorry, but you are a bunch of amateurs - that's all there is to it."
Because they depend on third parties? Surely so you do and so do everybody in business.
Provided (a big "if") their contract clearly stated they were offered 100TB per server, it would be the provider unable to hold to their contract the amateurish one (if not blatantly illegal).
I would look for legal advice since it seems they should be able to claim for contract breaching and damages for the lost business revenue on top of that.
"And I say that lowering the levels of CO2 will cause fairies to explode into flower scented farts. The burden of proof is on you to prove that it won't happen."
OK, I'll do. Lowering the levels of CO2 from current 380ppm to 280ppm will put them as they were by the year 1800 and it's known fairies didn't explode back then.
Your turn.
"I am a bit concerned that what appears to be an entire agency moving its operations toward complete dependency on a single commercial entity."
The point is that they probably *already* are completly dependent on a single commercial entity. Other poster said USDA is basically a Microsoft shop to a point they can't realistically go away from it (wonderful Microsoft lock-in) much less in these recession years. So going to Microsoft's cloud just makes sense: it will be cheaper and it will probably go a bit better managed (or at least, there's the hope that they will be able to sue if not).
The pity is that they went this short-sighted path instead of realizing what's the real culprit (Microsoft lock-in) and understand that the sooner they break the lock-in, the better and cheaper in the long run.
"There's plenty of proprietary source implementations of TCP/IP, you know."
And they work in a compatible way that allowed to its success due to the existance of an open source one to be tested against, you know.
"Which to me shows that what matters is an open API and public domain algorithms, not open source."
You are true, but it's a naive point of view. There's no way for anything more complex than a whistle that you can have an unambigous protocol definition. In order for the API or algorithms to gain wide adoption you need an open sourced implementation (BSD licensed if you want or expect close source distributing companies to adopt it).
"Open Source business model works well when your company model is a consulting based business."
Yes. But let's make "consulting" a wider word and it will become even truer. If your business is about contracting (I'll pay you to *do* this) instead of licensing (I'll pay you to be *allowed* to do this) then it makes quite a lot of sense.
It's not only consulting but developing on-demand too (and everything where "doing" is involved instead of "licensing").
"Suns model was product based business."
The problem is not what Sun's model *was* but that it wasn't able to adapt.
"IBM doesn't do many products any more"
So they *had* a business model and they *adapted* it to a changing environment.
"I've not worked as a programmer for, hmm, maybe 15 years and all of this was known way back even before I "retired" from that line of work."
Yes: there's an obvious problem with programming and it's that "we" as a guild are not building upon past experience. For the most part, the current generation of programmers are making the same kind of mistakes that where common -and learnt how to avoid, even 20 years ago. Can you imagine, say, aviation if you had to engineer an Airbus 380 all the way from Otto Lilienthal to-date instead of building on past knowledge? That's what I feel too many times when dealing with new development projects.
Now the question is: what could we do so new programmers start where the previous generation left instead of having to build their own knowledge almost all the way along again?
"The whole lot is full of contradictions"
No, it isn't. It goes "don't do that... but don't fall in the other extreme".
That's on line with his central idea that programming is "an art, one that requires a skilled hand and a creative mind to achieve a happy medium between problematic extremes".
"My employer uses F/OSS extensively - and as the sysadmin, I've started to notice a pattern."
I'd think you just noticed half the pattern.
"F/OSS products which scratch an individual or a small group of peoples' itch generally get developed to a certain point and then stagnate. If you're lucky, that point is acceptable to you."
If you are not so lucky, it's open source, go scratch your own itch and become part of the "group of people" that develop the product under your company's bankroll to its "certain point" and then let it stagnate. This way it'll grow tailor-made to your company's needs while still probably being cheaper and less risky than going the closed source path.
"The products that do really well - the "best of F/OSS", if you like - are almost invariably the sort which scratches a very common itch."
So there's a lot of people that do what I pointed out on my previous paragraph so you are less likely to be the one to step ahead into the "group of people" that makes it happen.
"They're usually bankrolled by a number of companies (the Linux kernel falls under this category) or become self-funding when the project leader sets up a company to sell a commercial version with support and possibly extra features."
For some of them it might be the case that your company is the one most interested on the project reaching a high level "scratching point", so it becomes a revenue opportunity for it.
"What I like about rolling releases is you get to deal with application incompatabilities one at a time as they come up"
Only if you can dedicate time each and every day to support efforts and you can afford not knowing when your system will be down for maintenance. Most people neither can nor want to afford that.
"rather than having to spend a week or few all at once when upgrading a distro."
False disyuntive. My experience with Debian (Woody, Sarge, Etch, Lenny) is that it takes me about four hours to calmly upgrade from one to the next, so it's either taking care of problems each and every day, not knowing when something will break, or about four hours about each two years.
"I think it's also probably better for security, as you get the latest patches for the software."
Almost no developer follows sounded software promotion practices, so you end up with security patches *and* new bugs at the same time. Not to talk about integration-related problems.
"If our allies suddenly can't trust our foreign service to keep classified information, well, classified, how likely do you think it is they will continue to trust our diplomats?"
Well, the obvious point here is that your foreign service is in fact unable to keep classified information, well, classified so why the rest of the world should trust your diplomats?
And, please, pay attention that your diplomats are not unable to keep secret things secret because of Assange but because of themselves, so why fingerpointing Assange instead of the real culprits, the diplomats, that's it?
"You can only take pride in being American if it took you some work to become one."
*Or* behave as one.
I can accept someone USA-born taking pride of his nationality if he really lives for what he percieves to be the "true nature" of being American (you know, the land of the free, the home of the brave and all that stuff) and it's up to the challenge of making sure those ideals reach intact or even embettered for those that will come after him.
I really don't know if that's the case here.
"Well said. You should have signed in."
Why? Would signing make his words any more true or false?
"the vast majority of the material that's been released shows no evidence of any government abuses"
There's more on the "abuse" word than "plain illegal", specially with respect to politics. If there's really no government abuses, why all the airing? Can it be because at least some people found the data supportive of at least questionable practices? And if it's indeed questionable practices at least for somebody, how can't it be considered whistleblowering?
"If he's innocent, he can go back and defend himself."
He *is* innocent. Do you remember that little thingie, "innocent until proven guilty"?
"he has little reason not to"
Given the timing of his accusation, one might thing that there is more than a "little reason" to be worried about.
"Is shipping the data to a country where it is known that you run an increased risk of agencies and interested 3rd parties taking a peek at the data 'due diligence'?"
Yes as long as local laws so they say.
And that's the case.
"It could be rather bad for a company to be found in breach."
But if its data is on a safe harbour and the contract says so, then they won't be found in breach. They certainly have done due dilligence.
"Let's play poker sometime. You play with your cards face up."
Exactly my point: modern foreign politics shouldn't be a poke game.