It's like the whole argument that the economy is more important than the environment while completely ignoring the fact that the economy can't exist without the environment - but taken to a new ridiculous level.
Sure, "The Ecosystem" is obviously more important if its ability to function as a whole is really threatened. DDT was never that sort of threat - sure, it screws up fish and birds when it bio-accumulates but that's pretty much it.
One thing that people are really bad at is understanding complex systems that can only be usefully described mathematically or with large numbers. That applies to the economy, the environment, and any discussion involving "millions" of anything.
The question with DDT is a very simple one: how many horribly sick birds does it take to really be worth letting millions of people die instead? There's a rational answer to that question based on the ecological realities involved. Any opinion on the topic that isn't based on a rational analysis of the actual trade-offs involved is dumb and wrong.
Romney, Edwards, and Clinton are all suggesting mandatory health insurance as a solution to the US healthcare problems. The only thing I was trying to argue against is that policy proposal.
One of the selling points of PHP is that one *can* fairly easily make libraries for it in C if need be.
This is true for most programming languages. It's very easy for the "high level scripting languages" like Perl, Python, Ruby, Lua, Pike, moderately harder for languages like OCaml and Erlang, and still possible for languages that live "off in their own world" like Java and most Lisps.
Odd... and here I though there were only 193 UN-recognized countries, plus a few "somewhat recognized" ones. So the US have permanent military bases in more countries than there are on this planet ?
See what I mean? When you actually verify numbers and what their implications are, all kinds of interesting things come up. Let me try again with real numbers: "500+ Military bases in 50+ countries".
Are you pro-war? Do you think the US needs to maintain permanent military bases in 200+ countries including Germany, Japan, and South Korea? Do you think that the the "security funding" in the US federal budget should exceed all other discretionary spending combined by 50%?
If not, then I can't see how you could support any candidate other than Paul or Kucinich. Sure, some of their positions on other issues are a little odd. Maybe the women of Iowa would have to drive 200 miles to get an abortion if Paul was president and got his way on Roe vs. Wade (which would be seriously unlikely). Maybe we'd end up with some overly expensive socialized medicine program and some silly commitments on global warming if Kucinich was president.
Personally, I just can't see those issues as being terribly important when $0.23 out of every tax dollar goes to either bombing foreigners, spying on Americans, or funding lobbyists to try to more money for those two practices. If you disagree with me, I suggest you spend some time looking at the numbers involved - either of deaths from our interventionist foreign policy or just how big these budget numbers really are. The military PR industry is damn good at what they do, and humans are really bad at intuitively grasping importance in numbers, but if you vote in the US you really have a responsibility to try.
You think all those 'advanced countries' would have learned by now that government is the *problem* not the *solution*.
And a government mandate that citizens pay money to certain government-approved companies is somehow a good idea? Mandatory health insurance is just a new tax payed directly to the campaign contributors. I mean, sure, the insurance industry and the pharmaceutical industry love this plan - but it's worse for the people than *either* legitimate socialized health care or an actual free market would be.
Cellular telephones use digital radio signals divided into "channels". One channel gets you a cellular-quality digital voice signal. Cellular broadband is just transmitting IP over 2 or 4 of those same channels.
Have you priced out just fixing the problem by buying a real connection?
T1 lines are damn cheap now - I frequently see prices around $400/month. Optical lines start in the low thousands. All it would take is a couple neighbors and setting up a WiFi or even DSL co-op becomes competitive with what you'd expect DSL/Cable to cost.
Now, that may not be the right answer for getting internet access occasionally at your parents house - but it absolutely is for anyone personally lives somewhere where the telcos won't provide service (or won't provide good service).
As marcosdumay mentioned, the worst case with an unrestricted "Version X or later" clause is that the FSF releases a more permissive GPL. With a proxy, the worst case is that the proxy fails to accept future versions of the license that are reasonable.
In one case, your software can be combined with proprietary software. In the other, your software *cannot* be combined with free software. Considering the chances of these scenarios (reasonably low for a malicious FSF, reasonably high for a ineffective proxy) and the harm that would come from them (an incompatible license could make your software worthless, an overly permissive license merely prevents your software from exerting pro-freedom pressure).
Basically, your choices break down like this:
You want your code to be part of the GPL commons, even if that means that the FSF has the power to - completely against their nature - allow proprietary exploitation of your code in the future.
You want to make absolutely sure that your code will never be exploited by proprietary software vendors, even if that means that it may not be reusable by other free software projects in the future.
Now, my argument absolutely does apply to using the New-BSD or Apache license rather than the GPL. The question is simply if there is value to the idea of copyleft - and if you want your code to be usefully copyleft (which applies reusability), the only rational choices are "GPLvX or Later" or maybe the Artistic License 2.0.
I'm not comfortable with the "or any later version" clauses many GPL programs have.
Consider very carefully what the actual potential costs and benefits of such a clause are before deciding not to use it.
One of the key advantages to using any version of the GPL is that your code can be combined with other code that was written separately and also released under the GPL. "Version X or later" code can always be combined. When the next version comes out, "Version X only" code will be uncombinable. That basically means that - unless your project is Linux sized and can get away with having its own license - "Version X or later" is the only answer that will allow your project to outlive your personal work on it.
Every GNU+Linux distribution (which includes Ubuntu and Red Hat) already ships a bunch of GPLv3 applications. From the perspective of companies that distribute general purpose operating systems, GPLv3 is strictly better than GPLv2 because of the internationalized wording and the "contributors can't screw the community with patents" provisions.
The only reason for these bodies to exist is to make sure the peasants don't care enough to pick up their pitchforks. If the peasants don't care, the research proceeds. If the peasants are pissed off, public education campaigns occur first.
In a perfect world, maybe.
Back in the real world, enough irrational protest can prevent valuable research from occurring - and leave the related commercial sector scraping by with dysfunctional, archaic, and dangerous technology. For a good example, see nuclear power in the USA.
If you read my example, it is an example of gaming. The person's actual preferences are this:
Obama: Yes
Hillary: Yes
Bush v3.0: No
If those were actually her preferences, she'd vote them. But in your example, those aren't her preferences. Instead, her preferences are these:
Obama: Maybe
Hillary: Yes
Bush: No
In a case like that, either voting Hillary-only or Hillary-and-Obama would be an honest vote. It would only be a dishonest vote if she voted for Obama and not Hillary or Bush and not Obama. Only dishonest votes count as gaming.
So how many megawatts of power generated do you need to fill the balloons for a child's birthday party? How much would Helium have to be worth for it to be an economically viable plan to change the reactor design to make non-radioactive Helium recoverable?
The fact that old computers can be recycled and used by people who might not otherwise have had their own computer *does not* imply that everyone should stop buying new computers. First, if people stopped buying new computers then there would be no used computers to give to your nieces and nephews. Second, if people stopped buying new computers then there would be no R&D budget to develop better hardware - and just because you personally are happy enjoying the results of 50 years of exponential progress in computer hardware doesn't mean that the rest of us should be deprived of the potential results of 60 or 100 years of that process.
I'm sure it'd help if it wasn't a frankensteinian mix of c, python and c++ then.
Most non-trivial programs end up being a mix a lot like that. It's not frankensteinian at all - that's how one normally develops software. You use Python (or Perl, or Ruby, or TCL, or Lua) for non-performance critical sections to save programmer time, and you use C/C++ for the performance critical sections or the sections that involve interfacing to anything that wasn't written in your high-level language of choice.
Truth by majority? Is that how we define truth these days? The fact that the majority infringes a law only says that people ignore the law. It doesn't influence it's validity.
Laws aren't about "truth". Every law is a social compromise between freedom and a livable society. In a democracy, which social compromises are made is decided by the people themselves - and their opinions will naturally change over time.
Why wouldn't a recount be possible with optical scan machines?
Re:The only measure I know of by which Rails sucks
on
Rails May Not Suck
·
· Score: 1
If you want a good measure of what sort of people are in a given community, go to their IRC channel and ask a question from their FAQ.
True enough, but remember: The set of communities with assholes in them and the set of communities that produce useful software tools overlap almost entirely.
Re:The only measure I know of by which Rails sucks
on
Rails May Not Suck
·
· Score: 1
PHP kicks everybody elses ass
Have you heard of Perl?
Now, it's possible that you're defining "Usable" as "Not in Perl", but Perl has more mature, high quality code available in it than any other "web language".
Re:My experience with RoR
on
Rails May Not Suck
·
· Score: 3, Informative
More than anything I'm worried about speed.
The standard story for dynamic language development applies:
Optimizing before you have a problem is a waste of time.
There are lots of ways to optimize, up to and including re-writing the bottlenecks in C/C++.
The most common optimization that's used with Rails is it's built-in support for caching, which can speed things up by quite a bit. You can get the same sort of results with a hand-optimized memcached setup any other dynamic language - but Rails gives it to you almost for free.
Slashdot Mirror
Sure, "The Ecosystem" is obviously more important if its ability to function as a whole is really threatened. DDT was never that sort of threat - sure, it screws up fish and birds when it bio-accumulates but that's pretty much it.
One thing that people are really bad at is understanding complex systems that can only be usefully described mathematically or with large numbers. That applies to the economy, the environment, and any discussion involving "millions" of anything.
The question with DDT is a very simple one: how many horribly sick birds does it take to really be worth letting millions of people die instead? There's a rational answer to that question based on the ecological realities involved. Any opinion on the topic that isn't based on a rational analysis of the actual trade-offs involved is dumb and wrong.
Where's the straw man?
Romney, Edwards, and Clinton are all suggesting mandatory health insurance as a solution to the US healthcare problems. The only thing I was trying to argue against is that policy proposal.
This is true for most programming languages. It's very easy for the "high level scripting languages" like Perl, Python, Ruby, Lua, Pike, moderately harder for languages like OCaml and Erlang, and still possible for languages that live "off in their own world" like Java and most Lisps.
See what I mean? When you actually verify numbers and what their implications are, all kinds of interesting things come up. Let me try again with real numbers: "500+ Military bases in 50+ countries".
Are you pro-war? Do you think the US needs to maintain permanent military bases in 200+ countries including Germany, Japan, and South Korea? Do you think that the the "security funding" in the US federal budget should exceed all other discretionary spending combined by 50%?
If not, then I can't see how you could support any candidate other than Paul or Kucinich. Sure, some of their positions on other issues are a little odd. Maybe the women of Iowa would have to drive 200 miles to get an abortion if Paul was president and got his way on Roe vs. Wade (which would be seriously unlikely). Maybe we'd end up with some overly expensive socialized medicine program and some silly commitments on global warming if Kucinich was president.
Personally, I just can't see those issues as being terribly important when $0.23 out of every tax dollar goes to either bombing foreigners, spying on Americans, or funding lobbyists to try to more money for those two practices. If you disagree with me, I suggest you spend some time looking at the numbers involved - either of deaths from our interventionist foreign policy or just how big these budget numbers really are. The military PR industry is damn good at what they do, and humans are really bad at intuitively grasping importance in numbers, but if you vote in the US you really have a responsibility to try.
And a government mandate that citizens pay money to certain government-approved companies is somehow a good idea? Mandatory health insurance is just a new tax payed directly to the campaign contributors. I mean, sure, the insurance industry and the pharmaceutical industry love this plan - but it's worse for the people than *either* legitimate socialized health care or an actual free market would be.
Nice troll. As a free-software supporting atheist, I can't think of a much more obnoxious mockery of those ideals than your post.
Cellular telephones use digital radio signals divided into "channels". One channel gets you a cellular-quality digital voice signal. Cellular broadband is just transmitting IP over 2 or 4 of those same channels.
Have you priced out just fixing the problem by buying a real connection?
T1 lines are damn cheap now - I frequently see prices around $400/month. Optical lines start in the low thousands. All it would take is a couple neighbors and setting up a WiFi or even DSL co-op becomes competitive with what you'd expect DSL/Cable to cost.
Now, that may not be the right answer for getting internet access occasionally at your parents house - but it absolutely is for anyone personally lives somewhere where the telcos won't provide service (or won't provide good service).
As marcosdumay mentioned, the worst case with an unrestricted "Version X or later" clause is that the FSF releases a more permissive GPL. With a proxy, the worst case is that the proxy fails to accept future versions of the license that are reasonable.
In one case, your software can be combined with proprietary software. In the other, your software *cannot* be combined with free software. Considering the chances of these scenarios (reasonably low for a malicious FSF, reasonably high for a ineffective proxy) and the harm that would come from them (an incompatible license could make your software worthless, an overly permissive license merely prevents your software from exerting pro-freedom pressure).
Basically, your choices break down like this:
Now, my argument absolutely does apply to using the New-BSD or Apache license rather than the GPL. The question is simply if there is value to the idea of copyleft - and if you want your code to be usefully copyleft (which applies reusability), the only rational choices are "GPLvX or Later" or maybe the Artistic License 2.0.
Consider very carefully what the actual potential costs and benefits of such a clause are before deciding not to use it.
One of the key advantages to using any version of the GPL is that your code can be combined with other code that was written separately and also released under the GPL. "Version X or later" code can always be combined. When the next version comes out, "Version X only" code will be uncombinable. That basically means that - unless your project is Linux sized and can get away with having its own license - "Version X or later" is the only answer that will allow your project to outlive your personal work on it.
Every GNU+Linux distribution (which includes Ubuntu and Red Hat) already ships a bunch of GPLv3 applications. From the perspective of companies that distribute general purpose operating systems, GPLv3 is strictly better than GPLv2 because of the internationalized wording and the "contributors can't screw the community with patents" provisions.
In a perfect world, maybe.
Back in the real world, enough irrational protest can prevent valuable research from occurring - and leave the related commercial sector scraping by with dysfunctional, archaic, and dangerous technology. For a good example, see nuclear power in the USA.
If those were actually her preferences, she'd vote them. But in your example, those aren't her preferences. Instead, her preferences are these:
Obama: Maybe
Hillary: Yes
Bush: No
In a case like that, either voting Hillary-only or Hillary-and-Obama would be an honest vote. It would only be a dishonest vote if she voted for Obama and not Hillary or Bush and not Obama. Only dishonest votes count as gaming.
So how many megawatts of power generated do you need to fill the balloons for a child's birthday party? How much would Helium have to be worth for it to be an economically viable plan to change the reactor design to make non-radioactive Helium recoverable?
The fact that old computers can be recycled and used by people who might not otherwise have had their own computer *does not* imply that everyone should stop buying new computers. First, if people stopped buying new computers then there would be no used computers to give to your nieces and nephews. Second, if people stopped buying new computers then there would be no R&D budget to develop better hardware - and just because you personally are happy enjoying the results of 50 years of exponential progress in computer hardware doesn't mean that the rest of us should be deprived of the potential results of 60 or 100 years of that process.
You can if you want to.
Most non-trivial programs end up being a mix a lot like that. It's not frankensteinian at all - that's how one normally develops software. You use Python (or Perl, or Ruby, or TCL, or Lua) for non-performance critical sections to save programmer time, and you use C/C++ for the performance critical sections or the sections that involve interfacing to anything that wasn't written in your high-level language of choice.
But that only works for X years after the disaster, where X appears to be a number larger than 7.
Laws aren't about "truth". Every law is a social compromise between freedom and a livable society. In a democracy, which social compromises are made is decided by the people themselves - and their opinions will naturally change over time.
Ethics may not be changeable, but that doesn't mean we have it right yet - and as we continue to work on the problem, ethics will appear to change.
Why wouldn't a recount be possible with optical scan machines?
True enough, but remember: The set of communities with assholes in them and the set of communities that produce useful software tools overlap almost entirely.
Have you heard of Perl?
Now, it's possible that you're defining "Usable" as "Not in Perl", but Perl has more mature, high quality code available in it than any other "web language".
The standard story for dynamic language development applies:
The most common optimization that's used with Rails is it's built-in support for caching, which can speed things up by quite a bit. You can get the same sort of results with a hand-optimized memcached setup any other dynamic language - but Rails gives it to you almost for free.