The only reall difference between SCSL and GPL in this instance is that with the GPL you can go your own way and distribute it anyway [if Linus doesn't like your change].
That is precisely what Free Software is all about. Remember, "free speech, not free beer".
This confusiong is what RMS warned us about a few months ago when he said the people arguing for "Open Source" were putting pragmatism ahead of the important things and that eventually people would forget which freedoms they were fighting for.
To put it another way, what good is having the hood of your car not welded shut if you have to sign a contract binding you to not ever have a third party mechanic look under there anyways.
1) RAM is relatively cheap compared to what it was before but it's still expensive enough that adding 5minutes of buffer would make a dent on a $300 device.
2) adding RAM would consume power itself. 3) spinning down the CD would be a tradeoff, spinning it up is more expensive than keeping it spinning for a while. I'm not sure where the tradeoff is.
4) Any number quoted for shock protection these days already takes into account compression. In fact even standard CD players with 40s shock protection are using lossy compression to give you that. The best ones allow you to disable it for a noticeable improvement in sound quality.
On the contrary Stallman has endorsed Linux as the current kernel for the GNU project many times. In fact the people at the FSF at the time were some of the early adopters of linux.
Work on the HURD continues, I'm not sure on Richard Stallman's views on the technical merits of the microkernel architecture, but in my opinion we're starting to see the real limitations of the monolithic kernel now. kernel modules only go so far.
Instead of a well-designed well-optimized microkernel we get either kernel modules which are nice in that they can be upgraded without restarting, but you can't try out a suspect one without trusting the stability of the entire system to it. And you can't test it concurrently with the stable one for production use. Or we get adhoc microkernel style subsystems like xfs/arla anyways.
I'm not sure how you can reconcile "never work in practice" with the manifest sucesses of microkernels like QNX and NT. Just because Digital Unix, sorry OSF/1, oh sorry True64Unix was so miserable doesn't mean it has to be that way.
Sigh, I count two uses of "K" in your algorithm. Note the prominent use of "One" in "One-Time-Pad" As soon as you use the pad a second time it's no longer a "One-Time" Pad...
What you describe is basically how a stream cipher works, except that K is supposed to be continuously changing based on an initial seed from the key. If the stream doesn't change then yes it's totally insecure.
Which isn't to say that's stopped people from using such things. Examples of such encryption are the MS SQL 7 login packet and the MS Windows.PWL file (the former it appears is intentional and the latter due to incompetence) To be fair to MS lots of other companies have marketed products that have similarly insecure ciphers.
The danger is that something like this succeeds
on
Beaming Money
·
· Score: 1
This is basically a way to make credit payments to people who can't accept credit card payments themselves. It's not for ebay or other real merchants. It's likely to be very popular and make them lots of money I'm afraid. Banks actually make more than just the interest on the accounts, they make money on the float, it's basically free money that the banking money synthesizes (Now that's the real start trek stuff, except it's the basis for the existing brick and morter banking business).
The problem is that the availability of faux-ecash systems like this will only make it less likely that we ever get true electronic cash. Mondex, et al are all just "better credit cards" that avoid having to pass your credit card in the clear but don't do anything to preserve your privacy. Do you really want your credit card company to have a list of everyone you exchanged cash with and every store you shop at?
well, flame #1: You released the source, but not under an open source license. Really you should contact the trademark holders and ask if your license is an open source license before you go claiming it is. As it is you are misusing the trade mark and the open source board is obligated to enforce their trademark. There's more to "open source" than merely releasing the source code. Users must be free to work with it freely, including not having to tell you every time they modify a file. The "Open Source" trademark violation aside, the privacy issues of requiring someeon contact you before they do programming in the privacy of their own home is chilling.
Slashdot Mirror
The only reall difference between SCSL and GPL in this instance is that with the GPL you can go your own way and distribute it anyway [if Linus doesn't like your change].
That is precisely what Free Software is all about. Remember, "free speech, not free beer".
This confusiong is what RMS warned us about a few months ago when he said the people arguing for "Open Source" were putting pragmatism ahead of the important things and that eventually people would forget which freedoms they were fighting for.
To put it another way, what good is having the hood of your car not welded shut if you have to sign a contract binding you to not ever have a third party mechanic look under there anyways.
1) RAM is relatively cheap compared to what it was before but it's still expensive enough that adding 5minutes of buffer would make a dent on a $300 device.
2) adding RAM would consume power itself.
3) spinning down the CD would be a tradeoff, spinning it up is more expensive than keeping it spinning for a while. I'm not sure where the tradeoff is.
4) Any number quoted for shock protection these days already takes into account compression. In fact even standard CD players with 40s shock protection are using lossy compression to give you that. The best ones allow you to disable it for a noticeable improvement in sound quality.
On the contrary Stallman has endorsed Linux as the current kernel for the GNU project many times. In fact the people at the FSF at the time were some of the early adopters of linux.
Work on the HURD continues, I'm not sure on Richard Stallman's views on the technical merits of the microkernel architecture, but in my opinion we're starting to see the real limitations of the monolithic kernel now. kernel modules only go so far.
Instead of a well-designed well-optimized microkernel we get either kernel modules which are nice in that they can be upgraded without restarting, but you can't try out a suspect one without trusting the stability of the entire system to it. And you can't test it concurrently with the stable one for production use. Or we get adhoc microkernel style subsystems like xfs/arla anyways.
I'm not sure how you can reconcile "never work in practice" with the manifest sucesses of microkernels like QNX and NT. Just because Digital Unix, sorry OSF/1, oh sorry True64Unix was so miserable doesn't mean it has to be that way.
Sigh, I count two uses of "K" in your algorithm. Note the prominent use of "One" in "One-Time-Pad"
.PWL file (the former it appears is intentional and the latter due to incompetence) To be fair to MS lots of other companies have marketed products that have similarly insecure ciphers.
As soon as you use the pad a second time it's no longer a "One-Time" Pad...
What you describe is basically how a stream cipher works, except that K is supposed to be continuously changing based on an initial seed from the key. If the stream doesn't change then yes it's totally insecure.
Which isn't to say that's stopped people from using such things. Examples of such encryption are the MS SQL 7 login packet and the MS Windows
This is basically a way to make credit payments to people who can't accept credit card payments themselves. It's not for ebay or other real merchants. It's likely to be very popular and make them lots of money I'm afraid. Banks actually make more than just the interest on the accounts, they make money on the float, it's basically free money that the banking money synthesizes (Now that's the real start trek stuff, except it's the basis for the existing brick and morter banking business).
The problem is that the availability of faux-ecash systems like this will only make it less likely that we ever get true electronic cash. Mondex, et al are all just "better credit cards" that avoid having to pass your credit card in the clear but don't do anything to preserve your privacy. Do you really want your credit card company to have a list of everyone you exchanged cash with and every store you shop at?
well, flame #1: You released the source, but not under an open source license. Really you should contact the trademark holders and ask if your license is an open source license before you go claiming it is. As it is you are misusing the trade mark and the open source board is obligated to enforce their trademark. There's more to "open source" than merely releasing the source code. Users must be free to work with it freely, including not having to tell you every time they modify a file. The "Open Source" trademark violation aside, the privacy issues of requiring someeon contact you before they do programming in the privacy of their own home is chilling.