And completely invaluable in two cases: when a website has something you need but happens to be down, and when the website is up but google is indexing a page that keeps changing, and when I get to the current page what I'm looking for isn't there anymore.
In the first case, you're losing visitors anyway because your site is down. So fix it.
In the second case, it helps me find whether your site indeed has what I'm looking for. I prefer the real website to the cached content, so I'll switch to that as soon as possible.
In both cases, having cached content increases the chances of me coming back or viewing more pages on your site if one of the things that I mentioned happens. If it's down and there's no cache (it can be disabled somehow), I'm not coming back ever.
Where does this idea of the peaceful alien come from? There has never been mutual cooperation between civilizations or species competing for the same resources. Among civilizations, it has always resulted in destruction or subjugation of the less technologically advanced civilization. We need to be keeping our ears open and our mouths shut.
I think that if anything can show up here and say hi they probably don't need anything from us. Unless they come from Proxima Centauri they can probably find whatever they need much closer, and sending anything from here back wherever they came from is probably mind boggingly expensive in energy expenditures.
For instance take the lack of interest in mining asteroids or the moon. We probably could if we had a good reason to, but it's so expensive it's not worth it.
If you're worried about any possible aliens' intentions, then SETI is precisely the right approach. You'd want to know if something is coming our way, and get at least some idea of what it might be like.
It also seems unlikely we can affect our visibility much. On one hand, we're absolutely tiny compared to other things happening in the universe. Any amount of energy we could send into space for instance is a drop in the bucket compared to what the Sun outputs. Anything we emit is unlikely to be received unless somebody is already looking in our direction for some other, more visible reason. But, on the other hand, if somebody is really looking, and capable of getting here, they almost certainly can figure out there's something here, and there's no way we can become quiet enough to pretend there isn't.
At this point we can barely get off this rock. If anything shows up, they almost certainly vastly surpass us just from the fact that they can travel all the way here. So if there's anything to do about that the best plan would seem to be to try to figure out if anybody is coming, and if they are use that information to come up with a plan.
No you don't. You just have to give it to the people that you give binaries to
Exact same thing I said. You have to give the code to anybody you gave the binaries to who asks for it.
Compare, for example, Yahoo! contributing changes to FreeBSD back and Google keeping their internal version of Linux private. The GPL did absolutely nothing to protect Linux. The BSDL did nothing to protect FreeBSD.
Yes, but Google had to necessarily give the source for their changes for Android, while just try and get the source from Apple. Darwin OSS development was pretty much inexistent last time I looked, and what little there is seems to be unusable.
Yahoo! gave code back because they determined it that the cost of maintaining a fork was greater than the competitive advantage gained by keeping the code private. Google kept their filesystem (among other things) private because they made the opposite decision.
Sure. But that means Google can't ever sell that. The moment Google sells something with GFS in it, somebody will be able to ask for the code.
I find this overall to be an acceptable compromise. I can't just walk into Google's office and check out if they're using any of my source there and if they changed anything. Enforcing anything on internal usage would be very difficult, impractical, and involve tactics and methods I disapprove of. The moment Google releases anything to the public however, it doesn't matter whether they want to give it or not, they're legally obligated to give the code.
If you find the GPL to be insufficient, there's the AGPL, which Google seems to really hate as it makes the keeping changes internally much harder.
So resuming what was said so far:
When changes are made only internally, with BSD they get contributed when convenient, and with the GPL as well, with AGPL any user over the network can have them. When changes are released externally, with BSD they get contributed when convenient, with GPL any user can have them, with AGPL any user can have them.
In light of that and my own preferences I therefore prefer the AGPL to the GPL where possible, and heavily prefer GPL to BSD.
So what you are saying is that you object to people stealing your imaginary property? How does them stealing it diminish your enjoyment of it? If 1000 people steal your code, do you have 1000 less code?
Wow, what a way to miss the point.
It's GPL licensed. People are free to use it all they want. If 1000 people use it, all the better, it makes more likely somebody will contribute or want to pay me.
It seems a very fair exchange to me -- you can have my code if I can have your.
For very limited uses only, and it can be quite impractical to keep up.
If it's code you only use internally, then nobody knows of your modifications and the codebase can evolve in ways that make it difficult to keep patching. And the moment you distribute it you have to give the code to anybody you distributed to who asks.
You may be able to play tricks like Tivo if you release a physical product, but that doesn't mean you can omit giving the source, and there's the GPL3 for that case. The AGPL covers the web usage case.
I steer clear of the BSD licenses when contributing.
For precisely the same reason you like it: there are no strings attached. Which means you can have my stuff without having to give anything back. I consider that such an arrangement effectively makes me an unpaid employee of your company, so I won't contribute anything significant under such terms.
I contribute under the GPL/AGPL because in such a case I do get something back: either somebody else's code, or money, if somebody wants a different license.
For the same reason, I preferentially use GPL licensed code. I might want to send a patch some day, but contributing something takes effort. I need to cleanup my code, figure out where to send it, perhaps discuss it on the mailing list with the project, and so on. That's quite a lot of boring work, so I expect to get something out of it.
Wouldn't zeroing the memory on start make things more predictable? If memory persists across reboots you could also have weird bugs due to bad state being kept after a crash, if the voltage unexpectedly drops for instance.
Also, it is my understanding that it's common practice to test after initializing memory with different values (all 0s, all 1s, etc) to see if that makes anything go wrong.
Scientists aren't competent in everything. There are plenty biologists who wouldn't want to go to Mars for research, and I imagine that people with an interest in space aren't that interested in forests either.
Also, if you neglect a field of research people with experience in it disappear, and are hard to replace later. We can't just forget about space for 20 years until the economy improves. If we do that the people who used to build engines, research rockets, investigate how to live in space, etc, will die and move on, and won't get replaced. Then we'll have to start from zero.
Hold that right there. You left of "legitimate, non-malware app".
No, I didn't. The user doesn't know if the application is legitimate, that's the entire point of antivirus-like systems.
Is this stops the user from installing a virus or whatever, that is good. Even if the user THOUGHT that s/he wanted to install it.
You've not dealt with the "normal user" enough, I see. There are few users with the mindset of "This is probably not safe, I'd better not do it" in situations like this. Most have the "damn computer getting into my way when I want to run this cute game I found", at which point it's very likely that the security system gets uninstalled.
Most people will happily override whatever they have to install Bonzi Buddy, because it's cute or something, then complain the antivirus isn't working because all this weird stuff is happening.
See above. You would spend your money with the more responsive vendor. Or you'd go through the hoops.
If you go through the hoops you're not protected anymore. It's like ignoring a certificate error when doing banking online. It might be that the bank has a moron administrating certificates. But that could be a man in the middle as well, in which case you're out of your money. The user, from their computer can't tell which it is.
Why would you need to? If the hashes and signatures and so forth aren't enough to show that that file came from that vendor, oh, wait, they would be.
You're not understanding. A hash certifies that foo.exe is the same file they got from Yoyodyne. Okay. But the AV company can't be completely certain that they're signing the right thing. Maybe somebody broke in and replaced the file Yoyodyne sent, or they downloaded it from the wrong server, or the application from the start had hidden malicious functionality.
At best, a signature certifies "this is the exact file $company ships". It doesn't certify however that the file isn't going to do anything harmful, as that would require examining the code.
For instance, Microsoft has in the past released buggy updates that caused the computer to crash (I think even fail to boot in at least one case). This update of course had Microsoft's signature on it.
That's because the anti-virus vendors don't have the LEGAL RIGHTS to do that.
Eh? That makes no sense. Antiviruses already hook deep enough in Windows that they could enforce pretty much any security model they wanted. I also don't see what would be illegal about releasing some sort of VM that wraps a single application in a sandbox.
Exactly. Which is why that needs to change. Instead of trying to chase the latest variant of a threat, why not save time and effort and identify the LEGITIMATE files? Then, if something is trying to write a file to the OS portion of your drive, and that file is not recognized, it should block it (and MAYBE allow the user to override it after a few hoops and maybe online comparisons with the latest threat databases).
And just how is that going to work?
The main threat are executables. You could require signatures. However, not everything will be signed. Heck, many drivers still aren't. So inevitably the user will run into something unsigned they want to run. At that point they'll ignore/disable the signature warning, and happily install any trojan that comes along.
Or you could reverse the antivirus idea, and build a giant database of checksums. It'll need a checksum for every obscure software out there, in every possible version. WoW released an update today? You can't play until the DB gets updated. At that the user will ignore/disable the signature warning, and happily install any trojan that comes along. Add to that that no company will analyze every byte of every binary, and them listing a trojaned version as valid is quite possible.
Even if that somehow worked perfectly, you still have to deal with exploits, like images crafted to expoloit the decoder. You can't possibly whitelist every legitimate image.
Any signature based system only works well within tight constraints that are impractical on desktop computers. Time would be much better spent on creating sandboxes, tightening permissions and fixing ways to exploit a program, so that if something gets in, it can't do anything anyway. But there's little interest for antivirus vendors in that, as if we got there there wouldn't be improved versions or database updates to sell.
Well, if having child porn is a crime of posession it'd be weird to get charged for failing to find a way to obtain some. It'd be like locking somebody up for having expressed at some point the desire to get some heroin, but failing to find a dealer.
Chromatic aberration and spherical aberration are mostly unimportant if all you need is a key's shape. So long it's not so blurry you can't figure out the shape, you're golden. The lens I linked won't win you any photography prizes, but it is perfectly usable for getting shots of things far away. Your biggest problem would be needing a good tripod to be able to aim it accurately enough.
High res images at a couple of hundred metres (high enough to read the peaks on your house key) between the time it takes you to take your keys from your pocket and put the key in the lock is well into the realm of serious photographic equipment and prowess (insuring your camera and lens for more than your car).
Maybe hundreds of meters is a bit too much, but ridiculous magnification isn't that expensive. $600 for a decent DSLR. $80 for a manual 800mm lens, that an APS sensor will bring to the equivalent of 1200mm.
Granted, that's a horrible lens. It's fully manual, has about 3 pieces of glass in it, is f/11 and gets horrible chromatic aberration just to start with. But if you want to make photos of things that are really far, it works perfectly fine so long you don't care about the quality and have plenty light.
No. I'm saying, somebody has to fight until the end, trying everything they can, for science to advance.
To put an extreme example, if people went "fuck, I'm screwed, might as well die now", popped a cyanide pill and died right there, it would be pretty hard to do any useful research done, if only because of the lack of test subjects.
Those people who spend those huge amounts of money trying to stay alive participate in medical trials, try the newest medication and newest techniques. It is in a big part thanks to them that science advances, some treatments are found to work, and people are able to live a normal life despite having a condition formerly terminal in 3 months. That's why I think all that money isn't wasted.
But if you're doing all that you're carefully generating the CSV, and got a proper parser, and aren't simply doing a 'print "$foo,$bar\n";' into a text file. You're making sure that 0.5 doesn't get written as "0,5" on a french system, and that quotes are doubled, and that any multiline fields are quoted, and deal with backslashes. And your parser has to understand all that of course.
Even if you do all that, the application you interoperate with might not do all of that, or do it in exactly the same way. It could think that the space after the comman in "foo, bar" is junk, while in your application that's part of the field. Even in your post there's already a problem: A newline might either be a \n or a literal newline. Those details are problematic.
At that point you might as well use a XML library instead.
If population growth is the only thing that improves, I might agree with you. Is that really the case though? What about measuring the amount of sick people that need treatment?
I'm not talking about the population, but about the amount of people to feed.
If you have to feed 10 million hungry people today, and 12 million next year, there is a problem. That can be either that the population grew, or that people are overall worse off and in more need of help. In the first case, the increased population is still failing to produce enough food, in the second, the food isn't really helpful. In any case something is wrong that isn't getting fixed by just sending free food to the population.
You act as if people is merely a burden rather than a resource.
If those people were a resource then next year there would be less people to feed, as those extra people would be growing food, for instance.
Only if all the following conditions exist: There actually IS a clothing industry in whatever place we're talking about.
Making clothes is easy. I'm not talking about designer clothes here. Some cloth, thread and a sewing machine can go a long way. Your grandma didn't need a room full of machinery to knit a sweater.
The clothing is actually donated rather than someone just buying clothing locally.
Lots of it is. I've seen many notices in many places along the lines of "Leave your old clothes here, we'll pick them up and send where needed". Apparently more than enough to screw up the industry as per the above article.
The people who can't afford clothing would have somehow went out and bought it rather than continue to wear rags.
Yes, but that's not a sustainable economy. Short term, somebody gets old designer jeans from the US. Long term, that means they don't need to buy locally, so the local maker goes out of business. In turn, that local merchant now isn't earning money to spend in the local economy. If they try to grow food, well, you dump truckloads of that for free in there as well.
Infra-structure doesn't help so much when everyone is dying of some easily preventable disease. Infra-structure is important, but you have to start with getting people relatively healthy first. Sick people ain't so productive, and are obviously very expensive. Is it really smart to build a factory, educate the workers, train them.. and then have a large portion of the people you just spent a lot of money on dying of malaria, or some other cheap to prevent disease? A healthy society is the base of productivity. In a very real sense it IS infra-structure.
I agree, but why is it that year after year people keep talking of that for $X dollars we could feed/clothe/vaccinate $Y millions in the third world? If we have to feed $Y millions this year, and 1.1 * $Y millions the next, and so on, we're not making any progress. We're even actually making things worse. For instance, the massive amount of donated clothes destroys the clothing industry in those places. Instead of helping what we do is screwing up their situation even more, and making them dependent on donations.
The "feed $Y millions" stuff should have ended decades ago, except for mass disasters. Feed people once, then make sure they can keep feeding themselves without further intervention.
CSV and tab delimited files are trouble though. For CSV you run into it as soon as you need to allow commas in a field. For tab delimited files it's tabs. For both you have issues with newlines, encodings, optional columns, additional columns, and I probably forgot something.
The nice thing about XML is that you don't have to care about any of that.
If you think "But, I will never have tabs/commas/quotes/newlines/foreign languages in my data!" then unless you have a very tightly defined format you're probably wrong. Been there, done that. Spent a few days debugging a fixed column format generating application that would sometimes generate a column with a character too many, shifting everything afterwards. After that I say screw it, just use XML and you'll never end up counting characters in a text editor.
In programming the best version is generally the third.
The first is the "I don't know what I'm doing" version, which gets written by trying without much thought, ugly hacks, and without a decent design. Sometimes it does work quite well however, as despite not being very pretty it does what it's supposed to.
The second is "V1 is crap, but now that I have figured it all out I can do better!". Often a horrible mess, due to things like wanting to make everything modular, adding every feature possible, and using the latest cool tech and design patterns where they don't belong. Turns out to be slow, huge, buggy and overly complicated to use. There's even a name for this: "second system effect".
Based on the lessons learned from the lack of planning in the first and the excesses of the second, the third version has a good chance of being actually decent.
Of course not all of it. Not even a large part of it probably.
But, if when people got cancer they decided to just take pain relief medication and die quietly, nobody would do any research. Somebody has to fight until the bitter end. And those people are precisely those who try the experimental medicine, in case it helps. Even if research was entirely funded by donations, somebody still has to volunteer to try it.
The first heart transplants were a desperate gamble, but now thanks to all the attempts many people can actually enjoy a normal life afterwards, and some even manage to climb mountains.
Sure, but if you travel/stay in space long enough it might weight less be worth it to use a single set of rechargeable ones instead of several of lithiums. On a trip you could use a small solar panel, and on the ISS there probably is a battery charger.
Though most DSLRs can work in tethered mode, so you can avoid needing batteries at all that way.
True, it's horrendously expensive. But that money pays for research, and that research will allow people to stay alive or even get cured a lot cheaper later.
People aren't going to pay half a million for battling cancer forever. At some point it'll be understood and become curable with a few of the right pills and injections. But for that to happen, somebody needs to try the less understood or experimental treatments and see if it works out.
Incidentally, I believe that paying for the "vaccine for nearly a quarter million children in developing countries" is on the long term a rather pointless thing. Doing it that way we'll just be shipping vaccines over there forever. Instead, money should be invested on infrastructure in those countries that need it, so that they can manufacture their own vaccines. Also, actually allowing those countries to manufacture them by eliminating the need to obey the patents would do a whole lot more of good.
Slashdot Mirror
And completely invaluable in two cases: when a website has something you need but happens to be down, and when the website is up but google is indexing a page that keeps changing, and when I get to the current page what I'm looking for isn't there anymore.
In the first case, you're losing visitors anyway because your site is down. So fix it.
In the second case, it helps me find whether your site indeed has what I'm looking for. I prefer the real website to the cached content, so I'll switch to that as soon as possible.
In both cases, having cached content increases the chances of me coming back or viewing more pages on your site if one of the things that I mentioned happens. If it's down and there's no cache (it can be disabled somehow), I'm not coming back ever.
I think that if anything can show up here and say hi they probably don't need anything from us. Unless they come from Proxima Centauri they can probably find whatever they need much closer, and sending anything from here back wherever they came from is probably mind boggingly expensive in energy expenditures.
For instance take the lack of interest in mining asteroids or the moon. We probably could if we had a good reason to, but it's so expensive it's not worth it.
SETI is a detector, not an emitter.
If you're worried about any possible aliens' intentions, then SETI is precisely the right approach. You'd want to know if something is coming our way, and get at least some idea of what it might be like.
It also seems unlikely we can affect our visibility much. On one hand, we're absolutely tiny compared to other things happening in the universe. Any amount of energy we could send into space for instance is a drop in the bucket compared to what the Sun outputs. Anything we emit is unlikely to be received unless somebody is already looking in our direction for some other, more visible reason. But, on the other hand, if somebody is really looking, and capable of getting here, they almost certainly can figure out there's something here, and there's no way we can become quiet enough to pretend there isn't.
At this point we can barely get off this rock. If anything shows up, they almost certainly vastly surpass us just from the fact that they can travel all the way here. So if there's anything to do about that the best plan would seem to be to try to figure out if anybody is coming, and if they are use that information to come up with a plan.
Exact same thing I said. You have to give the code to anybody you gave the binaries to who asks for it.
Yes, but Google had to necessarily give the source for their changes for Android, while just try and get the source from Apple. Darwin OSS development was pretty much inexistent last time I looked, and what little there is seems to be unusable.
Sure. But that means Google can't ever sell that. The moment Google sells something with GFS in it, somebody will be able to ask for the code.
I find this overall to be an acceptable compromise. I can't just walk into Google's office and check out if they're using any of my source there and if they changed anything. Enforcing anything on internal usage would be very difficult, impractical, and involve tactics and methods I disapprove of. The moment Google releases anything to the public however, it doesn't matter whether they want to give it or not, they're legally obligated to give the code.
If you find the GPL to be insufficient, there's the AGPL, which Google seems to really hate as it makes the keeping changes internally much harder.
So resuming what was said so far:
When changes are made only internally, with BSD they get contributed when convenient, and with the GPL as well, with AGPL any user over the network can have them.
When changes are released externally, with BSD they get contributed when convenient, with GPL any user can have them, with AGPL any user can have them.
In light of that and my own preferences I therefore prefer the AGPL to the GPL where possible, and heavily prefer GPL to BSD.
Wow, what a way to miss the point.
It's GPL licensed. People are free to use it all they want. If 1000 people use it, all the better, it makes more likely somebody will contribute or want to pay me.
It seems a very fair exchange to me -- you can have my code if I can have your.
Oh, misunderstood what you said.
Use without modification is something that's mostly neutral from my point of view. My interest is in development.
For very limited uses only, and it can be quite impractical to keep up.
If it's code you only use internally, then nobody knows of your modifications and the codebase can evolve in ways that make it difficult to keep patching. And the moment you distribute it you have to give the code to anybody you distributed to who asks.
You may be able to play tricks like Tivo if you release a physical product, but that doesn't mean you can omit giving the source, and there's the GPL3 for that case. The AGPL covers the web usage case.
I steer clear of the BSD licenses when contributing.
For precisely the same reason you like it: there are no strings attached. Which means you can have my stuff without having to give anything back. I consider that such an arrangement effectively makes me an unpaid employee of your company, so I won't contribute anything significant under such terms.
I contribute under the GPL/AGPL because in such a case I do get something back: either somebody else's code, or money, if somebody wants a different license.
For the same reason, I preferentially use GPL licensed code. I might want to send a patch some day, but contributing something takes effort. I need to cleanup my code, figure out where to send it, perhaps discuss it on the mailing list with the project, and so on. That's quite a lot of boring work, so I expect to get something out of it.
Wouldn't zeroing the memory on start make things more predictable? If memory persists across reboots you could also have weird bugs due to bad state being kept after a crash, if the voltage unexpectedly drops for instance.
Also, it is my understanding that it's common practice to test after initializing memory with different values (all 0s, all 1s, etc) to see if that makes anything go wrong.
Scientists aren't competent in everything. There are plenty biologists who wouldn't want to go to Mars for research, and I imagine that people with an interest in space aren't that interested in forests either.
Also, if you neglect a field of research people with experience in it disappear, and are hard to replace later. We can't just forget about space for 20 years until the economy improves. If we do that the people who used to build engines, research rockets, investigate how to live in space, etc, will die and move on, and won't get replaced. Then we'll have to start from zero.
No, I didn't. The user doesn't know if the application is legitimate, that's the entire point of antivirus-like systems.
You've not dealt with the "normal user" enough, I see. There are few users with the mindset of "This is probably not safe, I'd better not do it" in situations like this. Most have the "damn computer getting into my way when I want to run this cute game I found", at which point it's very likely that the security system gets uninstalled.
Most people will happily override whatever they have to install Bonzi Buddy, because it's cute or something, then complain the antivirus isn't working because all this weird stuff is happening.
If you go through the hoops you're not protected anymore. It's like ignoring a certificate error when doing banking online. It might be that the bank has a moron administrating certificates. But that could be a man in the middle as well, in which case you're out of your money. The user, from their computer can't tell which it is.
You're not understanding. A hash certifies that foo.exe is the same file they got from Yoyodyne. Okay. But the AV company can't be completely certain that they're signing the right thing. Maybe somebody broke in and replaced the file Yoyodyne sent, or they downloaded it from the wrong server, or the application from the start had hidden malicious functionality.
At best, a signature certifies "this is the exact file $company ships". It doesn't certify however that the file isn't going to do anything harmful, as that would require examining the code.
For instance, Microsoft has in the past released buggy updates that caused the computer to crash (I think even fail to boot in at least one case). This update of course had Microsoft's signature on it.
Eh? That makes no sense. Antiviruses already hook deep enough in Windows that they could enforce pretty much any security model they wanted. I also don't see what would be illegal about releasing some sort of VM that wraps a single application in a sandbox.
And just how is that going to work?
The main threat are executables. You could require signatures. However, not everything will be signed. Heck, many drivers still aren't. So inevitably the user will run into something unsigned they want to run. At that point they'll ignore/disable the signature warning, and happily install any trojan that comes along.
Or you could reverse the antivirus idea, and build a giant database of checksums. It'll need a checksum for every obscure software out there, in every possible version. WoW released an update today? You can't play until the DB gets updated. At that the user will ignore/disable the signature warning, and happily install any trojan that comes along. Add to that that no company will analyze every byte of every binary, and them listing a trojaned version as valid is quite possible.
Even if that somehow worked perfectly, you still have to deal with exploits, like images crafted to expoloit the decoder. You can't possibly whitelist every legitimate image.
Any signature based system only works well within tight constraints that are impractical on desktop computers. Time would be much better spent on creating sandboxes, tightening permissions and fixing ways to exploit a program, so that if something gets in, it can't do anything anyway. But there's little interest for antivirus vendors in that, as if we got there there wouldn't be improved versions or database updates to sell.
Well, if having child porn is a crime of posession it'd be weird to get charged for failing to find a way to obtain some. It'd be like locking somebody up for having expressed at some point the desire to get some heroin, but failing to find a dealer.
There are different kinds of quality.
Chromatic aberration and spherical aberration are mostly unimportant if all you need is a key's shape. So long it's not so blurry you can't figure out the shape, you're golden. The lens I linked won't win you any photography prizes, but it is perfectly usable for getting shots of things far away. Your biggest problem would be needing a good tripod to be able to aim it accurately enough.
Maybe hundreds of meters is a bit too much, but ridiculous magnification isn't that expensive. $600 for a decent DSLR. $80 for a manual 800mm lens, that an APS sensor will bring to the equivalent of 1200mm.
Granted, that's a horrible lens. It's fully manual, has about 3 pieces of glass in it, is f/11 and gets horrible chromatic aberration just to start with. But if you want to make photos of things that are really far, it works perfectly fine so long you don't care about the quality and have plenty light.
Well, I think the experimental drugs are given out for free. But you don't get to that point without trying the current non-free approaches first.
No. I'm saying, somebody has to fight until the end, trying everything they can, for science to advance.
To put an extreme example, if people went "fuck, I'm screwed, might as well die now", popped a cyanide pill and died right there, it would be pretty hard to do any useful research done, if only because of the lack of test subjects.
Those people who spend those huge amounts of money trying to stay alive participate in medical trials, try the newest medication and newest techniques. It is in a big part thanks to them that science advances, some treatments are found to work, and people are able to live a normal life despite having a condition formerly terminal in 3 months. That's why I think all that money isn't wasted.
Sure, it can be done.
But if you're doing all that you're carefully generating the CSV, and got a proper parser, and aren't simply doing a 'print "$foo,$bar\n";' into a text file. You're making sure that 0.5 doesn't get written as "0,5" on a french system, and that quotes are doubled, and that any multiline fields are quoted, and deal with backslashes. And your parser has to understand all that of course.
Even if you do all that, the application you interoperate with might not do all of that, or do it in exactly the same way. It could think that the space after the comman in "foo, bar" is junk, while in your application that's part of the field. Even in your post there's already a problem: A newline might either be a \n or a literal newline. Those details are problematic.
At that point you might as well use a XML library instead.
I'm not talking about the population, but about the amount of people to feed.
If you have to feed 10 million hungry people today, and 12 million next year, there is a problem. That can be either that the population grew, or that people are overall worse off and in more need of help. In the first case, the increased population is still failing to produce enough food, in the second, the food isn't really helpful. In any case something is wrong that isn't getting fixed by just sending free food to the population.
If those people were a resource then next year there would be less people to feed, as those extra people would be growing food, for instance.
Making clothes is easy. I'm not talking about designer clothes here. Some cloth, thread and a sewing machine can go a long way. Your grandma didn't need a room full of machinery to knit a sweater.
But, even that goes nowhere when you ship truckloads of your old stuff and dump it on those people for free. What little industry was there dies, then they start needing those donations because you just drove the producers out of business.
Lots of it is. I've seen many notices in many places along the lines of "Leave your old clothes here, we'll pick them up and send where needed". Apparently more than enough to screw up the industry as per the above article.
Yes, but that's not a sustainable economy. Short term, somebody gets old designer jeans from the US. Long term, that means they don't need to buy locally, so the local maker goes out of business. In turn, that local merchant now isn't earning money to spend in the local economy. If they try to grow food, well, you dump truckloads of that for free in there as well.
I agree, but why is it that year after year people keep talking of that for $X dollars we could feed/clothe/vaccinate $Y millions in the third world? If we have to feed $Y millions this year, and 1.1 * $Y millions the next, and so on, we're not making any progress. We're even actually making things worse. For instance, the massive amount of donated clothes destroys the clothing industry in those places. Instead of helping what we do is screwing up their situation even more, and making them dependent on donations.
The "feed $Y millions" stuff should have ended decades ago, except for mass disasters. Feed people once, then make sure they can keep feeding themselves without further intervention.
CSV and tab delimited files are trouble though. For CSV you run into it as soon as you need to allow commas in a field. For tab delimited files it's tabs. For both you have issues with newlines, encodings, optional columns, additional columns, and I probably forgot something.
The nice thing about XML is that you don't have to care about any of that.
If you think "But, I will never have tabs/commas/quotes/newlines/foreign languages in my data!" then unless you have a very tightly defined format you're probably wrong. Been there, done that. Spent a few days debugging a fixed column format generating application that would sometimes generate a column with a character too many, shifting everything afterwards. After that I say screw it, just use XML and you'll never end up counting characters in a text editor.
In programming the best version is generally the third.
The first is the "I don't know what I'm doing" version, which gets written by trying without much thought, ugly hacks, and without a decent design. Sometimes it does work quite well however, as despite not being very pretty it does what it's supposed to.
The second is "V1 is crap, but now that I have figured it all out I can do better!". Often a horrible mess, due to things like wanting to make everything modular, adding every feature possible, and using the latest cool tech and design patterns where they don't belong. Turns out to be slow, huge, buggy and overly complicated to use. There's even a name for this: "second system effect".
Based on the lessons learned from the lack of planning in the first and the excesses of the second, the third version has a good chance of being actually decent.
Of course not all of it. Not even a large part of it probably.
But, if when people got cancer they decided to just take pain relief medication and die quietly, nobody would do any research. Somebody has to fight until the bitter end. And those people are precisely those who try the experimental medicine, in case it helps. Even if research was entirely funded by donations, somebody still has to volunteer to try it.
The first heart transplants were a desperate gamble, but now thanks to all the attempts many people can actually enjoy a normal life afterwards, and some even manage to climb mountains.
Sure, but if you travel/stay in space long enough it might weight less be worth it to use a single set of rechargeable ones instead of several of lithiums. On a trip you could use a small solar panel, and on the ISS there probably is a battery charger.
Though most DSLRs can work in tethered mode, so you can avoid needing batteries at all that way.
True, it's horrendously expensive. But that money pays for research, and that research will allow people to stay alive or even get cured a lot cheaper later.
People aren't going to pay half a million for battling cancer forever. At some point it'll be understood and become curable with a few of the right pills and injections. But for that to happen, somebody needs to try the less understood or experimental treatments and see if it works out.
Incidentally, I believe that paying for the "vaccine for nearly a quarter million children in developing countries" is on the long term a rather pointless thing. Doing it that way we'll just be shipping vaccines over there forever. Instead, money should be invested on infrastructure in those countries that need it, so that they can manufacture their own vaccines. Also, actually allowing those countries to manufacture them by eliminating the need to obey the patents would do a whole lot more of good.