It's no different from any other non-one-time-pad stream cipher. Yes, the randomness is limited, it's limited to the initial secret's entropy and MD5's cutoff (128 bits), which isn't that bad.
AES also isn't a one time pad, but there is a mode for it, pretty similar to the one described by the grandparent poster, where you feed it a key, and you keep reencrypting the previous result against it to get another block of bits; the intermediate results you use for XORring against your plaintext. This is referred to as using another "mode" of AES, stream mode. (there's also counter mode and OFB etc., but I'm a bit rusty, so I'll shut up now. pick up applied cryptography by bruce schneier as an introduction)
Have you forgotten that typical emails will pass between a number of hosts unencrypted as it is being delivered? Where's the advantage in encrypting the last leg of the journey if none of the others are encrypted?
The last leg is either at home, in which case you don't care much typically, or at work. At work, you don't want your sysadmin snooping your mail, or if you're at a client's, have a client's sysadmin snoop mail that might be all about other clients. The last leg is where a webmail connection isn't just one connection out of a zillion, all from people you don't know, but a webmail connection from that guy, there, at that table.
Of course, that's not just white collar. 10+ doesn't seem too bad, actually, I commute 10+ hours a week, and I live in a pretty densely populated country, The Netherlands.
I was gonna mod you up, but I'm gonna have to join in the rave, right here. I've got a pair of Koss The Plug headphones that I really like -- it takes some getting used to stuffing all that foam in your ear, but it isolates outside sound great. Not 44dB, but still impressive. Apparently sony also have an affordable, and better-sounding, isolated headphone, which is surely on my list of stuff to get. I've tried Sony's active noise cancelling headphones, and the effect is amazing; you'd be standing next to a subway wagon speeding past hearing almost nothing. Kinda scary, even. Neatest thing is a switch to turn noise cancelling on or off so you can hear the effect (or perhaps switching noise cancelling "off" just turns on a noise generator;-).
I'd say, go for some sony plugs - they're really not bad, and you can spend as much or as little as you like. (I also like Sennheiser, for non-isolated/noise-cancelling headphones for everyday use, the low end stuff is great value for money, especially compared to low end crap from Philips).
Usually runner-up brands mention their competitor only by reference, such as "10% cheaper than the Leading Brand"; so as to avoid giving the Leading Brand some Free Advertising.
Some manufacturers even advertise comparisons against the "Leading Brand" when they also own the Leading Brand, and in some cases they even mention the Leading Brand reference even though the product advertised IS the Leading Brand! (This was the case with Tide, I think).
When "Death to America, death to Ireal" is chanted at the end of Irans parliamentry sessions much in the same fasion you might say "Amen" in church...do you really thing we should not take them seriously?
I'd expect shouting "Death to America" has as much bearing on a person's actual willingness to go and do a bit of terrorism as shouting "Amen" had bearing on that person's willingness to follow the Ten Commandments. I betcha a lot more people say Amen, and then go out for some killing (thou shalt not), e.g. soldiers in Iraq, than Iranians shouting "death to America".
But the whole SCO case amply demonstrates that Microsoft has a point. The GPL is certainly good for creating a SCO like FUD lawsuit that can be used to obtain discovery powers and burn huge quantities of legal fees. The best corporate lawyers I have worked with are the ones who avoid the lawsuits in the first place. From that point of view the GPL is a real tar baby and RMS has told me personally that this was essentially his intention all along.
This is no different from any other copyrighted work; if the copyright can be infringed, it can lead to infringement suits. Even whether there is no source, copyright can still be infringed, though using source to link in is slightly easier.
A friend of mine [hazelthompson.com] is a photojournalist, and she says that standard digital SLR is still not high resolution enough to be blown up to 6ft on the wall of an art gallery - for that, you need medium format or at a push 35mm slide film. Sure, resolutions will go up and up, but it's likely to be a few years before digital is good enough for artistic/professional photographers.
It is however great for photojournalism (go figure), since you can take a zillion pictures on a single CF mini harddrive (and use an image tank to pull it empty while you plug in your second CF card); and then the pics are easily transmitted over the 'net.
Also, you can't blow up 35mm to wall-size, as others have pointed out. 35mm will still serve a purpose, but mostly for use with instant cameras and longer term strictly with 35mm SLR. 35mm P&S and not-quite-yet-a-real-SLRs will die, not to mention APS.
Some of the HP IJPs require a 59MB download to install one 37k driver. And 39MB of slow, clunky, and unreliable "Print management" admin software doodads. Which do not want to uninstall themselves.
If you care about such things though, that 59MB download ("printing system" for e.g. laserjet 4100) does include more than 139 free, professional, legal typefaces (the postscript3 types and some more) as truetype fonts. Nice.
OS/2, the Amiga, GeOS etc. were all much more mature, consistent and easy to use than windows 3.1 or even 95. Actually, windows is still a mess, though obviously dead OSses aren't competing against it anymore. The internal inconsistencies even in Microsoft's Office suite are laughable. Powerpoint, Excel, Word, they all work completely differently, even when all you're trying to do is make a table. (Of the lot, I like powerpoint's usability best actually, except for tables. Access is the worst of the lot. It should be taken out and shot.)
However, all the channels -are- already on the cable line. So, at least in theory, you could store all the data coming across it. It'd take some massive storage space, but it's doable.
What's the bandwidth on your friendly local cable though? About 50-850 Mhz, so 800Mhz of bandwidth in total? Remember, you'd be storing everything, the inter-channel noise included!
It's a far better idea to apply some sort of OFDM like scheme where you use an FFT to "tune" into all channels simultaneously, and only record the bits you need. Still a lotta bandwidth, but slightly more finite.
How has YOUR LIFE been effected by the Patriot Act?? I don't want to hear your wild insane "Bush can knock down my door without a warrant" theories, I want to know how YOU have been negatively effected by the patriot act.
I'd tell you, but I'm locked away indefinately without access to lawyers or public hearings.
Yes, I've read that... But how is it actually possible? I mean.. I can understand it with source code - it's just a diff-patch, but what about executeables? How can you avoid downloading the whole thing again, if it's already a compiled executeable?
On my machine, the firefox folder alone contains 12.dll files, with 7 more in the components folder. Also, 3.exes. (not counting plugins and uninstall) Then there are the files in chrome/ etc.
If a change occurs in one DLL only, you only need to distribute that one DLL. Or even, through the magic of binary diffs, just a patch.
It's even conceivable that firefox may be split into even more DLLs or components in future. Or to have binary diffs that know about the structure of DLLs so they can replace a single exposed function.
In windows (XP at least), you drag your object to the taskbar. Whichever task you're hovering over will be brought to the foreground. Then drag to that window and drop.
They have effective control anyway. If they all decided to point their DNS servers to a certain place, then that would be adefacto domain name registry. I'm sure the same applies to IP addresses.
And in actual fact, the EU guideline on telecommunications (which is implemented in National laws) states quite clearly that government is in charge of any and all adressing (be it numeric or names) for all public networks.
It just happens that for now, European governments don't feel the need to intervene and let ICANN/Verisign/ccTLDs and RIPE/ARIN/APNIC do their thing. In future, European governments might just choose to cede control to a different body, such as the EU, or the ITU. Whether this happens tomorrow or in 50 years remains to be seen; this is probably mostly dependent on how much ICANN can screw things up.
Legally though, in the EU, the national governments are in charge of appointing the naming/numbering authorities. They just haven't bother to appoint any one.
This law is about search engines making copyrighted material available (probably google cache, for example). It is not about preventing websites from being found by searching on google. What right do google cache and the internet archive have to reproduce my copyrighted content?
Fair use.
Let's not forget that a blanket ban would also prohibit the syntax-highlighted snippets that allow you to skip search engine spam-sites. Also, they don't cache sites indefinately, and don't even include images, so you're better off going to the original site anyway - if it's reachable and the content is still available, that is.
And let's not forget, any time I visit a website, I make a copy of it. No one gave me permission for that. It just happens because you happen to publish it on a publically accesible website with no access controls whatsoever. Common sense would dictate that you want me to see it, or you wouldn't have put it online. Now, if I have permission to copy it and to cache it, and obviously my ISP has permission to cache it in its proxy, why shouldn't google? Try defining a distinction in legalese that doesn't cut on both sides. Or how about you don't and you let judges decide where the line is, based on current legislation, on a case-by-case basis. If the balance sways too much in one direction in case law, THEN the laws can be fixed. (No judge would let you archive "all the current books, newspaper articles, movies, and music" under existing law.)
What we're seeing now are corporations lobbying for laws to be changed pre-emptively to redress an imbalance in favor of the public that DOES NOT YET EXIST. In theory at least. Of course in actual fact they're just lobbying for the balance to go entirely their way so they can nickle and dime ordinary Joes like us every living breathing moment. Legislators will figure it out, but it'll probably take 40 years, unless we, the public, also lobby them.
Remember, radio stations were outlawed when they started up in the US, because of copyright. Now they're recognized as a straight up advertising channel for the record companies (payola).
His primary concern (and what he believes he's invented) is a method to be backward compatible with browers that have JavaScript turned off, and/or browsers that lack CSS support. (You actually want to support those guys?)
Yes. Yes, you do. Many portable devices don't have those features. Think (smart)phones, PDAs, etc.
Many times JavaScript is shut off because of a security policy, or because it's just plain abused too many times to make websites even more annoying.
In fact, I'd go so far as to say that webdesigners who insist their audience should have javascript switched on are annoying. Quit it, already.
Slashdot Mirror
It's no different from any other non-one-time-pad stream cipher. Yes, the randomness is limited, it's limited to the initial secret's entropy and MD5's cutoff (128 bits), which isn't that bad.
AES also isn't a one time pad, but there is a mode for it, pretty similar to the one described by the grandparent poster, where you feed it a key, and you keep reencrypting the previous result against it to get another block of bits; the intermediate results you use for XORring against your plaintext. This is referred to as using another "mode" of AES, stream mode. (there's also counter mode and OFB etc., but I'm a bit rusty, so I'll shut up now. pick up applied cryptography by bruce schneier as an introduction)
Have you forgotten that typical emails will pass between a number of hosts unencrypted as it is being delivered? Where's the advantage in encrypting the last leg of the journey if none of the others are encrypted?
The last leg is either at home, in which case you don't care much typically, or at work. At work, you don't want your sysadmin snooping your mail, or if you're at a client's, have a client's sysadmin snoop mail that might be all about other clients. The last leg is where a webmail connection isn't just one connection out of a zillion, all from people you don't know, but a webmail connection from that guy, there, at that table.
Mean communte time: 25 minutes (one way, so that's about 5 hours a week).
Of course, that's not just white collar. 10+ doesn't seem too bad, actually, I commute 10+ hours a week, and I live in a pretty densely populated country, The Netherlands.
I was gonna mod you up, but I'm gonna have to join in the rave, right here. I've got a pair of Koss The Plug headphones that I really like -- it takes some getting used to stuffing all that foam in your ear, but it isolates outside sound great. Not 44dB, but still impressive. Apparently sony also have an affordable, and better-sounding, isolated headphone, which is surely on my list of stuff to get. I've tried Sony's active noise cancelling headphones, and the effect is amazing; you'd be standing next to a subway wagon speeding past hearing almost nothing. Kinda scary, even. Neatest thing is a switch to turn noise cancelling on or off so you can hear the effect (or perhaps switching noise cancelling "off" just turns on a noise generator ;-).
I'd say, go for some sony plugs - they're really not bad, and you can spend as much or as little as you like. (I also like Sennheiser, for non-isolated/noise-cancelling headphones for everyday use, the low end stuff is great value for money, especially compared to low end crap from Philips).
Usually runner-up brands mention their competitor only by reference, such as "10% cheaper than the Leading Brand"; so as to avoid giving the Leading Brand some Free Advertising.
Some manufacturers even advertise comparisons against the "Leading Brand" when they also own the Leading Brand, and in some cases they even mention the Leading Brand reference even though the product advertised IS the Leading Brand! (This was the case with Tide, I think).
a scantily-clad hoodcap-model.
When "Death to America, death to Ireal" is chanted at the end of Irans parliamentry sessions much in the same fasion you might say "Amen" in church...do you really thing we should not take them seriously?
I'd expect shouting "Death to America" has as much bearing on a person's actual willingness to go and do a bit of terrorism as shouting "Amen" had bearing on that person's willingness to follow the Ten Commandments. I betcha a lot more people say Amen, and then go out for some killing (thou shalt not), e.g. soldiers in Iraq, than Iranians shouting "death to America".
I'm sure if you could buy PC27-whatever memory you could get gigs at pennies each... but what's the point?
Actually, cheap PC133 SDRAM + battery = still faster than hard drive (1066MB/s vs 150Mb/s plus no seektime), and almost as non-volatile as flash.
So perhaps DRAM isn't too bad compared to flash..
Isn't that like going to France to learn French, and then being horrified that the locals don't speak English fluently?
But the whole SCO case amply demonstrates that Microsoft has a point. The GPL is certainly good for creating a SCO like FUD lawsuit that can be used to obtain discovery powers and burn huge quantities of legal fees. The best corporate lawyers I have worked with are the ones who avoid the lawsuits in the first place. From that point of view the GPL is a real tar baby and RMS has told me personally that this was essentially his intention all along.
This is no different from any other copyrighted work; if the copyright can be infringed, it can lead to infringement suits. Even whether there is no source, copyright can still be infringed, though using source to link in is slightly easier.
A friend of mine [hazelthompson.com] is a photojournalist, and she says that standard digital SLR is still not high resolution enough to be blown up to 6ft on the wall of an art gallery - for that, you need medium format or at a push 35mm slide film. Sure, resolutions will go up and up, but it's likely to be a few years before digital is good enough for artistic/professional photographers.
It is however great for photojournalism (go figure), since you can take a zillion pictures on a single CF mini harddrive (and use an image tank to pull it empty while you plug in your second CF card); and then the pics are easily transmitted over the 'net.
Also, you can't blow up 35mm to wall-size, as others have pointed out. 35mm will still serve a purpose, but mostly for use with instant cameras and longer term strictly with 35mm SLR. 35mm P&S and not-quite-yet-a-real-SLRs will die, not to mention APS.
Some of the HP IJPs require a 59MB download to install one 37k driver. And 39MB of slow, clunky, and unreliable "Print management" admin software doodads. Which do not want to uninstall themselves.
If you care about such things though, that 59MB download ("printing system" for e.g. laserjet 4100) does include more than 139 free, professional, legal typefaces (the postscript3 types and some more) as truetype fonts. Nice.
OS/2, the Amiga, GeOS etc. were all much more mature, consistent and easy to use than windows 3.1 or even 95. Actually, windows is still a mess, though obviously dead OSses aren't competing against it anymore. The internal inconsistencies even in Microsoft's Office suite are laughable. Powerpoint, Excel, Word, they all work completely differently, even when all you're trying to do is make a table. (Of the lot, I like powerpoint's usability best actually, except for tables. Access is the worst of the lot. It should be taken out and shot.)
Anything involving the reduction of scope for C/Z/net to grow is good. Rarely in my life have I ever come across such a poor source of information.
3...2...1..
"You must be new here!"
Actually the current thinking is that we will probably not, in fact, be fucked.
Well, you would think that, what with posting on slashdot and all..
However, all the channels -are- already on the cable line. So, at least in theory, you could store all the data coming across it. It'd take some massive storage space, but it's doable.
What's the bandwidth on your friendly local cable though? About 50-850 Mhz, so 800Mhz of bandwidth in total? Remember, you'd be storing everything, the inter-channel noise included! It's a far better idea to apply some sort of OFDM like scheme where you use an FFT to "tune" into all channels simultaneously, and only record the bits you need. Still a lotta bandwidth, but slightly more finite.
How has YOUR LIFE been effected by the Patriot Act?? I don't want to hear your wild insane "Bush can knock down my door without a warrant" theories, I want to know how YOU have been negatively effected by the patriot act.
I'd tell you, but I'm locked away indefinately without access to lawyers or public hearings.
Please phone your rep and let him know he did the right thing.
Yes, I've read that... But how is it actually possible? I mean.. I can understand it with source code - it's just a diff-patch, but what about executeables? How can you avoid downloading the whole thing again, if it's already a compiled executeable?
.dll files, with 7 more in the components folder. Also, 3 .exes. (not counting plugins and uninstall) Then there are the files in chrome/ etc.
On my machine, the firefox folder alone contains 12
If a change occurs in one DLL only, you only need to distribute that one DLL. Or even, through the magic of binary diffs, just a patch.
It's even conceivable that firefox may be split into even more DLLs or components in future. Or to have binary diffs that know about the structure of DLLs so they can replace a single exposed function.
Can it run Linux?
It find Mac/OS to be easier to use.
In windows (XP at least), you drag your object to the taskbar. Whichever task you're hovering over will be brought to the foreground. Then drag to that window and drop.
They have effective control anyway. If they all decided to point their DNS servers to a certain place, then that would be adefacto domain name registry. I'm sure the same applies to IP addresses.
And in actual fact, the EU guideline on telecommunications (which is implemented in National laws) states quite clearly that government is in charge of any and all adressing (be it numeric or names) for all public networks.
It just happens that for now, European governments don't feel the need to intervene and let ICANN/Verisign/ccTLDs and RIPE/ARIN/APNIC do their thing. In future, European governments might just choose to cede control to a different body, such as the EU, or the ITU. Whether this happens tomorrow or in 50 years remains to be seen; this is probably mostly dependent on how much ICANN can screw things up.
Legally though, in the EU, the national governments are in charge of appointing the naming/numbering authorities. They just haven't bother to appoint any one.
I think you mean altavista. And of course webcrawler.
Fair use.
Let's not forget that a blanket ban would also prohibit the syntax-highlighted snippets that allow you to skip search engine spam-sites. Also, they don't cache sites indefinately, and don't even include images, so you're better off going to the original site anyway - if it's reachable and the content is still available, that is.
And let's not forget, any time I visit a website, I make a copy of it. No one gave me permission for that. It just happens because you happen to publish it on a publically accesible website with no access controls whatsoever. Common sense would dictate that you want me to see it, or you wouldn't have put it online. Now, if I have permission to copy it and to cache it, and obviously my ISP has permission to cache it in its proxy, why shouldn't google? Try defining a distinction in legalese that doesn't cut on both sides. Or how about you don't and you let judges decide where the line is, based on current legislation, on a case-by-case basis. If the balance sways too much in one direction in case law, THEN the laws can be fixed. (No judge would let you archive "all the current books, newspaper articles, movies, and music" under existing law.)
What we're seeing now are corporations lobbying for laws to be changed pre-emptively to redress an imbalance in favor of the public that DOES NOT YET EXIST. In theory at least. Of course in actual fact they're just lobbying for the balance to go entirely their way so they can nickle and dime ordinary Joes like us every living breathing moment. Legislators will figure it out, but it'll probably take 40 years, unless we, the public, also lobby them.
Remember, radio stations were outlawed when they started up in the US, because of copyright. Now they're recognized as a straight up advertising channel for the record companies (payola).
Yes. Yes, you do. Many portable devices don't have those features. Think (smart)phones, PDAs, etc. Many times JavaScript is shut off because of a security policy, or because it's just plain abused too many times to make websites even more annoying.
In fact, I'd go so far as to say that webdesigners who insist their audience should have javascript switched on are annoying. Quit it, already.