Actually, a single counterexample does invalidate a general claim. That's the point.
What you mean is that there may exist ciphers where encrypting twice does increase security. That's a much less general claim than the one the original poster was seeking to invalidate.
Take the result, and encrypt it again with a different 16 bit number.
Now, how many possible keys do you have to go through to get back the original data?
(Hint: Try single decryption using key3 = key1 XOR key2.)
This counterexample invalidates the claim that encrypting twice with an N-bit key is equivalent to encrypting with a 2N-bit key. It also demonstrates the worst possible case, that encrypting twice may be no more secure than encrypting once.
What GPLv3 might do is ensure that vendors like IBM continue to at least make some general-purpose computers available for sale, if only for business use.
Send them a DMCA takedown. Get one drafted by a lawyer so it looks nice and official, turn it into a template. Send them by postal mail. Their standard complaint channels aren't going to achieve anything; they're just there to shut people up.
seems in cases where the purchaser uses paypal to complete the transaction that ebay should automatically positive feedback/reputation points as they have successfully confirmed that the buyers has fulfilled their part of the transaction.
Absolutely not. There are all kinds of ways a buyer can dick me around after sending money via PayPal.
If the buyer (say) pretends he didn't get the item and demands a refund, even though I have delivery confirmation saying the item was delivered, I need a way to be able to leave appropriate feedback.
Then I also hate being asked for my password repeatedly when shifting between eBay and PayPal, buying and selling, etc.
Well, being security minded I will not use the same account for banking that I use for something else. That's why I won't touch Google's e-commerce offerings.
Think about it--would you walk into a store, hand over your wallet at the door and say "Hey, I trust you, take whatever's appropriate"?
Yup. I've been buying more stuff from Amazon Marketplace and less from eBay, because prices on eBay are ridiculous once you factor in the usual price gouging on the shipping.
On the other hand, I haven't had any problem selling stuff on eBay, because I pick a reasonable starting price and charge fair shipping rates.
Imagine a future world where every PC has Trusted Computing installed, and is unable to run any code not signed by Microsoft. Microsoft takes a bunch of GPL code (say, GCJ) and starts selling it as a product, signed with their secret key.
Sure, you have the source--but you can't run it on any PC. You could theoretically run it on some piece of hardware you built yourself. Can you not see how that would be destroying the freedoms the GPL was supposed to protect?
If ODF becomes the worldwide standard GP predicts, MS will support the standard in their product, and MS Office will continue to prevail.
I don't care if Microsoft Office continues to occupy the majority of the market, so long as I am not forced to use it.
The problem is documents being available only in closed, proprietary formats. If Office is made to use ODF, and I can read Office-made documents in my choice of application, them everyone's happy.
Except Microsoft, because they know that without lock-in, they'll have to drop the price of Office dramatically in order to compete. And it's one of the few things they make a profit on, and they're already facing shareholder discontent, so...
The "spirit of the GPL" is that if I own the hardware and the software, I should be able to modify the software as I wish, run the modified version on my hardware, and distribute the modified version.
It's spelled out pretty clearly in the many documents RMS has written on the topic. Example:
The freedom to run the program, for any purpose (freedom 0).
In the TiVo case, I do not have the freedom to run modified TiVo software for the purpose of controlling my TiVo. Whether I have lost the freedom due to copyright or due to restrictive hardware is beside the point; preservation of freedom 0 is one of the explicit goals of the GPL.
I think Linus is 100% wrong on this one, just like he was 100% wrong when he chose BitKeeper.
Not doing anything about it is suicide too. If you're right any we end up in a world where only binaries signed by hardware manufacturers can be executed, then OSS is dead anyway.
The 68000 used 3x 16 bit ALUs chained together to simulate a 32 bit ALU. It was basically a 16 bit chip internally, built to execute an architecture designed to be 32 bit.
I found it wasn't CRTs' flicker I was allergic to; it was the fuzziness of the vertical lines on conventional tubes. I could handle a CRT, but only a Trinitron tube. Anything else would give me a headache inside half an hour.
This was kinda hard to explain at work, unfortunately. I ended up wandering the halls looking for spare monitors and trying them out until I found one with a Trinitron tube. They were (and still are) too cheap to buy LCDs, but fortunately I've been moved to a laptop now.
It was also a pain in that I really wanted an iMac at the time when Apple was only selling CRT iMacs. I ended up paying significantly more for a G4, which was hellishly noisy.
I got rid of the last CRT from the house early this year. Glad to see the back of the damn things.
One advantage I can see is that your documents will be available anywhere you can get access to the web, which can be a pretty compelling argument.
Yes, but I can do that with a wiki. For most documents that I would want to have everywhere, I don't need any formatting that isn't available in plain old HTML (and hence wiki markup).
Oh, I'm sure lots of people will use Google's Writely and spreadsheet. But I'll wait for a free version I can run on my own hosting, and not give all my data to Google. And in the meantime, I'll use tracks and typo and wiki software.
The original Macintosh used a Motorola 68000, which was 16 bit internally (16 bit ALUs, 16 bit data bus, 16 bits of address space accessible at a time, 16 bit instructions), albeit with addresses stored as 32 bits in preparation for later 32 bit models.
Not ditto for Unix. The 16/32-bit transition had merely occured much earlier. It was "ancient" history by the time most people began talking about 32-bit.
Xenix was launched on 16 bit CPUs in 1983. 32 bit support was added by 1987, so the 16 bit version was hardly ancient history by the time people were talking about 32 bits.
Minix was still a 16 bit OS in 1988. If it hadn't been, Linus wouldn't have started writing his kernel.
In 1982 SunOS was launched on Sun 100 machines, which had 68000 CPUs. In 1983, Sun 200 series machines ran on 68010s which was partially 32 bit. So again, I don't think 1982 was ancient history by 1983.
If you mean that the Mac and Unix had transitioned to 32 bit long before Windows, well, yes, I didn't suggest otherwise; I just said that Mac and Unix transitions were uneventful compared to the massive changes and improvements in Windows.
Slashdot Mirror
How about creating an Amazon honor system tip jar?
t ml/ref=zm_pb_h_09/058-4962647-3756209
http://zme.amazon.com/exec/varzea/subst/fx/home.h
Actually, a single counterexample does invalidate a general claim. That's the point.
What you mean is that there may exist ciphers where encrypting twice does increase security. That's a much less general claim than the one the original poster was seeking to invalidate.
Well, that would certainly keep her busy at Microsoft.
Consider the (naive and insecure) XOR encryption.
Take your file, encrypt it with a 16 bit number.
Take the result, and encrypt it again with a different 16 bit number.
Now, how many possible keys do you have to go through to get back the original data?
(Hint: Try single decryption using key3 = key1 XOR key2.)
This counterexample invalidates the claim that encrypting twice with an N-bit key is equivalent to encrypting with a 2N-bit key. It also demonstrates the worst possible case, that encrypting twice may be no more secure than encrypting once.
What GPLv3 might do is ensure that vendors like IBM continue to at least make some general-purpose computers available for sale, if only for business use.
Not your credit card, your wallet, including your cash and your ATM card and its PIN.
Remember, PayPal is not a bank, and for many people PayPal is linked to a real bank account that can be drained.
Send them a DMCA takedown. Get one drafted by a lawyer so it looks nice and official, turn it into a template. Send them by postal mail. Their standard complaint channels aren't going to achieve anything; they're just there to shut people up.
Consider the buyer who receives the item, then lies that he didn't, even though I have delivery confirmation saying I did.
Is it unfair that he didn't get immediate positive feedback when he sent the money?
Absolutely not. There are all kinds of ways a buyer can dick me around after sending money via PayPal.
If the buyer (say) pretends he didn't get the item and demands a refund, even though I have delivery confirmation saying the item was delivered, I need a way to be able to leave appropriate feedback.
Well, being security minded I will not use the same account for banking that I use for something else. That's why I won't touch Google's e-commerce offerings.
Think about it--would you walk into a store, hand over your wallet at the door and say "Hey, I trust you, take whatever's appropriate"?
Yup. I've been buying more stuff from Amazon Marketplace and less from eBay, because prices on eBay are ridiculous once you factor in the usual price gouging on the shipping.
On the other hand, I haven't had any problem selling stuff on eBay, because I pick a reasonable starting price and charge fair shipping rates.
Imagine a future world where every PC has Trusted Computing installed, and is unable to run any code not signed by Microsoft. Microsoft takes a bunch of GPL code (say, GCJ) and starts selling it as a product, signed with their secret key.
Sure, you have the source--but you can't run it on any PC. You could theoretically run it on some piece of hardware you built yourself. Can you not see how that would be destroying the freedoms the GPL was supposed to protect?
Nope, that's the whole point. TiVo make the hardware refuse to run any modified software, even though that software is covered by the GPL.
...artists whose work is only available as DRM-laden iTunes downloads. Like chunks of Richard H. Kirk's back catalog, and Funkstörung's latest.
Me. Actually, I have 5 of their albums, but I'm planning on getting more.
I don't care if Microsoft Office continues to occupy the majority of the market, so long as I am not forced to use it.
The problem is documents being available only in closed, proprietary formats. If Office is made to use ODF, and I can read Office-made documents in my choice of application, them everyone's happy.
Except Microsoft, because they know that without lock-in, they'll have to drop the price of Office dramatically in order to compete. And it's one of the few things they make a profit on, and they're already facing shareholder discontent, so...
The "spirit of the GPL" is that if I own the hardware and the software, I should be able to modify the software as I wish, run the modified version on my hardware, and distribute the modified version.
It's spelled out pretty clearly in the many documents RMS has written on the topic. Example:
In the TiVo case, I do not have the freedom to run modified TiVo software for the purpose of controlling my TiVo. Whether I have lost the freedom due to copyright or due to restrictive hardware is beside the point; preservation of freedom 0 is one of the explicit goals of the GPL.
I think Linus is 100% wrong on this one, just like he was 100% wrong when he chose BitKeeper.
Not doing anything about it is suicide too. If you're right any we end up in a world where only binaries signed by hardware manufacturers can be executed, then OSS is dead anyway.
The 68000 used 3x 16 bit ALUs chained together to simulate a 32 bit ALU. It was basically a 16 bit chip internally, built to execute an architecture designed to be 32 bit.
The fact that the registers and address space were 32 bit doesn't alter the fact that all processing occurred 16 bits at a time.
The Z80 had 16 bit registers and a 16 bit address bus. Guess what? It still wasn't a 16 bit CPU.
The Z80 had 16 bit registers and a 16 bit address bus. I guess that means it was a 16 bit CPU, eh?
I found it wasn't CRTs' flicker I was allergic to; it was the fuzziness of the vertical lines on conventional tubes. I could handle a CRT, but only a Trinitron tube. Anything else would give me a headache inside half an hour.
This was kinda hard to explain at work, unfortunately. I ended up wandering the halls looking for spare monitors and trying them out until I found one with a Trinitron tube. They were (and still are) too cheap to buy LCDs, but fortunately I've been moved to a laptop now.
It was also a pain in that I really wanted an iMac at the time when Apple was only selling CRT iMacs. I ended up paying significantly more for a G4, which was hellishly noisy.
I got rid of the last CRT from the house early this year. Glad to see the back of the damn things.
Yes, but I can do that with a wiki. For most documents that I would want to have everywhere, I don't need any formatting that isn't available in plain old HTML (and hence wiki markup).
Oh, I'm sure lots of people will use Google's Writely and spreadsheet. But I'll wait for a free version I can run on my own hosting, and not give all my data to Google. And in the meantime, I'll use tracks and typo and wiki software.
The original Macintosh used a Motorola 68000, which was 16 bit internally (16 bit ALUs, 16 bit data bus, 16 bits of address space accessible at a time, 16 bit instructions), albeit with addresses stored as 32 bits in preparation for later 32 bit models.
Xenix was launched on 16 bit CPUs in 1983. 32 bit support was added by 1987, so the 16 bit version was hardly ancient history by the time people were talking about 32 bits.
Minix was still a 16 bit OS in 1988. If it hadn't been, Linus wouldn't have started writing his kernel.
In 1982 SunOS was launched on Sun 100 machines, which had 68000 CPUs. In 1983, Sun 200 series machines ran on 68010s which was partially 32 bit. So again, I don't think 1982 was ancient history by 1983.
If you mean that the Mac and Unix had transitioned to 32 bit long before Windows, well, yes, I didn't suggest otherwise; I just said that Mac and Unix transitions were uneventful compared to the massive changes and improvements in Windows.
Only for DOS/Windows users. For Mac users it was largely a non-event, bar some software incompatibilities. Ditto most flavors of Unix.