I'm not sure you can change your license from "v2 or later" to "v2 only" without permission. If you accepted contributions to a project under a "v2 or later" license, then your contributors were placing their code under *that* license, and a change to a "v2 only" license would constitute releasing contributors' code under a license they didn't approve.
Umm, yes they did. You have to watch your inclusions here. "v2 only" is a subset of "v2 or later". Also note that the "v2 or later" language is not actually part of the license (or a license type itself); it's just a phrase that is traditionally used to introduce the license in many releases. It's equivalent to saying "This software is simultaneously released under v2, and v3, and v4, and v5, and... v_n -- take your pick".
Once you have that situation, the release under v3 does not 'trump' or efface the release under v2; it is just an additional license possibility.
In the case we're discussing, anybody who picks up the code to reuse it can choose to either release the result under "v2 only" or under "v2 or later". In the the first case, they're picking up the "v2" option of the initial package. Remember, the initial package is still out there in the ether, available under v2, v3, v4,..., vn, so the restriction only applies to the mod.
Note that this reading relies heavily on the use of the word OR in "... or later". It provides the choice to the person receiving the code as to which license they'll accept. If the original stipulation said "v2 AND any later version", we'd be in the sort of "v3 trumps v2" situation everyone seems to be worrying about - but if these issues ever came to trial, I seriously doubt any judge would accept as valid a release under a license that did not yet exist at the time the license was granted. (IANAL).
I know, I figured as much, but I couldn't resist making the nerd-ween joke. Which says a lot about where I'd be found on the Slashdot bellcurve, compared to the people who read it and said nothing...
OTOH, if you considered my previous post 'castigation', you're much too thin-skinned to be posting to Slashdot. I'd class it at the 'friendly gibe' level, myself.
In truth it's Time acknowledging we are a narcacistic society.
... it perfectly captures our unholy stew of sarcasm and narcissism, as in, "Sure, we're all equal... sure, anybody can grow up to be president... sure, the law treats everybody the same..."
With a little effort, we could make 'narcacistic' completely cromulent.
Pure FUDD... If you follow the nested links to the actual hatin' on the OLPC, you find out that most of the $970 figure is a $542 estimate of the cost of internet access, per laptop, spread over 5 years. The other estimates (training, lossage) may be reasonable, but this knocks it into lala land.
Agreed. When I first saw these sample 'innovations', I had to double-check to make sure I wasn't reading one of those look-and-feel satires on The Onion or someplace. These are the strongest arguments he can fill his column-inches with? This is what an annual 7 Billion Dollars of R&D money gets you? Sheesh.
"beat up on..." is a colloquialism, most heavily used in rural and/or uneducated language groups ("yobs", I believe, would be your word, right?). More interesting in general is the addition of the preposition to change meaning, which is standard usage on both sides of the pond. "to beat up" is a 'phrasal verb' -- a term I didn't know until I came across advanced English instruction books in France. Think about the different meanings foreigners have to learn for:
to put, to put on, to put over, to put up, to put up with, to put upon, to put through (e.g. school), to put across, to put to ("I put it to him...") , to put in,.... etc.
These are not euphemisms; they're just distinct meanings depending on the associated preposition. It's basically the same thing the Germans do with prepositional prefixes to their verbs (which sometimes get detached).
As long as we're airing pet peeves about regional English... I can deal with dropped articles ( "I had to go to hospital.../ to university"), but when are you going to do something about verb-subject agreement?
"The committee are proposing..."
I mean really, there's just one committee, it's a singular noun, it takes a singular verb!
I'm just having you on... let's go make fun of Australians now!
Agreed - I was going to make a similar comment until I saw yours. Here's a related calculation -- what is the cost of their failure rate?
Producing 15 billion per year, with 18 defects per million, gives us:
1.5e10 x ( 18 / 1e6 ) = 270,000 pieces per year.
If we assume their marginal cost of production for one brick is 1 cent (a generous estimate, I would think), this means that the cost of their defects per annum is... $ 2,700.
I consider that impressive, as a measure of their process engineering. If their defect rate was 180 per million, would you accept the project to get it down to 18 per million, with a budget of $ 27,000 per year for the improvements?
Basically, you can have an item in a multi-star system -- is it in a stable orbit around one of them, or is it just doing a few loop-de-loops on its way through? Can it orbit 2, 3,..., n stars at once? http://imagine.gsfc.nasa.gov/docs/ask_astro/answer s/980122c.html With multiple star systems with large interstellar distances, its orbit could be millions of years -- by which time the stars have changed relative position, in which case 'stable orbit' loses its meaning. Where do you draw they line between 'planet in complex orbit' and 'floating rock interfering in a multi-star system' ? The folks who want a dynamical component to the definition are wrong on this one, I think.
There has been an undoubted increase in bristlecone pine ring widths in the 20th
century. Graybill and Idso [1993] explicitly stated it is greater than could be explained
by temperature.
Graybill and Idso are the source of the original pine measurements:
Graybill, D.A., and S.B. Idso. 1993. Detecting the aerial fertilization effect of atmospheric CO2
enrichment in tree-ring chronologies. Global Biogeochemical Cycles Vol. 7, pp. 81-95.
Even one of Mann's co-authors in MBH98 raises the question: (again, from M&M)
Despite the reliance of MBH98 on the North American PC1, the validity of this
series as a temperature proxy was not independently established in peer-reviewed
literature. Co-author Hughes stated later [Hughes and Funkhouser, 2003] that the
anomalous growth rate of bristlecone pines was a "mystery", which should have raised
questions about the PC1.
Hughes, M.K. and G. Funkhouser. 2003. Frequency-dependent climate signal in upper and
lower forest border trees in the mountains of the Great Basin. Climatic Change: Vol. 59,
pp. 233-244
That's a few to get you started... check out the M&M PDF I linked to above; lots of references to others that place doubt on bristlecone pine reliance are there, starting on page 14. Issues have to do with the fact that the species is anomalous, and MBH98 seem to claim that CO2 had more of an effect in the 19th century than in the 20th. (Sure, M&M are lsiting these references, but at least it's a start -- we can assume at least Hughes isn't biased in their favor.)
If you keep asking me questions and sending me out to do more research, you may end up turning me into a GW denier in spite of myself !:)
in other words how close to a 100% majority should the "collective" (government) need in order to impose coercion as their means.
I say 99.99999% - what do you think?
I think I'd like to know the size of the population you're taking a percentage of, so I know how many members I need in my soon-to-be-formed criminal conspiracy, to guarantee its safety. Thanks!
Heywaitaminnit.... impose coercion ? As opposed to getting people to submit to it voluntarily?
By destroying the credibility of the Monckton article in several of its major assertions, it makes any claim by Monckton suspect. This is basic skepticism 101.
Note where in my original comment I said we should use higher standards of rhetoric. I agree with you that several claims Monckton made were discredited, but that only discredits his conclusion, not all his subsidiary claims. (You are applying the standards of law, where only the main conclusion matters, not those of rhetoric, where subsidiary claims can be considered separately. ) If this were a legal case, and his PDF were his argument, he would lose -- but by the standards of scientific debate, we should consider his other claims. What if someone else wrote another article, leaving out his errors -- would that be more convincing?
We apparently agree that Monckton's main conclusion is wrong -- we only disagree about the extent to which he must be proven wrong. To take an example or two, if someone argues that evolution happens, because it's evident that giraffes have long necks due to generations of stretching them to reach tasty leaves, you would accept his conclusion but reject his supporting argument. Similarly, you'd probably reject astrological conclusions drawn from an (otherwise correct) description of planetary motions.
There is educational value in making the case for GW clear and bulletproof -- it's not enough to say 'the specialists agree; we just need to believe them.'. I'm still waiting to find out what's wrong with Monckton's assertion that bristlecone-pine climate data was overweighted by a factor of 390 in determining historical temperatures in order to make the medieval warming period go away in the hockey-stick graph. I believe there's a refutation of that; I just don't know what it is. The nerd in me would like to know the details. For example, here's a list of publications in refereed journals, some of which Monckton does refer to: http://www.uoguelph.ca/~rmckitri/research/trc.html
Note also Monckton's claim that the UN recommended against relying on bristlecone-pine data, but it is the principal support for the hockey stick. (Here's a detailed reference on this issue: http://www.multi-science.co.uk/mcintyre-mckitrick. pdf These sorts of claims are worth refuting in themselves, even if we disagree with his conclusion, and even if some other claims in the same article have been discredited already.
My point is not about the details of climate science, but about the scientific process. If the case for GW is so open-and-shut (and I believe it is) we should be able to formulate a rebuttal of the Monckton article that meets each of his points head-on.
What I would like to see is the sort of this-is-why-relativity-is-true presentation of GW, that can convince everyone but the irrational habitual sceptics, that pre-refutes all of the objections Monckton is making. From all appearances, Monckton is at least trying to hold a civilized discussion on the subject. Being wrong doesn't make him irrational or uncivilized, and we do science no service by treating him as such.
True, nobody says "I believe in gravity", but... people can be wrong and still educable. Assuming that we're not facing someone with intelligent-design-type intransigence in the face of rational discussion, this is usually worth doing. It is not only a service to the original proponent, it helps anyone who's been following the discussion.
Every generation needs to learn anew that relativity is counter-intuitive but still right, that light is modeled as both a particle and a wave, and insisting on one or the other leads to error; that you can't trisect an angle or square the circle, that pi is transcendental, that continental drift happened, etc. By my lights, one of the best ways to do that is to expose people to the debates over the issues.
Agreed that political opposition to Kyoto motivates some people; they'd be better served by highlighting that than by trying to undercut the science. I was hearing about the GW debate for quite a while before I learned about the developing-countries Kyoto exemption.
The first article did NOT make an argument. Instead it attempted to convince everyone he was correct by saying how X,Y and Z arguments are false. But there are 100's of arguments for global warming. OF COURSE some of them are false.
This is a pretty low standard... I believe in global warming. But it's also possible that the arguments for GW are simultaneously 1)true and 2) overhyped. The first article was the least shrill, least tendentious, attempt I have read to present the 'case against'. If we're right about GW, it shouldn't be hard to disprove it, using the same or higher standards of both rhetoric and logic. I notice that TFA didn't say anything specific about the 390x overweighting of bristlecone-pine climate data and its use in erasing the the warm period during the middle ages. I'm willing to believe the original article got it wrong; since science is on our side, it should be easy to explain how.
Further -- the Monbiot article says that "climate sensitivity is an equilibrium concept" -- meaning that CO2 release precedes its effects by several decades. Nice, but the original (Monckton) article claimed that the problem was that warming preceded CO2 rise, which means Monbiot didn't really rebut him. There are many, many specific claims in the original article (and the linked-to PDF is even more detailed); Monbiot tackles very few of them adequately. Rather than slamming a journalist for lack of sufficient credentials, he should be congratulated for attempting to meet the scientists halfway by speaking their own language, and set right where he needs to be set right. The truth has nothing to fear from polemic.
Umm, yes, that is how it works. Read the excerpt again:
If the Program specifies a version number of this License which applies to it and "any later version", you have the option of following the terms and conditions either of that version or of any later version published by the Free Software Foundation.
That's direct from the GPLv2. Note the "you have the option" phrase. What a copyright holder can do is say "this is released under GPL2 and any future version", which means you can choose to accept either the GPLv2 or the GPLv3 license (just as you can license MySQL under a GPL or a proprietary license). The existence of one license path does not cancel the existence of the other.
What you are saying would make a nonsense of contract law - "I'm releasing this code under terms which may change in the future, and are not known even to me at the present time. At the moment, these terms here [GPLv2] are valid, but they may become invalid at some point in the future due to actions taken by entities not a party to this contract." That wuold be laughed out of court in any jurisdiction. Even the GPL explains that the choice of which version to use is the recipients if the 'any future version' clause is used to present the license.
About projects Foo and Bar -- even if the holder of Bar decides to re-release Foo (modified or not) exclusively under GPLv3, it doesn't make the initial GPLv2 release go away. And my claim (which I think would at least merit a hearing in court) is that GPLv2's provision about 'no added restrictions' might lead to a successful legal argument that GPLv3 is a restriction which means that the owner of project Bar loses his right to use Foo code because he's violated the GPLv2 provision.
Your 'two points to remember' are basically saying that somebody can release code under the GPL, a modification of the GPL, or an entirely different license. True, but irrelevant. Once something is released under GPLv2, it stays released under GPLv2 for all time, no matter what else happens to it. That's the whole point of the GPL.
Anything released by me in the future will carry a modified GPLv2 that does not permit the use of any future version of the GPL simply because this is a deliberate railroading of the purpose of the GPL.
You don't need to do this, just specify "GPLv2 ONLY" when you describe your license. Read on:
If the Program specifies a version number of this License which applies to it and "any later version", you have the option of following the terms and conditions either of that version or of any later version published by the Free Software Foundation. (GPLv2)
The 'any later version' stipulation is not part of GPLv2, it's part of the text that many people use to announce which license they're using. So you don't need to modify it (furthermore, you can't, if you're releasing a mod rather that something you coded up from scratch).
More interesting to me is article 6 of the GPLv2:
Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions. You may not impose any further restrictions on the recipients' exercise of the rights granted herein.
I can see a number of lawyers making hay of that, saying that GPLv3 introduces further restrictions. Note that when someone releases under "GPLv2 or any future version", the choice belongs to the recipient, not the licensor. The licensor can't grant rights under a license that doesn't yet exist when he makes the grant.
So, what do we get? All existing GPLv2 software is, and will forever continue to remain, licenseable as GPLv2. Even a new version of an existing program released under GPLv3 will still have its prior version and source available under GPLv2. And the first lawyer for IBM/RedHat/Novell who cares enough will use article 6 of GPLv2 to declare the extra restrictions of GPLv3 invalid for the new version of the existing program.
It's a shame that Stallman has gone on this crusade, but my money says GPLv2 is here to stay. Not many people will be releasing GPLv3 code, and the GPLv2 mainstream will fork them into irrelevancy.
Aha, rackmount UPS, ok, that's different. I was picturing one of those big behemoths between the outside power and the under-floor 30-amp circuit.
As for Loews, agreed, but its a crapshoot. If I was in a Loews, and there was a blackout lasting more than 10 minutes, i'd be buying one if I didn't have one at home already. And good luck finding 10, especially after you've been living on battery power for an hour already and have decided it may last longer and you need to mitigate...
But really, you got me, I was just posting to work in the reference to the Seinfeld Frogger episode, the apropos-ness of which was diluted somewhat by those damn rackmount UPSs.
Mmmm, okay, they don't have to run in phase. So, assuming that you're talking about the emergency scenario described above (a multi-day blackout, batteries run out), and not talking about stocking them in advance, you run into a few issues:
1. Lowes tends to run out of generators during blackouts. They're big, and they don't stock that many. Don't count on finding 10 of them during a blackout to run your 10 mission-critical apps.
2. Assuming you find the generators, even forgetting about phase, how exactly are you going to bring them on-line, with their little 110V outlets, without interrupting operations? We're not talking about Costanza's Frogger Game here. You are talking about a multiserver platform for an enterprise app -- you have to bring it down and back up again in a very structured way, with hours of downtime. Aaand... by your own comment (GP? GGP?) the $1M in losses came from only hours of downtime.
Absolutely! Hey, Gary Larson was thrilled to have an owl louse named after him:
Strigiphilus garylarsoni Clayton, ~1989 (owl louse) "I considered this an extreme honor. Besides, I knew no one was going to write and ask to name a new species of swan after me. You have to grab these opportunities when they come along." - Gary Larson
Your argument is a general argument against using any new technology, at any time, because at some point it will get even better. I'm not sure what your amortization worry is, either. You pay a fixed price for the windmill, which becomes a sunk cost. You are taking a certain amount of risk that the windmill will last 20 years -- but it will either last, or it won't. That doesn't change if suddenly some new technology comes along. You're still getting 20 years worth of electricity covered by your initial investment, which is what you wanted. (The amortization schedules of Opteron servers don't suddenly change once Intel releases quad-cores, and they're still putting out as much processing power as they did the day before).
If your question is, 'what is the best way to manage $XK dollars today to give me 20 years worth of electricity at minimal cost' (buy a windmill today, or buy market power for 5 years and then buy a much cheaper device), that is an interesting but different question unrelated to amortization. Given the fluctuations of the energy market, you could probably still buy your windmill today, buy some long-term electrical-power-price options, and make a profit or at least come out even if the price of power drops significantly in 5 years.
Slashdot Mirror
Once you have that situation, the release under v3 does not 'trump' or efface the release under v2; it is just an additional license possibility.
In the case we're discussing, anybody who picks up the code to reuse it can choose to either release the result under "v2 only" or under "v2 or later". In the the first case, they're picking up the "v2" option of the initial package. Remember, the initial package is still out there in the ether, available under v2, v3, v4,
Note that this reading relies heavily on the use of the word OR in "... or later". It provides the choice to the person receiving the code as to which license they'll accept. If the original stipulation said "v2 AND any later version", we'd be in the sort of "v3 trumps v2" situation everyone seems to be worrying about - but if these issues ever came to trial, I seriously doubt any judge would accept as valid a release under a license that did not yet exist at the time the license was granted. (IANAL).
... You must LOOK UP to see the source of the GIANT WHOOSHING SOUND
I know, I figured as much, but I couldn't resist making the nerd-ween joke. Which says a lot about where I'd be found on the Slashdot bellcurve, compared to the people who read it and said nothing...
OTOH, if you considered my previous post 'castigation', you're much too thin-skinned to be posting to Slashdot. I'd class it at the 'friendly gibe' level, myself.
You're right, it has to be true; I read it in Wikipedia!
With a little effort, we could make 'narcacistic' completely cromulent.
Sarcasm detector? Now that's a really useful invention.
Pure FUDD... If you follow the nested links to the actual hatin' on the OLPC, you find out that most of the $970 figure is a $542 estimate of the cost of internet access, per laptop, spread over 5 years. The other estimates (training, lossage) may be reasonable, but this knocks it into lala land.
Agreed. When I first saw these sample 'innovations', I had to double-check to make sure I wasn't reading one of those look-and-feel satires on The Onion or someplace. These are the strongest arguments he can fill his column-inches with? This is what an annual 7 Billion Dollars of R&D money gets you? Sheesh.
"beat up on..." is a colloquialism, most heavily used in rural and/or uneducated language groups ("yobs", I believe, would be your word, right?). More interesting in general is the addition of the preposition to change meaning, which is standard usage on both sides of the pond. "to beat up" is a 'phrasal verb' -- a term I didn't know until I came across advanced English instruction books in France. Think about the different meanings foreigners have to learn for:
.... etc.
... I can deal with dropped articles ( "I had to go to hospital .../ to university"), but when are you going to do something about verb-subject agreement?
..."
... let's go make fun of Australians now!
to put, to put on, to put over, to put up, to put up with, to put upon, to put through (e.g. school), to put across, to put to ("I put it to him...") , to put in,
These are not euphemisms; they're just distinct meanings depending on the associated preposition. It's basically the same thing the Germans do with prepositional prefixes to their verbs (which sometimes get detached).
As long as we're airing pet peeves about regional English
"The committee are proposing
I mean really, there's just one committee, it's a singular noun, it takes a singular verb!
I'm just having you on
Agreed - I was going to make a similar comment until I saw yours. Here's a related calculation -- what is the cost of their failure rate?
... $ 2,700.
Producing 15 billion per year, with 18 defects per million, gives us:
1.5e10 x ( 18 / 1e6 ) = 270,000 pieces per year.
If we assume their marginal cost of production for one brick is 1 cent (a generous estimate, I would think), this means that the cost of their defects per annum is
I consider that impressive, as a measure of their process engineering. If their defect rate was 180 per million, would you accept the project to get it down to 18 per million, with a budget of $ 27,000 per year for the improvements?
Here's why:
..., n stars at once? http://imagine.gsfc.nasa.gov/docs/ask_astro/answer s/980122c.html With multiple star systems with large interstellar distances, its orbit could be millions of years -- by which time the stars have changed relative position, in which case 'stable orbit' loses its meaning. Where do you draw they line between 'planet in complex orbit' and 'floating rock interfering in a multi-star system' ? The folks who want a dynamical component to the definition are wrong on this one, I think.
http://www.2dplay.com/orbit/orbit.swf
Basically, you can have an item in a multi-star system -- is it in a stable orbit around one of them, or is it just doing a few loop-de-loops on its way through? Can it orbit 2, 3,
( http://www.multi-science.co.uk/mcintyre-mckitrick
and found this:
Graybill and Idso are the source of the original pine measurements:
Even one of Mann's co-authors in MBH98 raises the question: (again, from M&M) That's a few to get you started
If you keep asking me questions and sending me out to do more research, you may end up turning me into a GW denier in spite of myself !
Heywaitaminnit
We apparently agree that Monckton's main conclusion is wrong -- we only disagree about the extent to which he must be proven wrong. To take an example or two, if someone argues that evolution happens, because it's evident that giraffes have long necks due to generations of stretching them to reach tasty leaves, you would accept his conclusion but reject his supporting argument. Similarly, you'd probably reject astrological conclusions drawn from an (otherwise correct) description of planetary motions.
There is educational value in making the case for GW clear and bulletproof -- it's not enough to say 'the specialists agree; we just need to believe them.'. I'm still waiting to find out what's wrong with Monckton's assertion that bristlecone-pine climate data was overweighted by a factor of 390 in determining historical temperatures in order to make the medieval warming period go away in the hockey-stick graph. I believe there's a refutation of that; I just don't know what it is. The nerd in me would like to know the details. For example, here's a list of publications in refereed journals, some of which Monckton does refer to:
http://www.uoguelph.ca/~rmckitri/research/trc.htm
Note also Monckton's claim that the UN recommended against relying on bristlecone-pine data, but it is the principal support for the hockey stick. (Here's a detailed reference on this issue: http://www.multi-science.co.uk/mcintyre-mckitrick
What I would like to see is the sort of this-is-why-relativity-is-true presentation of GW, that can convince everyone but the irrational habitual sceptics, that pre-refutes all of the objections Monckton is making. From all appearances, Monckton is at least trying to hold a civilized discussion on the subject. Being wrong doesn't make him irrational or uncivilized, and we do science no service by treating him as such.
True, nobody says "I believe in gravity", but ... people can be wrong and still educable. Assuming that we're not facing someone with intelligent-design-type intransigence in the face of rational discussion, this is usually worth doing. It is not only a service to the original proponent, it helps anyone who's been following the discussion.
Every generation needs to learn anew that relativity is counter-intuitive but still right, that light is modeled as both a particle and a wave, and insisting on one or the other leads to error; that you can't trisect an angle or square the circle, that pi is transcendental, that continental drift happened, etc. By my lights, one of the best ways to do that is to expose people to the debates over the issues.
Agreed that political opposition to Kyoto motivates some people; they'd be better served by highlighting that than by trying to undercut the science. I was hearing about the GW debate for quite a while before I learned about the developing-countries Kyoto exemption.
Further -- the Monbiot article says that "climate sensitivity is an equilibrium concept" -- meaning that CO2 release precedes its effects by several decades. Nice, but the original (Monckton) article claimed that the problem was that warming preceded CO2 rise, which means Monbiot didn't really rebut him. There are many, many specific claims in the original article (and the linked-to PDF is even more detailed); Monbiot tackles very few of them adequately. Rather than slamming a journalist for lack of sufficient credentials, he should be congratulated for attempting to meet the scientists halfway by speaking their own language, and set right where he needs to be set right. The truth has nothing to fear from polemic.
What you are saying would make a nonsense of contract law - "I'm releasing this code under terms which may change in the future, and are not known even to me at the present time. At the moment, these terms here [GPLv2] are valid, but they may become invalid at some point in the future due to actions taken by entities not a party to this contract." That wuold be laughed out of court in any jurisdiction. Even the GPL explains that the choice of which version to use is the recipients if the 'any future version' clause is used to present the license.
About projects Foo and Bar -- even if the holder of Bar decides to re-release Foo (modified or not) exclusively under GPLv3, it doesn't make the initial GPLv2 release go away. And my claim (which I think would at least merit a hearing in court) is that GPLv2's provision about 'no added restrictions' might lead to a successful legal argument that GPLv3 is a restriction which means that the owner of project Bar loses his right to use Foo code because he's violated the GPLv2 provision.
Your 'two points to remember' are basically saying that somebody can release code under the GPL, a modification of the GPL, or an entirely different license. True, but irrelevant. Once something is released under GPLv2, it stays released under GPLv2 for all time, no matter what else happens to it. That's the whole point of the GPL.
More interesting to me is article 6 of the GPLv2: I can see a number of lawyers making hay of that, saying that GPLv3 introduces further restrictions. Note that when someone releases under "GPLv2 or any future version", the choice belongs to the recipient, not the licensor. The licensor can't grant rights under a license that doesn't yet exist when he makes the grant.
So, what do we get? All existing GPLv2 software is, and will forever continue to remain, licenseable as GPLv2. Even a new version of an existing program released under GPLv3 will still have its prior version and source available under GPLv2. And the first lawyer for IBM/RedHat/Novell who cares enough will use article 6 of GPLv2 to declare the extra restrictions of GPLv3 invalid for the new version of the existing program.
It's a shame that Stallman has gone on this crusade, but my money says GPLv2 is here to stay. Not many people will be releasing GPLv3 code, and the GPLv2 mainstream will fork them into irrelevancy.
Aha, rackmount UPS, ok, that's different. I was picturing one of those big behemoths between the outside power and the under-floor 30-amp circuit.
As for Loews, agreed, but its a crapshoot. If I was in a Loews, and there was a blackout lasting more than 10 minutes, i'd be buying one if I didn't have one at home already. And good luck finding 10, especially after you've been living on battery power for an hour already and have decided it may last longer and you need to mitigate...
But really, you got me, I was just posting to work in the reference to the Seinfeld Frogger episode, the apropos-ness of which was diluted somewhat by those damn rackmount UPSs.
Agreed on the outside-the-box, though.
Mmmm, okay, they don't have to run in phase. So, assuming that you're talking about the emergency scenario described above (a multi-day blackout, batteries run out), and not talking about stocking them in advance, you run into a few issues:
... by your own comment (GP? GGP?) the $1M in losses came from only hours of downtime.
1. Lowes tends to run out of generators during blackouts. They're big, and they don't stock that many. Don't count on finding 10 of them during a blackout to run your 10 mission-critical apps.
2. Assuming you find the generators, even forgetting about phase, how exactly are you going to bring them on-line, with their little 110V outlets, without interrupting operations? We're not talking about Costanza's Frogger Game here. You are talking about a multiserver platform for an enterprise app -- you have to bring it down and back up again in a very structured way, with hours of downtime. Aaand
Absolutely! Hey, Gary Larson was thrilled to have an owl louse named after him:
Strigiphilus garylarsoni Clayton, ~1989 (owl louse) "I considered this an extreme honor. Besides, I knew no one was going to write and ask to name a new species of swan after me. You have to grab these opportunities when they come along." - Gary Larson
If your question is, 'what is the best way to manage $XK dollars today to give me 20 years worth of electricity at minimal cost' (buy a windmill today, or buy market power for 5 years and then buy a much cheaper device), that is an interesting but different question unrelated to amortization. Given the fluctuations of the energy market, you could probably still buy your windmill today, buy some long-term electrical-power-price options, and make a profit or at least come out even if the price of power drops significantly in 5 years.