Maybe this is a good argument to keep it all on the server (although until recently my corporate email kept quotas and prevented you doing that).
Which is exactly what I do...:)
$ cd ~/.maildir ; find . | wc -l 22454 $ du -s . 715632 .
Just run courier-imap and I can access my mail from thunderbird, a webmail client, or whatever the email-program-de-jure is. File corruptions? Forget it - they're just text files. And backing up 715MB of text files is completely trivial - probably compresses to 50MB. I backup everything in/home that isn't trivially replacable (the latter goes in a designated directory tree and is excluded).
You don't even need to run your own domain to do all this - you can set this up using fetchmail to pull your mail out of gmail/ISP/yahoo/hotmail, or whatever else freepopsd supports.
There is no reason to delete email in this day and age unless you are a business covering your tracks (err, complying with a record retention policy). If somebody had to take the time to type it, then it can't possibly cost more than a microcent to store.
I'm not intimately familiar with Fedora, but many distros backport security patches. So, even though you might be two versions behind bleeding-edge you may in fact be completely secure. Backported patches have the advantage of being less likely to break things.
Re:Flash as an Application Development Platform? N
on
The Future of Flash
·
· Score: 1
The only reason that YouTube, Google Video et al adopted Flash as their video player client was because Flash is pretty much universal
As long as you're not running 64-bit... I end up running vmware or firefox in a chroot whenever I need to access a page that runs flash. Both are a major pain. You can't even fire off konqueror in a chroot as it tries to talk to the 64-bit versions that are already running and gets confused.
This is the problem with proprietary protocols - you are supported only if the vendor feels it meets their business plans.
What is wrong with just putting an avi file using an open codec on your website? Why does it HAVE to be flash? And if you need vector graphics what is wrong with SVG?
Does it really NEED to cost millions of dollars or is this our government's way to protect business models?
Well, if you got rid of the requirement to test pharmaceuticals before marketing them then the costs would fall DRASTICALLY. On the other hand you'd have a million brands of white pills and no real way to figure out which ones work and what side-effects they might have. The government would undoubtedly try to do trials, but it would be impossible to keep up (there would be many times as many drugs on the market as there are now - imagine doing a comparative study of 500 different cholesterol-lowering drugs). On the other hand, the instant that a particular drug is shown to work better than the others then everybody would copy it and the company which came up with it wouldn't get any particular reward. As a result, companies would not have any incentive to make drugs that actually work, just drugs that plausibly sound like they could work.
If there were a way to generate the existing required safety and efficacy data more efficiently somebody would be doing it. While people generally only hear a couple of big names when it comes to major pharmaceuticals, the fact is that outsource suppliers of almost all phases of development are available, from animal testing to formulation and manufacture and QA testing to clinical trials. These are mostly small companies and they compete vigorously for contracts with the big pharma outfits (who often use them to offload work they can't handle at the moment). If one of these companies found a more efficient way to do any of these development phases they would get a ton of business and all pharma development costs would drop as a result. Big pharma is in a bit of a cost crisis right now - new drugs aren't being developed fast enough to fuel their balance sheets and most companies are looking for ways to cut costs as price increases are not well-tolerated by the public and are made difficult by insurers.
Agree on the 3100cn. I got sick and tired of cleaning the inkjet and having the family have to fuss with it (and inevitably run into issues while I was gone). The 3100cn is a bit large, but it has great color output, is very fast, and is very economical (1.5 cents bw, 4.5 cents color per page). I figure that it will pay for itself in no time since the toner cartridges don't dry up and cost a lot less per page. Color wasn't an absolute necessity (I get my photos done at Walmart), but at $300 shipped on sale how could I not order one?
Re:A GNOME user converts.
on
KDE 3.5.4 Released
·
· Score: 2, Informative
Beagle, Fast User Switching, apps like Rhythmbox and Evince, GTK/Cairo with good Python bindings -- these all provide real value to me on a daily basis.
FYI - Fast user switching is in fact available in KDE. It just launches a new kdm session on a new X display and optionally locks the current one. Works just fine for me.
Most of the other stuff are just applications - I use evince all the time on KDE. I don't care what set of widgets an app uses so much as how the overall window-manager and integrated apps like a browser work. The rest you can really just swap out all you want.
OK, so we slingshot it around a few times. Orbital mechanics may not be well understood by me, but the folks at NASA seem to have it down pretty good.
I have every confidence that the NASA guys could get a payload to the sun. It actually only takes high school physics (well, it also takes state-of-the-art engineering).
The problem is cost. You have to generate a delta-V of approximately the Earth's orbital velocity - that is 30 km/s. The parent post suggested that the Sun would do most of the work. This is incorrect - the sun will only do work once the craft has no solar orbital velocity - then it will just fall straight down into the sun. The real work is getting rid of 30 km/s of orbital velocity - if you don't do that the waste will just orbit the sun very close to the earth, and sooner or later it will come back (even if it had escape velocity - that just gets it out of earth's orbit - if it swings around the sun a few times and comes back at us it will still re-enter earth orbit).
The problem is trivial to surmount - you just need a really big rocket. But then again, keeping the waste on earth just needs a really big hole in the ground. The only real decision is which engineering project is more expensive or risky - and most likely it will turn out to be the hole in the ground.
Slingshotting it around a few times is not really a great solution - you still need a ton of energy to get to anything to slingshot off in the first place. The other problem is launch windows - if you want to do multiple slingshots then you have to be really patient for a window. The craft will also need a lot of course corrections - if you're going to launch thousands of waste containers that is a lot of manpower to keep them all on course (unless you just want to drop them on Venus - but even that needs accuracy if you don't want to risk slingshotting it back into solar orbit near the earth). So, maybe with some fancy slingshotting you might only need 15-20 km/s of delta-V - that is still a lot of impulse. The Saturn-V was good for about 7.5GNs - so that is good for about 300 metric tons of payload if you only need 15 km/s (plus escape velocity). Well, that is actually a major overestimate - it doesn't factor in the weight of the rocket itself (which is an exercise in calculus which I'm not bored enough to do). It certainly can be done, but you're talking about a lot of HUGE rockets.
All in all, the hole in the ground is probably the best bet.
but if they believe your health exists entirely in a vacuum without affecting anybody else, they have no understanding of how social costs are... well... social.
Yes, but in the real world people will tend to act in the intrests of self-advancement.
For example, most people would consider me to have above-average abilities, and as a result I can be much more productive than many others when I work. Wouldn't it be in the interests of society to compel me to work a much longer day than others who are less productive? After all, many poorly educated workers produce less useful output in a day of hard work than some well-educated workers might produce in an hour - so why bother having those folks work at all? Why not have them man the whips and guns and they can see to it that the more educated members of society so that the overall level of output is higher?
The problem is that in the real world this doesn't work - the educated will simply find a way to escape and set up a more capitalistic society where the folks reaping the most benefits of work are those doing the most work.
Obviously this is an imperfect analogy, and is not meant to describe any kind of functional society, and I would acknowledge that some forms of manual labor can be very high in value-output and that some forms of educated labor may not. The point is just that when I work I do so to benefit myself and my family first, and others as I am able to voluntarily pitch in. If somebody put a gun to my head to get me to work harder in the interests of the better good I would quickly figure out how to game the system and get to be the person holding the gun and not the person it is pointed at (and this is why all the brains in a communistic society aim for party leadership and not doing real work).
None of this is to say that there can't be a balance, but ultimately if somebody tells me that I can't do something because it lowers my health and therefore my labor value to society I'm going to resent society for treating me merely as a resource to be exploited. It would be one thing if I were on the dole - but if I'm paying my taxes then my debt to society is paid in full - it is not appropriate for the government to act as a taskmaster and determine if I'm working as hard as I could...
Of course it is - but that is what Linus gets for not keeping his licensing flexible. The FSF requires assignment of copyright and the "or a later version" wording precisely so that the license can be adapted as needed.
The FSF shouldn't have to set themselves in stone because a few big third parties became dependant on their code without taking proper legal precautions...
When you buy commercial software (well most of it) - you don't buy to own it, you buy to own a license.
Uh, care to cite the appropriate law here? And I don't mean the EULA (an after-the-fact contract that lacks consideration).
If you buy a book you OWN it. You don't have the right to copy and distribute it, but you do in fact own it. You can resell it.
Ditto for software - if it looks like a duck and walks like a duck you can't put a piece of paper in the box that says it is really a giraffe.
Sure, you won't find any mainstream software vendor who supports this view, but I'm sure if you talk to folks at the EFF or FSF (who are the ones writing the GPL that is the topic of this discussion) they would generally hold to the idea that software is sold. I think you'll find very little law in support of any other concept - just a lot of big-software-company press releases.
This would only be a problem if some newer version of glibc under GPLv3 comes up with some super-fancy implementation of memcpy and the kernel wants to use it. In that case they could not copy the code (though they could clean-room reimplement it).
They could continue to use the old GPLv2 version as long as they wanted to...
Modifying an autopilot by smashing it with a big rock prior to installation might cause a disaster. What safeguards are in place to prevent this? Sure, they might put 1" armor on it, but I'm sure that some high explosive will still damage the device.
Ultimately the users of safety-critical hardware are the ones most responsible for using it correctly. Their first step should be to screen their vendors for systems that are adequately tested. However, if they feel that they can modify the hardware to make it more suitable they should be able to do so (although they might have to apply for various government certifications in some industries to use the resulting new product). The user has every incentive to exercise due care with the equipment.
Airlines spend tons of money on maintenance to make sure their planes don't just fall out of the sky. Why would anybody think that they would just start flashing their firmware willy-nilly simply because the hardware allows it. If they did develop their own firmware they would almost certainly test the living daylights out of it and install it under thorough controls. The FAA would be able to tell what firmware was running just by inspecting the aircraft's maintenance logs.
Safety-critical industries are generally very aware of the impact of the software they use, and since the costs of testing their software right is less than the cost of a planefull of wrongful death lawsuits they're going to be careful about flashing their EPROMS - regardless of vendor support.
What will happen if the FSF gets its way is that manufactures will simply not use OSS code at all. Not only will the user not be able to modify the code, there will also be no open code to look at. The FSF will lose and users will be back where they were before there ever was a GPL.
How is this a loss for the FSF? Manufacturers might not use GPL software, but this will come at a cost that will need to be passed onto consumers. Other manufacturers might continue to use GPL software with the added strings, and they might have a lower cost basis, and therefore a marketplace advantage.
All those manufacturers use GPL software right now because it works well and it is free-as-in-beer. Those two incentives will still remain, and having your product 80% done before you start on it with no licensing costs is an advantage that will not lightly be dismissed.
If you don't like DRM then don't buy or use devices that implement it.
Well, until there are no devices that do not implement it. In fact, the new GPL will help raise the costs of DRM hardware (due to the need to potentially license another OS and/or deal with a less mature product than GPL software) - which will help slow its adoption and make those devices that don't implement it a little more accessible.
In any case - there is nothing to fight over. If you don't like the new GPL don't use it - just don't expect much help from those who do use it or the ability to leverage their software.
It is implied that you can twaek the source code. Not the way how the hardware works. OpenSource Software license should not force how the HARDWARE should work. I think it needs separate license/certification for hardware. Like OpenHardware. This could solve the problem - then you will known - this hardware is OK, it is compatible, we like it. That other hardware is piece of smelly crap and we don't like it. Simple.
Great - so one day consumers might have a choice between a 15GHz Intel PC for $400, or a 1GHz OpenPC for $1500. Some choice - nobody will buy open hardware unless it is mainstream, and it will never be mainstream unless the average person can see the benefits - most average people don't change their OS.
If I write a software package I would prefer that it not be usable by the customers of closed hardware. This means that the vendors of closed hardware will need to reinvent my software package or open their hardware. This makes the open hardware more competitive (either by making more mainstream hardware open, or raising the costs of closed hardware). If I spend my time writing software it isn't so that some proprietary vendor can use it in their effort to lock down their hardware.
"Trusted computing" is simple an implementation of hardware that can be controlled to execute software or not based on the fact who you trust.
Define "you". If I buy a computer and I am provided all keys embedded in the hardware (and their associated related keys) then it functions as you describe - I can sign software and the computer will trust it, or I can sign data and a remote computer that trusts my computer will trust the data.
However, this is not the model envisioned in Trusted Computing. In general the keys and associated related keys will not be given to computer owners, but will be maintained by a 3rd party. Therefore, it isn't about who "you" trust, but rather who "they" trust. The "they" in this case might be a media distributor who doesn't trust "you" but rather wants to communicate directly with your computer hardware, which they trust only because you don't have control of it.
If my computer had a chip in it that would track all software running on it and certify this list using a keypair that I provided it with, and if the chip would only allow software I signed to be run, then I would consider this an EXCELLENT feature. It gives me more control over my own hardware and makes rootkit detection foolproof and remote software installation impossible (without my authorization). The key is whoever knows the keys controls the whole scheme. I submit that the keys should be owned by whoever pays for the hardware.
Disclaimer - I'm not the grandparent poster so I can't say why he would want to use it. But, here are a couple of reasons why developers might want to use it:
It reduces the amount of free-as-in-beer software available to manufacturers of locked-down / DRM hardware - thus it increases the cost of such hardware and makes open hardware more competitive. Just think how much more expensive a Tivo-like appliance might be if they didn't get to use linux - they might consider allowing their hardware to run unsigned binaries as a result rather than licensing an alternative OS.
It ensures that if somebody makes public commercial use of my code and generates a lot of improvements that I will be able to incorporate those improvements into my code. Right now they could just run it on a webserver and not distribute their source.
Keep in mind that the time free software developers is valuable - many would like to receive compensation for it. If a private user with shallow pockets wants to use their software, well then fine. If a big megacorp is going to make a mint using the software they should at least give a little something back to the original developer - either their source code or a negotiated non-GPL license fee. Think MySQL - it would never work without the GPL (that is - their business model).
Nobody is forced to accept the GPL - they just need to reinvent the wheel (or at least the parts of the wheel that aren't available under BSD/etc).
Re:Positively fantastic news
on
Growing Insulin
·
· Score: 1
Well, I guess it is better than no insulin at all (even if they are not nearly as safe as recombinant insulin). Just crush up pancreas and run it through a column (or whatever they do).
Kind of like hospitals donating expired drugs to 3rd-world countries. There are definite safety and efficacy problems with this, but they're still helping more people than they are harming.
Re:Positively fantastic news
on
Growing Insulin
·
· Score: 1
If there was a conspiracy to raise prices, then some company that wasn't in on it would apply to make generic insulin the old way and charge the old prices for it. While making generic drugs isn't super-cheap, there aren't many artificial barriers to entry.
So many companies currently make insulin that it would be unlikely that they could form a trust. All one company needs to do is undercut the rest.
In the real world. price fixing is very hard to maintain, and it is usually caught when it happens (eventually).
When was the last time that a new techonlogy RAISED the cost of generic medications? Non-patented medications are very inexpensive, are made in numerous countries (and are allowed to be imported into the US), and the US has some of the cheapest prices for these in the first world.
Patented drugs are a different story, but even these only can support lofty prices if they are substantially better than generic medicines. Just watch Pfizer's earnings with Lipitor once generic simvastatin floods the market. Lipitor is shown to be better, but for the minor gains many insurers will opt for cheaper prices, and that will drive down all the costs.
Re:Positively fantastic news
on
Growing Insulin
·
· Score: 1
No, if safflower insulin wants to sell for ten times the price of other brands of insulin, all they have to do is convince the doctors, not the patients,
All insurance companies have preferred formularies, and they WILL substitute drugs that are bio-equivalent. The doctor prescribes insulin - not a particular brand. If they do prescribe an insulin-derivative there is usually some reason for it. This company is making plain old insulin - nothing that couldn't be substituted.
though they could get bonus points for showing that it's safer than pig insulin or the current engineered bacteria that produce it
Well, nobody uses pig insulin anymore I think. It would certainly be safer than pig insulin, as would any other route of manufacturing recombinant human insulin. If it is safer for some reason than other sources of human insulin then perhaps they should be allowed to make a profit for their work in making life safer for diabetics.
Most likely this technology will just be used to make insulin cheaper to manufacture, and all prices will go down as a result. Not everything is a big conspiracy - especially where generic drugs are concerned. Non-branded drugs are very inexpensive, and are often cheaper in the US than in other nations which do not use market pricing. Now, branded drugs are a different story - but nobody has to use those - they could just use 10-year old cheap technology (with the associated reduced benefits).
All the moaning about expensive drugs is like moaning about expensive plasma TVs. I own a cheap CRT set and I don't feel like anybody owes it to me to sell a 50" plasma for $300. There are lots of cheap drugs out there, and like anything else you get what you're willing to pay for. If it weren't for branded drugs most likely you would be taking the cheap drugs anyway, as that would be all that was available.
Vmware runs fine on linux, and overhead isn't too bad (but keep in mind that if you want really good performance you'll need enough RAM to run the combined total of both OS's). However, not really ideal for gaming, as vmware does not support Direct3D/OpenGL well (I think there is some experimental support for it, but I can't vouch for how stable it is). Note that I've gotten my share of blue-screens using vmware (granted, this is on amd64 which is less than rock-solid for most software), and even the odd kernel panic (it does run at least partially in kernel space with a few modules - not sure why it would have to do that).
Still, I find it great for running those one-off windows-only applications without needing to dual-boot (which of course takes my server offline).
I doubt many SPARC users will care, but I'm sure many of us amd64 users wouldn't mind having our platform supported as something other than an afterthought.
If you're going to read and write files or network streams, you shouldn't be using data types like int/etc. Use the defined-size types, and then you don't have to worry about portability! It is just something that needs to be kept in mind, and it doesn't have to be a hardship. Plus it saves a shower of patching after the fact...
You also can't hide from a different installation of Windows that has the infected disk mounted. Rootkits hide themselves by hooking into the running kernel/fs drivers - inspect the disk with a clean install and they can't hide then either.
Completely true, but it is a lot more convenient to scan with a linux boot-CD than to keep two windows partitions on the same computer and take turns scanning one from the other. I don't believe there are any windows boot-CD solutions, and I believe that the reason is licensing (and of course lack of interest on the part of MS).
If this driver matures you can bet that it will become the technique of choice for supporting NTFS on a boot disk - you'll see companies like AV and disk imaging vendors jumping on it.
The principle being that if they want, they can stop providing the service and the contract ends at that point, if you have already paid for a period after that time then your going to have trouble getting you money back.
If they terminated the contract without cause (ie you didn't cause them any harm), then they would almost certainly have to pro-rate any fees collected. It would probably be a simple matter in small-claims court to collect a partial refund.
In fact, most contracts I've read and written in exactly that way.
The general principle is that if you give somebody something, you should get something for it. Most companies have get-out-for-any-reason clauses, but they typically state that they will give a pro-rated refund if they exercise this option.
The phone service issue you brought up is a separate one. If the phone company terminated service without cause they would almost certainly need to stop billing, although you might have to return any free hardware they gave you. If you terminate early the matter comes down to whether they gave you a lot of free hardware up front. Those "$20" phones might really cost the provider $150, and if they can show that between hardware and setup costs they spent $300 on you, but only collected $10, then they're probably going to be able to collect most of their early-termination fee. If all they provided was service and no hardware then an early termination fee will be much harder for them to enforce.
Usually these contracts tend to be enforced in a reasonable manner. Basically, if you get something they can charge you for it, and if you didn't get something they need to refund it if they just decided to cut you off. Also, if the deal was payments over a long time and they are the ones to cut you off, then you may even be able to keep the up-front benefits without ever paying for them - since they were the ones to terminate.
Most ISP contracts are month-to-month unless they recently gave you hardware that you get to keep. If you lease your hardware then it should definitely be month-to-month.
The easiest way to determine whether a nation is a nice one to live in is to try to walk across the border to leave. In the US, you won't even have to go through a checkpoint. In many other democratic nations there is an immigrations stop, but it is mostly to collect entry visas and keep track of the departure of visitors - so that they can keep track of who is still in their country.
If citizens of a nation can be told they cannot leave (except of course for outstanding warrents, court appearances, etc) then it is most likely not a place that you would want to live. If large-scale barriers to sneaking out are in place, then it is definitely not a place you would not want to live - obviously there must be some reason that people are trying to escape.
If a political system truly were utopian then they wouldn't need to use force to compel people to live under it.
I think the fundamental issue is one of freedom. Even if it doesn't benefit society as a whole at all, an argument can be made for libertarianism on the basis that it allows for greater freedom - this is in itself a benefit to society.
Any time you collect taxes you are infringing on the freedom of people to spend their money how they will. Now, to some degree taxes are necessary, and there are things like infrastructure which are natural monopolies where a government is the best form of administration. Likewise, there are externalities like national defence where everybody benefits and therefore everybody should pay.
Can I prove that people should have the right to spend money how they will? Of course not! I can point to societies that are socialized to the point of communism and the resulting carnage (no incentive to work, brutal use of government force due to the involuntary system of economic control, imprisonment of entire populations to prevent people from fleeing to "less progressive" nations).
Socialism is just a limited form of communism - and as a result it is subject to many of the same problems, although to a lesser degree. This is why it must be carefully regulated and reigned in - lest it get out of control to a degree that nations can't afford it.
Last time I checked the concept that people are endowed by their creator with inalienable rights was just an opinion as well. If you want solid proof the only thing you'll find is that the group with the biggest guns gets to write the rules - by this logic it doesn't matter which side is right or best, just which side is more motivated to use force and has it in the greatest quantity.
Slashdot Mirror
Maybe this is a good argument to keep it all on the server (although until recently my corporate email kept quotas and prevented you doing that).
:)
/home that isn't trivially replacable (the latter goes in a designated directory tree and is excluded).
Which is exactly what I do...
$ cd ~/.maildir ; find . | wc -l
22454
$ du -s .
715632 .
Just run courier-imap and I can access my mail from thunderbird, a webmail client, or whatever the email-program-de-jure is. File corruptions? Forget it - they're just text files. And backing up 715MB of text files is completely trivial - probably compresses to 50MB. I backup everything in
You don't even need to run your own domain to do all this - you can set this up using fetchmail to pull your mail out of gmail/ISP/yahoo/hotmail, or whatever else freepopsd supports.
There is no reason to delete email in this day and age unless you are a business covering your tracks (err, complying with a record retention policy). If somebody had to take the time to type it, then it can't possibly cost more than a microcent to store.
I'm not intimately familiar with Fedora, but many distros backport security patches. So, even though you might be two versions behind bleeding-edge you may in fact be completely secure. Backported patches have the advantage of being less likely to break things.
The only reason that YouTube, Google Video et al adopted Flash as their video player client was because Flash is pretty much universal
As long as you're not running 64-bit... I end up running vmware or firefox in a chroot whenever I need to access a page that runs flash. Both are a major pain. You can't even fire off konqueror in a chroot as it tries to talk to the 64-bit versions that are already running and gets confused.
This is the problem with proprietary protocols - you are supported only if the vendor feels it meets their business plans.
What is wrong with just putting an avi file using an open codec on your website? Why does it HAVE to be flash? And if you need vector graphics what is wrong with SVG?
Does it really NEED to cost millions of dollars or is this our government's way to protect business models?
Well, if you got rid of the requirement to test pharmaceuticals before marketing them then the costs would fall DRASTICALLY. On the other hand you'd have a million brands of white pills and no real way to figure out which ones work and what side-effects they might have. The government would undoubtedly try to do trials, but it would be impossible to keep up (there would be many times as many drugs on the market as there are now - imagine doing a comparative study of 500 different cholesterol-lowering drugs). On the other hand, the instant that a particular drug is shown to work better than the others then everybody would copy it and the company which came up with it wouldn't get any particular reward. As a result, companies would not have any incentive to make drugs that actually work, just drugs that plausibly sound like they could work.
If there were a way to generate the existing required safety and efficacy data more efficiently somebody would be doing it. While people generally only hear a couple of big names when it comes to major pharmaceuticals, the fact is that outsource suppliers of almost all phases of development are available, from animal testing to formulation and manufacture and QA testing to clinical trials. These are mostly small companies and they compete vigorously for contracts with the big pharma outfits (who often use them to offload work they can't handle at the moment). If one of these companies found a more efficient way to do any of these development phases they would get a ton of business and all pharma development costs would drop as a result. Big pharma is in a bit of a cost crisis right now - new drugs aren't being developed fast enough to fuel their balance sheets and most companies are looking for ways to cut costs as price increases are not well-tolerated by the public and are made difficult by insurers.
Agree on the 3100cn. I got sick and tired of cleaning the inkjet and having the family have to fuss with it (and inevitably run into issues while I was gone). The 3100cn is a bit large, but it has great color output, is very fast, and is very economical (1.5 cents bw, 4.5 cents color per page). I figure that it will pay for itself in no time since the toner cartridges don't dry up and cost a lot less per page. Color wasn't an absolute necessity (I get my photos done at Walmart), but at $300 shipped on sale how could I not order one?
Beagle, Fast User Switching, apps like Rhythmbox and Evince, GTK/Cairo with good Python bindings -- these all provide real value to me on a daily basis.
FYI - Fast user switching is in fact available in KDE. It just launches a new kdm session on a new X display and optionally locks the current one. Works just fine for me.
Most of the other stuff are just applications - I use evince all the time on KDE. I don't care what set of widgets an app uses so much as how the overall window-manager and integrated apps like a browser work. The rest you can really just swap out all you want.
OK, so we slingshot it around a few times. Orbital mechanics may not be well understood by me, but the folks at NASA seem to have it down pretty good.
I have every confidence that the NASA guys could get a payload to the sun. It actually only takes high school physics (well, it also takes state-of-the-art engineering).
The problem is cost. You have to generate a delta-V of approximately the Earth's orbital velocity - that is 30 km/s. The parent post suggested that the Sun would do most of the work. This is incorrect - the sun will only do work once the craft has no solar orbital velocity - then it will just fall straight down into the sun. The real work is getting rid of 30 km/s of orbital velocity - if you don't do that the waste will just orbit the sun very close to the earth, and sooner or later it will come back (even if it had escape velocity - that just gets it out of earth's orbit - if it swings around the sun a few times and comes back at us it will still re-enter earth orbit).
The problem is trivial to surmount - you just need a really big rocket. But then again, keeping the waste on earth just needs a really big hole in the ground. The only real decision is which engineering project is more expensive or risky - and most likely it will turn out to be the hole in the ground.
Slingshotting it around a few times is not really a great solution - you still need a ton of energy to get to anything to slingshot off in the first place. The other problem is launch windows - if you want to do multiple slingshots then you have to be really patient for a window. The craft will also need a lot of course corrections - if you're going to launch thousands of waste containers that is a lot of manpower to keep them all on course (unless you just want to drop them on Venus - but even that needs accuracy if you don't want to risk slingshotting it back into solar orbit near the earth). So, maybe with some fancy slingshotting you might only need 15-20 km/s of delta-V - that is still a lot of impulse. The Saturn-V was good for about 7.5GNs - so that is good for about 300 metric tons of payload if you only need 15 km/s (plus escape velocity). Well, that is actually a major overestimate - it doesn't factor in the weight of the rocket itself (which is an exercise in calculus which I'm not bored enough to do). It certainly can be done, but you're talking about a lot of HUGE rockets.
All in all, the hole in the ground is probably the best bet.
but if they believe your health exists entirely in a vacuum without affecting anybody else, they have no understanding of how social costs are... well... social.
Yes, but in the real world people will tend to act in the intrests of self-advancement.
For example, most people would consider me to have above-average abilities, and as a result I can be much more productive than many others when I work. Wouldn't it be in the interests of society to compel me to work a much longer day than others who are less productive? After all, many poorly educated workers produce less useful output in a day of hard work than some well-educated workers might produce in an hour - so why bother having those folks work at all? Why not have them man the whips and guns and they can see to it that the more educated members of society so that the overall level of output is higher?
The problem is that in the real world this doesn't work - the educated will simply find a way to escape and set up a more capitalistic society where the folks reaping the most benefits of work are those doing the most work.
Obviously this is an imperfect analogy, and is not meant to describe any kind of functional society, and I would acknowledge that some forms of manual labor can be very high in value-output and that some forms of educated labor may not. The point is just that when I work I do so to benefit myself and my family first, and others as I am able to voluntarily pitch in. If somebody put a gun to my head to get me to work harder in the interests of the better good I would quickly figure out how to game the system and get to be the person holding the gun and not the person it is pointed at (and this is why all the brains in a communistic society aim for party leadership and not doing real work).
None of this is to say that there can't be a balance, but ultimately if somebody tells me that I can't do something because it lowers my health and therefore my labor value to society I'm going to resent society for treating me merely as a resource to be exploited. It would be one thing if I were on the dole - but if I'm paying my taxes then my debt to society is paid in full - it is not appropriate for the government to act as a taskmaster and determine if I'm working as hard as I could...
Of course it is - but that is what Linus gets for not keeping his licensing flexible. The FSF requires assignment of copyright and the "or a later version" wording precisely so that the license can be adapted as needed.
The FSF shouldn't have to set themselves in stone because a few big third parties became dependant on their code without taking proper legal precautions...
When you buy commercial software (well most of it) - you don't buy to own it, you buy to own a license.
Uh, care to cite the appropriate law here? And I don't mean the EULA (an after-the-fact contract that lacks consideration).
If you buy a book you OWN it. You don't have the right to copy and distribute it, but you do in fact own it. You can resell it.
Ditto for software - if it looks like a duck and walks like a duck you can't put a piece of paper in the box that says it is really a giraffe.
Sure, you won't find any mainstream software vendor who supports this view, but I'm sure if you talk to folks at the EFF or FSF (who are the ones writing the GPL that is the topic of this discussion) they would generally hold to the idea that software is sold. I think you'll find very little law in support of any other concept - just a lot of big-software-company press releases.
This would only be a problem if some newer version of glibc under GPLv3 comes up with some super-fancy implementation of memcpy and the kernel wants to use it. In that case they could not copy the code (though they could clean-room reimplement it).
They could continue to use the old GPLv2 version as long as they wanted to...
Modifying an autopilot by smashing it with a big rock prior to installation might cause a disaster. What safeguards are in place to prevent this? Sure, they might put 1" armor on it, but I'm sure that some high explosive will still damage the device.
Ultimately the users of safety-critical hardware are the ones most responsible for using it correctly. Their first step should be to screen their vendors for systems that are adequately tested. However, if they feel that they can modify the hardware to make it more suitable they should be able to do so (although they might have to apply for various government certifications in some industries to use the resulting new product). The user has every incentive to exercise due care with the equipment.
Airlines spend tons of money on maintenance to make sure their planes don't just fall out of the sky. Why would anybody think that they would just start flashing their firmware willy-nilly simply because the hardware allows it. If they did develop their own firmware they would almost certainly test the living daylights out of it and install it under thorough controls. The FAA would be able to tell what firmware was running just by inspecting the aircraft's maintenance logs.
Safety-critical industries are generally very aware of the impact of the software they use, and since the costs of testing their software right is less than the cost of a planefull of wrongful death lawsuits they're going to be careful about flashing their EPROMS - regardless of vendor support.
What will happen if the FSF gets its way is that manufactures will simply not use OSS code at all. Not only will the user not be able to modify the code, there will also be no open code to look at. The FSF will lose and users will be back where they were before there ever was a GPL.
How is this a loss for the FSF? Manufacturers might not use GPL software, but this will come at a cost that will need to be passed onto consumers. Other manufacturers might continue to use GPL software with the added strings, and they might have a lower cost basis, and therefore a marketplace advantage.
All those manufacturers use GPL software right now because it works well and it is free-as-in-beer. Those two incentives will still remain, and having your product 80% done before you start on it with no licensing costs is an advantage that will not lightly be dismissed.
If you don't like DRM then don't buy or use devices that implement it.
Well, until there are no devices that do not implement it. In fact, the new GPL will help raise the costs of DRM hardware (due to the need to potentially license another OS and/or deal with a less mature product than GPL software) - which will help slow its adoption and make those devices that don't implement it a little more accessible.
In any case - there is nothing to fight over. If you don't like the new GPL don't use it - just don't expect much help from those who do use it or the ability to leverage their software.
It is implied that you can twaek the source code. Not the way how the hardware works. OpenSource Software license should not force how the HARDWARE should work. I think it needs separate license/certification for hardware. Like OpenHardware. This could solve the problem - then you will known - this hardware is OK, it is compatible, we like it. That other hardware is piece of smelly crap and we don't like it. Simple.
Great - so one day consumers might have a choice between a 15GHz Intel PC for $400, or a 1GHz OpenPC for $1500. Some choice - nobody will buy open hardware unless it is mainstream, and it will never be mainstream unless the average person can see the benefits - most average people don't change their OS.
If I write a software package I would prefer that it not be usable by the customers of closed hardware. This means that the vendors of closed hardware will need to reinvent my software package or open their hardware. This makes the open hardware more competitive (either by making more mainstream hardware open, or raising the costs of closed hardware). If I spend my time writing software it isn't so that some proprietary vendor can use it in their effort to lock down their hardware.
"Trusted computing" is simple an implementation of hardware that can be controlled to execute software or not based on the fact who you trust.
Define "you". If I buy a computer and I am provided all keys embedded in the hardware (and their associated related keys) then it functions as you describe - I can sign software and the computer will trust it, or I can sign data and a remote computer that trusts my computer will trust the data.
However, this is not the model envisioned in Trusted Computing. In general the keys and associated related keys will not be given to computer owners, but will be maintained by a 3rd party. Therefore, it isn't about who "you" trust, but rather who "they" trust. The "they" in this case might be a media distributor who doesn't trust "you" but rather wants to communicate directly with your computer hardware, which they trust only because you don't have control of it.
If my computer had a chip in it that would track all software running on it and certify this list using a keypair that I provided it with, and if the chip would only allow software I signed to be run, then I would consider this an EXCELLENT feature. It gives me more control over my own hardware and makes rootkit detection foolproof and remote software installation impossible (without my authorization). The key is whoever knows the keys controls the whole scheme. I submit that the keys should be owned by whoever pays for the hardware.
Disclaimer - I'm not the grandparent poster so I can't say why he would want to use it. But, here are a couple of reasons why developers might want to use it:
It reduces the amount of free-as-in-beer software available to manufacturers of locked-down / DRM hardware - thus it increases the cost of such hardware and makes open hardware more competitive. Just think how much more expensive a Tivo-like appliance might be if they didn't get to use linux - they might consider allowing their hardware to run unsigned binaries as a result rather than licensing an alternative OS.
It ensures that if somebody makes public commercial use of my code and generates a lot of improvements that I will be able to incorporate those improvements into my code. Right now they could just run it on a webserver and not distribute their source.
Keep in mind that the time free software developers is valuable - many would like to receive compensation for it. If a private user with shallow pockets wants to use their software, well then fine. If a big megacorp is going to make a mint using the software they should at least give a little something back to the original developer - either their source code or a negotiated non-GPL license fee. Think MySQL - it would never work without the GPL (that is - their business model).
Nobody is forced to accept the GPL - they just need to reinvent the wheel (or at least the parts of the wheel that aren't available under BSD/etc).
Well, I guess it is better than no insulin at all (even if they are not nearly as safe as recombinant insulin). Just crush up pancreas and run it through a column (or whatever they do).
Kind of like hospitals donating expired drugs to 3rd-world countries. There are definite safety and efficacy problems with this, but they're still helping more people than they are harming.
If there was a conspiracy to raise prices, then some company that wasn't in on it would apply to make generic insulin the old way and charge the old prices for it. While making generic drugs isn't super-cheap, there aren't many artificial barriers to entry.
So many companies currently make insulin that it would be unlikely that they could form a trust. All one company needs to do is undercut the rest.
In the real world. price fixing is very hard to maintain, and it is usually caught when it happens (eventually).
When was the last time that a new techonlogy RAISED the cost of generic medications? Non-patented medications are very inexpensive, are made in numerous countries (and are allowed to be imported into the US), and the US has some of the cheapest prices for these in the first world.
Patented drugs are a different story, but even these only can support lofty prices if they are substantially better than generic medicines. Just watch Pfizer's earnings with Lipitor once generic simvastatin floods the market. Lipitor is shown to be better, but for the minor gains many insurers will opt for cheaper prices, and that will drive down all the costs.
No, if safflower insulin wants to sell for ten times the price of other brands of insulin, all they have to do is convince the doctors, not the patients,
All insurance companies have preferred formularies, and they WILL substitute drugs that are bio-equivalent. The doctor prescribes insulin - not a particular brand. If they do prescribe an insulin-derivative there is usually some reason for it. This company is making plain old insulin - nothing that couldn't be substituted.
though they could get bonus points for showing that it's safer than pig insulin or the current engineered bacteria that produce it
Well, nobody uses pig insulin anymore I think. It would certainly be safer than pig insulin, as would any other route of manufacturing recombinant human insulin. If it is safer for some reason than other sources of human insulin then perhaps they should be allowed to make a profit for their work in making life safer for diabetics.
Most likely this technology will just be used to make insulin cheaper to manufacture, and all prices will go down as a result. Not everything is a big conspiracy - especially where generic drugs are concerned. Non-branded drugs are very inexpensive, and are often cheaper in the US than in other nations which do not use market pricing. Now, branded drugs are a different story - but nobody has to use those - they could just use 10-year old cheap technology (with the associated reduced benefits).
All the moaning about expensive drugs is like moaning about expensive plasma TVs. I own a cheap CRT set and I don't feel like anybody owes it to me to sell a 50" plasma for $300. There are lots of cheap drugs out there, and like anything else you get what you're willing to pay for. If it weren't for branded drugs most likely you would be taking the cheap drugs anyway, as that would be all that was available.
Vmware runs fine on linux, and overhead isn't too bad (but keep in mind that if you want really good performance you'll need enough RAM to run the combined total of both OS's). However, not really ideal for gaming, as vmware does not support Direct3D/OpenGL well (I think there is some experimental support for it, but I can't vouch for how stable it is). Note that I've gotten my share of blue-screens using vmware (granted, this is on amd64 which is less than rock-solid for most software), and even the odd kernel panic (it does run at least partially in kernel space with a few modules - not sure why it would have to do that).
Still, I find it great for running those one-off windows-only applications without needing to dual-boot (which of course takes my server offline).
I doubt many SPARC users will care, but I'm sure many of us amd64 users wouldn't mind having our platform supported as something other than an afterthought.
If you're going to read and write files or network streams, you shouldn't be using data types like int/etc. Use the defined-size types, and then you don't have to worry about portability! It is just something that needs to be kept in mind, and it doesn't have to be a hardship. Plus it saves a shower of patching after the fact...
You also can't hide from a different installation of Windows that has the infected disk mounted. Rootkits hide themselves by hooking into the running kernel/fs drivers - inspect the disk with a clean install and they can't hide then either.
Completely true, but it is a lot more convenient to scan with a linux boot-CD than to keep two windows partitions on the same computer and take turns scanning one from the other. I don't believe there are any windows boot-CD solutions, and I believe that the reason is licensing (and of course lack of interest on the part of MS).
If this driver matures you can bet that it will become the technique of choice for supporting NTFS on a boot disk - you'll see companies like AV and disk imaging vendors jumping on it.
The principle being that if they want, they can stop providing the service and the contract ends at that point, if you have already paid for a period after that time then your going to have trouble getting you money back.
If they terminated the contract without cause (ie you didn't cause them any harm), then they would almost certainly have to pro-rate any fees collected. It would probably be a simple matter in small-claims court to collect a partial refund.
In fact, most contracts I've read and written in exactly that way.
The general principle is that if you give somebody something, you should get something for it. Most companies have get-out-for-any-reason clauses, but they typically state that they will give a pro-rated refund if they exercise this option.
The phone service issue you brought up is a separate one. If the phone company terminated service without cause they would almost certainly need to stop billing, although you might have to return any free hardware they gave you. If you terminate early the matter comes down to whether they gave you a lot of free hardware up front. Those "$20" phones might really cost the provider $150, and if they can show that between hardware and setup costs they spent $300 on you, but only collected $10, then they're probably going to be able to collect most of their early-termination fee. If all they provided was service and no hardware then an early termination fee will be much harder for them to enforce.
Usually these contracts tend to be enforced in a reasonable manner. Basically, if you get something they can charge you for it, and if you didn't get something they need to refund it if they just decided to cut you off. Also, if the deal was payments over a long time and they are the ones to cut you off, then you may even be able to keep the up-front benefits without ever paying for them - since they were the ones to terminate.
Most ISP contracts are month-to-month unless they recently gave you hardware that you get to keep. If you lease your hardware then it should definitely be month-to-month.
The easiest way to determine whether a nation is a nice one to live in is to try to walk across the border to leave. In the US, you won't even have to go through a checkpoint. In many other democratic nations there is an immigrations stop, but it is mostly to collect entry visas and keep track of the departure of visitors - so that they can keep track of who is still in their country.
If citizens of a nation can be told they cannot leave (except of course for outstanding warrents, court appearances, etc) then it is most likely not a place that you would want to live. If large-scale barriers to sneaking out are in place, then it is definitely not a place you would not want to live - obviously there must be some reason that people are trying to escape.
If a political system truly were utopian then they wouldn't need to use force to compel people to live under it.
I think the fundamental issue is one of freedom. Even if it doesn't benefit society as a whole at all, an argument can be made for libertarianism on the basis that it allows for greater freedom - this is in itself a benefit to society.
Any time you collect taxes you are infringing on the freedom of people to spend their money how they will. Now, to some degree taxes are necessary, and there are things like infrastructure which are natural monopolies where a government is the best form of administration. Likewise, there are externalities like national defence where everybody benefits and therefore everybody should pay.
Can I prove that people should have the right to spend money how they will? Of course not! I can point to societies that are socialized to the point of communism and the resulting carnage (no incentive to work, brutal use of government force due to the involuntary system of economic control, imprisonment of entire populations to prevent people from fleeing to "less progressive" nations).
Socialism is just a limited form of communism - and as a result it is subject to many of the same problems, although to a lesser degree. This is why it must be carefully regulated and reigned in - lest it get out of control to a degree that nations can't afford it.
Last time I checked the concept that people are endowed by their creator with inalienable rights was just an opinion as well. If you want solid proof the only thing you'll find is that the group with the biggest guns gets to write the rules - by this logic it doesn't matter which side is right or best, just which side is more motivated to use force and has it in the greatest quantity.
It's just an opinion. Sorry. I am just not interested in opinions.
:)
Well, you sure seem to be on the wrong website then...
What exactly were you looking for here? You certainly don't mind sharing your opinion...