I know I posted late in the game. I know I have almost *no* chance of being moderated up. But I don't want a whole lot for Christmas. It's not my type of thing, y'know?
Anyway, my mother asked me the other day: "What do you want for Christmas? Something for your computer? What about books?"
Now, granted, a 128 meg DIMM would rock. And the Camel book would be cool. But they always make me feel so-- empty. I don't feel so much accomplishment in them.
Rather, I'd have old tech or something to create with. I would absolutely *adore* a blank book-- I can write anything in one of those; from debugging notes on my latest project to (horrid) poems that I occasionally set to paper.
If you must get me something math- or science-related, get me a slide rule-- I *adore* them. Or an adding machine, I've been wanting one of those. Or even an abacus. If you want to give me something *really* special, just buy me the beads that I can use to *make* a good abacus. Ye Gods, I would *love* the person who did that.
If you must get me something for my computer, don't get anything that requires some sort of installation. I have no PCI slots left, no ISA slots to begin with (!). All three of my memory slots (it's a cheap-ass motherboard, okay?) have 64 meg DIMMS-- I'm happy. I have a bigass hard drive, more than I'll ever need. Software has to be *for* *Linux*, but you'll likely *never* get that right. In other words, buy me a keyboard. PS/2 (yeah, yeah, quit yer griping, I told you it was a cheapass mobo). That's $15 for a decent one (no ergonomic stuff); not bad.
But please, don't get me anything. Call me instead. Come on over, we'll drink egg nog and work on something together (screenplays to pseudorandom number generators-- we can spend time together doing it). That's what a good friend is, and that's what I'd do for you.
One reason that people use stream ciphers is often a speed factor-- it's much faster to simply XOR the data stream with the keystream from a stream cipher than it is to actually involve the plaintext in the permutation. This is especially true if you implement the stream cipher on, say, a separate chip.
Given the fact that public-key ciphers are generally slower than symmetric ciphers, the main reason to use a stream cipher is pretty much defeated. Why not just use a pulic-key cryptosystem to send the key, then send the rest of the message encrypted with a conventional stream cipher?
Also, cryptologists prefer to work on things that have practical qualities (well, *most* do). Perhaps a public-key stream cipher hasn't been designed because there is very little practical advantage to doing so. Or maybe some of the major cryptologists out there haven't even considered the implications.
As part of your thesis, perhaps you could give some reasons and legitimate uses for a stream cipher. Are there any protocols where this would be useful? Are there applications that would be greatly helped by this? Could the design of a public-key stream cipher give new insights into the mathematical workings of other public-key ciphers such as ElGamal and RSA? Can you show that there is a legitimate reason to use a PKSC instead of key exchange, followed by a conventional stream cipher?
I think it'd be interesting to see a paper on this subject. If you do include a section on this topic as part of your thesis, I think people would be *very* interested in seeing it. Go for it!
The way Gore said it-- "...be worth *two* trillion dollars..."-- makes it sound like Gates is worth *one* trillion right now.
Dan Quayle may need some spelling lessons, but Gore needs a lesson in basic mathematics-- perhaps he could use a refresher course on scientific notation... Or counting, since CTRL-ALT-DEL is a *three* key combination, last time I checked...
Then again, he said that the whole "creating the Internet" thing was due to a "lack of sleep". Perhaps he'll finally fess up and attribute these screw ups to "lack of clue".
To confirm that Mindcraft's previous testing was unbiased and representative of Linux's performance."
It seems to me that their main intent was to validate their results. If the specific intent is to validate previous results, then I will guarantee that bias exists. Will Mindcraft publish results that validate the suspicions that they were biased from the start? No-- that would totally destroy the shreds of credibility they have left.
But I'm not going to totally dismiss the results, either. They followed a definite process, and we can't simply ignore what they say because we don't like it. I think that Microsoft *does* have some areas where it beats Linux-- we need to work on these. Mindcraft is only bringing it to the public eye.
But if Mindcraft is bring problems with Linux into the public eye, that means that developers can show why Linux *is* better. Bug fixes are going to pour in. Changes will be made, enhancements created, performance tweaks and changes will make their way into the kernel and into Apache. Performance will improve. The developers will gain more experience, allowing them to enhance their creations even more.
"...Microsoft, whose trailblazing Windows is the operating system for most personal computers today..."
I am totally against the DoJ on this issue. I'm not a libertarian (as far as I know), but I do tend to think that the best way to make a marketplace better is to make it more free. I also believe that lawsuits like that of the DoJ are symptomatic of a society which wishes to take no responsibility-- nobody was *forced* to enter into a contract with Microsoft, and nobody was *forced* to use Internet Explorer. I'm not going to stay quiet on these issues-- I hold my freedom too dear to not say anything.
But no matter *what* I do, I will *never* support the people over at microsoft.aynrand.org. They spend as much time putting Gates up on a pedestal as they do trashing the DoJ. They seem to think that, if they support Microsoft's right to do business, then they must show Microsoft to be an upstanding and wonderful company. This isn't true-- Microsoft isn't a good company; their software sucks and their prices are inflated.
Put it in perspective: if KKK members were to be banned from demonstrating or exercising their rights to free speech, I would be upset. But that doesn't mean I would put Klan members up on a pedestal and say that they are pillars of society-- that would be bullshit!
Perhaps a better paper to read is this one, which was written (surprisingly) by Eric S. Raymond-- yes, that's right, the same guy who wrote "The Cathedral and the Bazaar". I don't totally agree with this one, either, but at least it doesn't try to make Bill Gates into some sort of demigod while trashing the DoJ until it looks like pure evil.
Besides that, what does product quality have to do with defending the company? You should fight for the freedom of *any* entity whose freedom is being taken away unjustly-- even those that say despicable things or make atrocious products. It doesn't matter-- you should be willing to stand up for freedom.
Okay, I have problems with antitrust law-- I don't consider them right. But I also won't hold up Bill Gates as a visionary who creates "trailblazing" operating systems (http://microsoft.aynrand.org -- ugh!). But I'm going to put aside the moral and political issues for the moment and say this:
I really wanted to see Linux kick Microsoft's ass. I wanted to see Linux, NetBSD, OpenBSD, FreeBSD, and other open source alternatives (even GNU/Hurd) take the place of Windows. And I think it would have happened.
Let's take a look at the following: Western Union was in the business of communications. Messages were telegraphs sent via morse code over copper wires. Along came this goldurned telephone thingy-- it couldn't *possibly* compete with telegraphs! It was an interesting toy, but it had no practical value, and nobody wanted to use it. But telephones are now so commonplace that you have very few situations in which you need to use Western Union. So now WU isn't so big. Why? Because they forgot-- they were in the communications industry, not the telegraph industry. They forgot what business they were in.
Or how about railroads? They're still a big method of transportation. It's still profitable to be in the railroad biz. But not as profitable as it used to be. Why? Because the railroad companies thought that they were in the railroad business-- not the transportation industry. Along came trucks. Trucks were extremely regulated in the beginning, due in part to the legislation and lobbying done by the railroads. But trucking won out-- it's a more common mode of transportation today. The railroads lost because they forgot what business they were in.
Now look at Microsoft. They're in the software business, right? Well, that's the business they *could* be in. But they're in the Windows business, not software. And this Linux thing, this Open Source stuff-- it's the equivalent of trucking to railroads, of telephones to telegraphs. While railroads and telegraphs are still used to a small extent today, telephones and trucks are more common.
Microsoft was on its way to doing massive damage to itself. It wasn't going to go away completely, but damned if it wasn't going to suffer like every other company that forgets its origins.
But now, if the appeals exhaust and the DoJ has its way, Microsoft is going to be dismantled by the government. And that's a lot less exciting to watch. It's like watching all sorts of character development in a tragedy-- you know the hero's tragic flaw, and you *see* how his downfall will come-- only to have the hero killed by a bit player. It really *sucks*.
Oh well... enough of my ranting. I'm not trying to advocate a political, social, or economic viewpoint here, I'm just saying that I'm going to be much less amused at Microsoft's slow death because of this. Such is life, I suppose...
I was in second grade when the art teacher asked us to draw the best scarecrow we could think of. It was near Halloween, and there were a number of scary movies out. I had also just received a Ghostbusters toy for my birthday-- a Mummy. Being the weird kid I was, my thought processes were like so:
The purpose of a scare crow is to scare birds away. If you need something to scare away the birds, they obviously are not very easily scared. If it's hard to scare these particular birds away, then the best scarecrow will be the *scariest* scarecrow. There isn't anything scarier than a mummy, so a scarecrow that looks like a mummy would be the best scarecrow possible. QED (Well, I didn't think "QED", it seemed irrefutable to me at the time).
I then proceeded to draw what looked like a mummy with a big stick up his ass stuck out in the plains of Nebraska. My art teacher thought it was a great picture, and put it in my "portfolio" (students were *required* to enter at least two drawings into the art contest each year, both out of his or her "portfolio"). That January, I only had two drawings in my portfolio, and both of them were entered (by default).
The reaction was astounding. I was told that my drawing was "unacceptable" and my parents were told that it was disturbing. I refused to change it (they tried to make me), and my parents wouldn't force me (they told the school to fuck off-- God bless 'em, I couldn't have asked for better!). So, while I was absent from class one day with the flu, the teacher had another student draw *over* my mummy to turn it into a normal scarecrow. The school counselor was then called in, and (without informing my parents) I was told to meet with Ms. Corfield, the disturbingly cheery bitch who passed for a school counselor. To make a long story short, my parents found out and hell broke loose-- I never had to meet with the counselor again!
So we know that schools have been overreacting since before Columbine. I'm not posting it out of any "pity me, I was opressed" sentiment, but to show that schools have historically reacted without thinking through the consequences. It has always been that way.
The heart of the issue is the extent to which the school reacted. Sending a letter home to the parents would most likely be "okay" in this situation-- even though the situation doesn't (IMHO) warrant a letter home, the letter wouldn't wind up on the kid's permanent criminal record. I guess you could call a letter home a "socially acceptable reaction" (SAR). Arresting the kid is a "socially unnacceptable reaction" (SUR).
While I don't think we can stop SARs, I think that we need to at least make schools realize that we won't allow them to continue this pattern of SURs. Showing one school's stupid reactions to a harmless (if ill-written) story is a good story. Maybe a few stories like this will make schools think twice before they have kids arrested on nonexistent grounds.
I thought that the vote was 234 AGAINST the bill, with 122 FOR it. Since it needs a two-thirds majority to pass, 122 *would* be about 1/2 the proper amount (the real number of votes would be 118 or 119). As it is, 234 + 122 = 356, and two thirds of that is 237.3, which we'll round up to get 238. So, actually, it needed four more votes to pass, as a number of people have pointed out-- thanks!
So, based on my (mis) understanding of the article, I came to a mathematically correct conclusion based upon a faulty assumption-- I don't need more math, I need more common sense!
No, a digital signature is *not* just "piece of data". You seem to be confusing the idea of regular signatures with digital signatures.
A regular signature is static-- you put it on one document the same way you put it on another (well, with minor variations, but the overall whole of the thing stays the same). It is difficult for somebody else to imitate it, due to those idiosynchrasies.
However, a digital signature is *not* static. If you digitally sign a document, I cannot create another document and simply "cut and paste" the signature of the other document onto mine-- it would be immediately recognizable as bogus. The reason is that a digital signature *changes* with the contents of the document being signed and with the person signing it. If you and I sign the same document, the signatures are different and unique to each of us. If you sign two different documents, the signatures are different-- and unique to *you*.
If you want to learn more, I'd recommend the original RSA paper, at http://theory.lcs.mit.edu/~rivest/rsapaper.ps , as well as a copy of the DSS (Digital Signature Standard), available from NIST at http://csrc.nist.gov/fips/fips1861.pdf , as well as Bruce Schneier's "Applied Cryptography".
Okay, I think that the article made a little bit more out of the whole CAL thing that is really there. The pricing scheme still sucks, but it's not quite as bad as the article makes it out to be.
One thing, though: $219 for an upgrade from Win9x to Win2K? I get the feeling that this is a bit too much for even Microsoft to be getting away with. And Microsoft knows it.
I figure it this way: By making it look like Windows 2000 is meant only for serious, high-end computing, Microsoft fosters the image that only the technological "in-crowd" will have it (ha!). People will consider Win2K to be a status symbol. Then, blammo, within a couple of months, the price drops from $220 back down to good ol' $89.99. Well, for $89.99, who *wouldn't* want the latest software?
I guess it would be sort of akin to the people who put the latest kernel on their machine, just for the bragging rights-- not because it offers them anything new that they need or want. Except Linux is free anyway, so it doesn't matter too much.
Look out for the bait-and-switch... Microsoft has used it effectively before!
We actually have multiple naming systems. We name hardware development's computers after classic cartoons (Snidely, Dudley, Rocky, Bullwinkle). We name software's computers after people that the softare guys have crushes on (Pam_Lee, Jodi_Foster [that dude worries me...], and my own machine, Barbara_Walters...). When we hook up the machines we're developing to the network, we call 'em all Fred-- Fred1, Fred2, etc. It's boring, but you don't have to worry about accidentally logging onto one of the development machines and screwing it up royally...
Of course, at home, it's a different story. Shakespeare and the Simpsons-- Shakespeare for Linux, Simpsons for NT. Apu, Verona, Moe, and Benvolio. And, no, the NT machines *aren't* mine...
Oh well... I gotta go... Barbara's acting up again... she's such a naughty girl...
Hm... I think Linux has great potential in the area of "information appliances". Linux has a number of distinct advantages:
It's portable. Linux will most likely run on any of the systems that Gateway comes out with, as long as they don't ship Windows-only hardware with the boxes. Any portability issues can be resolved quickly, in-house, for the simple reason that the companies will have the source code.
It's cheaper. How can you undercut free? I don't see any companies trying to *pay* OEMs to use their OS. Though BeOS is supposedly free to OEMs that will use it on their products...
It's more stable. Hey, I know. Sometimes Linux crashes, but if these computers are just "information appliances", whose sole purpose in life is Web surfing and e-mail, then it wouldn't be too tough to fine-tune the system until it becomes pretty damn hard to make it go down.
Linux handles networks a hell of a lot better. If your ISP uses some sort of weird system for authentication, Linux probably supports it a lot more easily than Windows. And if, for some ungodly reason, you need to use non-standard protocols to communicate with somebody, you don't need $300 3rd-party add-ons to make the two systems talk.
Linux is just as user-friendly (if you do it right). Gnome, KDE, and a number of other projects make for interfaces that either match Windows (sometimes by looking just like it!), or even surpass it (KDE, WindowMaker, Enlightenment). Once people get past the "dear god, it's not like everything else I've ever fucking used" factor, it will feel like second nature.
I think Linux has quite a potential in the market. Funny that the article didn't mention that.
Face recognition has really cool applications, such as more intuitive and user-friendly interfaces. It also is perceived as a threat due ot the fact that you could (in theory) be recognized just about anywhere that digital video cameras are.
The Internet has allowed us to communicate and exchange ideas in a quicker, more efficient manner than ever before. People wonder about whether geeks are starting to show symptoms "similar to those of autism".
The Mosaic-2000 issue cuts two ways. If it's written well, and used well, it could be *very* helpful in identifying children who *do* need help-- and that's definitely something good. When either one of these conditions is taken away, though, it is likely that kids are going to suffer.
The real reason that geeks react so strongly is because of an inherent belief that exposing kids to more potential pressure is bad. Geeks seem to hold this belief very dear, perhaps because of hideous memories of hellish high school experiences. But geeks aren't the only ones.
All kids are going through hell during their teenage years. They're learning to act independently of their parents. They are worrying about college or work. They are in a pressure cooker as they physically, emotionally, mentally, and often academically mature. It's a tough time. And a system like Mosaic, when not used properly, puts *all* these kids at risk.
So the question is: is the chance of "healthy" kids suffering bad enough to overshadow the chance of "unhealthy" kids getting help?
Some say yes-- any chance of suffering is large enough to cancel out the benefits of the latter. Others say no-- we can minimize the risk of the former.
Since the system is being deployed, I think that it's almost a moot point. Apparently, enough people say that the potential for bad does not outweigh the potential for good. For those who disagree, we've entered the realm of damage control-- how do we minimize the chance of a particular kid being forced to suffer unjustly?
Well, for starters, there's education. Let the administrators know that the software is *not* a piece of screening software to be used for labelling and pigeonholing (is that a word?) students. Secondly, we need to verify the validity of the software. Now, I doubt Mosaic-2000 is gonna be opensourced, but we can at least perform periodic statistical tests. If the numbers fall outside an agreed-upon "acceptable" range, then the software is scrapped until the numbers become "acceptable" again.
Listen, I'm not saying I'm a supporter of this software. I wouldn't want it deployed if I could stop it. But it's happening. So we need to do our best to reduce the chances of the software's misuse. Maybe we can at least maximize the good while we minimize the bad.
You have stated, time and again, that while picking a good cryptographic algorithm with an adequate key length is important to security, it is only one link in the chain. There are numerous examples of this, including the attacks on Netscape's PRNG's and attacks against smart cards that measure power consumption, timing, etc. to determine the key. Any one of these methods can effectively render the rest of the system useless.
Now for the question: what do you think is the most overlooked aspect of designing a secure system? For example, PRNGs, ineffective key management, mismanaged trust, bad authentication, etc... What can people writing software do (aside from peer review, which is a *must*) to reduce the risks of common problems?
Note: Statements below are not neccessarily my opinions-- they're just illustrations.
The Internet has allowed us to communicate more and more with other people, through forums like Slashdot. Questions are now raised over privacy, censorship, sexually explicit or violent material, and the social impacts that the Internet has on geeks (are we all hermits?)
Guns allowed people a method to protect their homes against criminals. Guns have also allowed criminals the ability to use deadly force while committing crimes.
Radio brought into existence an entirely new area of journalism, allowing information to be spread nearly instantaneously through the airwaves. The same journalists also brought close-ups of grieving families in the post-Columbine hysteria, inciting a nation towards a series of knee-jerk reactions.
I can see face recognition as a cool area of study. It has some cool applications-- sitting down in front of a computer and having it "recognize" you would be cool. But, as with guns, radio, and the Internet, there's a downside. The technology can be used for things that people don't like-- including the invasion of privacy.
I had searched the patent database for something simple and ridiculous to have a patent on. I skimmed over the entry that was labeled (quite simply) "Wheel barrow". I skimmed a little too quickly, and I screwed up...
My apologies. I don't wish to be perpetuating the flow of incorrect information, and I was *seriously* out of line with that one-- moderators take note, the information *is not accurate*. I will *definitely* be more careful from now on...
Because the idea has been around longer than the patent. For example, the wheel barrow has been patented as recently as 1990 (see http://patent.womplex.ibm.com/details?pn=US0492130 5__ ). Can the gentleman who patented it go to all the companies who manufacture wheel barrows and tell them that the wheel barrow is his intellectual property? No-- the wheel barrow has been in use for hundreds of years. It would never hold up in court.
The same situation applies for "retrieval and organization" of data from a database. Hell, that would mean that ATM's are older than terminals-- technically, using vi to view a textfile via a telnet session is retrieving an organized piece of information from a database.
This *is* slightly different, however, from the LZW/gif compression thing. The patent existed *before* it was used for gifs-- so the company (even though they sat on the patent) can still enforce it. Just one of the idiosynchrasies (sp?) of patents...
But, then again, IHNLE (I have no legal experience).
Somebody pointed out that Unisys used a bait-and-switch technique; they didn't start enforcing their patent until it had already become an established standard and there were *TONS* of programs that used the LZW compression, etc... By allowing these companies to do this, did Unisys temporarily abandon their patent? I know it's possible to lose trademarks and copyrights for stuff like this, but what about patents? I can't really see a reasonable defense-- it would be kind of hard to claim that Unisys never knew about Netscape, Gimp, Microangelo, and a million other programs...
So my question is this: Can it be contested that Unisys abandoned their patent, and therefore gave up (some of) their rights to it?
When I worked for a company that was contracted for government work, I was informed that my job was to "make the TAMO Chip work". When I asked what a "TAMO Chip" was, I was surprised by the answer. The engineers had chosen a very cheap, incapable, and badly designed microcontroller to handle a horrendously complex task, and I was to write the firmware. "TAMO" stood for "Then A Miracle Occurs". Fascinating stuff, government contracting...
Putting issues of whether or not it's a good idea for kids to have net access, there is already a group like this in the Baltimore area. I commented on this a while back, but I was so far down in the comment list, I'm not sure anybody saw it.
The group is called "Geeks Into The Streets", and the homepage is at http://linux.umbc.edu/ You can follow the links from there.
If we can keep people from treating the internet as an electronic babysitter, I don't think bringing internet access to kids will be much of a problem-- who knows, if we hook 'em up with Linux boxes, we might raise an entire generation of SysAdmins!
Right now, it's only in the Baltimore, MD area. That's where the UMBC LUG (which is basically sponsoring the project) is located.
Currently, they're working on the Agape House-- an inner-city Baltimore after school club. Anybody have some other projects they want to suggest?
The ironic part is, a lot of groups seem to think that by giving kids a copy of MS Word, they are doing them a favor by making them learn the most well-known word processor. But when you give a kid Linux, he or she will learn System Administration, Programming, User Management, Networking, and Configuration.
Anyway, check out Geeks Into The Streets. It's a pretty neat bunch of guys, IMHO.
Let's face it: Microsoft only stregths are (a) its ruthlessness (b) its ability to make money any which way they can and (c) its ability to churn out evil FUD and buggy program.
Your point is?
Lots of companies are ruthless. "Playing nice" may make the world a better place overall, but that's not what business is about. Business is about money.
Being able to make money any which way you can is not something that counts against Microsoft. It's something we should look up to. If the same tactics were to be used against Microsoft, the company doing so would be called "creative" or at most, "aggressive".
And what's wrong with being ruthless? When somebody is being ruthless, he is being self-protective at the expense of others. Just like a cornered dog biting at anybody who comes too close. Is this something we should be upset over? No-- it's natural.
So what if MS writes buggy software? Nobody *has* to use MS, just like nobody *has* to use Linux. The whole "pre-installation monopoly" bit is bull. Most people would buy Windows anyway, even if it *weren't* pre-installed.
And FUD? It's not like your rant about the MS monopoly isn't just as bad.
The movie was... decent. It was spooky. I was definitely edgy while watching it-- kinda nervous, but not scared.
The movie was, I'll admit, absolutely predictable. It followed a standard cookie-cutter "hey-let's-make-a-horror-flick" template. But let's be fair-- all scary movies do that.
The ending sucked. It just gave no closure. And not only was it the most predictable part of the movie, it was the least well-done. Of course you expect to see Mike standing in the corner just before you hear a "THUD" and see the camera fall to the floor.
I will say, though, that the movie was better than most horror flicks. Why? Because no matter how unbelievable the characters (there were a few times when they became really campy), I somehow felt a connection to them that was stronger than with any other movie.
Is the connection there because of the first-person perspective? Maybe, maybe not. Is it because of the acting? A little. Is it because of the fact that there isn't a shitload of eerie pipe organ music and a bunch of guys with weapons? Perhaps.
I can't recommend this movie to anybody but a few people who really like horror flicks anyway. The film makes a single, novel innovation. After that, it's just a horror flick.
I'd like to see some more films of this type, but BWP is sort of a one-time only deal. Which sucks-- this movie could have been great. It just needed a little more work-- maybe the same guys who came up with the idea of doing it all in first person should have done a little bit more of the same with their script.
I was discussing this issue with a friend of mine not too long ago.
One interesting argument is this: Consider a movie theatre owner. At the front door, he charges admission (or makes people watch advertisements, or whatever). However, his theatre has a side door that is always unlocked. The theatre owner does have the right to post signs telling people not to come in through the side door-- it may not prevent people from coming in, but he can at least prosecute those who do.
This is all well and good for theatre owners.
However, there is a major distinction between theatres and the Web, and the distinction is this: the Web is a public forum.
When placing a work in a public forum, you are the one responsible for restricting access to that work. Plain and simple.
I understand UP's position, but it's their responsibility to restrict access to the site, not the responsibility of everybody else on the web.
Best wishes to MovieList-- stick with it, this is an important issue!
Slashdot Mirror
I know I posted late in the game. I know I have almost *no* chance of being moderated up. But I don't want a whole lot for Christmas. It's not my type of thing, y'know?
Anyway, my mother asked me the other day: "What do you want for Christmas? Something for your computer? What about books?"
Now, granted, a 128 meg DIMM would rock. And the Camel book would be cool. But they always make me feel so-- empty. I don't feel so much accomplishment in them.
Rather, I'd have old tech or something to create with. I would absolutely *adore* a blank book-- I can write anything in one of those; from debugging notes on my latest project to (horrid) poems that I occasionally set to paper.
If you must get me something math- or science-related, get me a slide rule-- I *adore* them. Or an adding machine, I've been wanting one of those. Or even an abacus. If you want to give me something *really* special, just buy me the beads that I can use to *make* a good abacus. Ye Gods, I would *love* the person who did that.
If you must get me something for my computer, don't get anything that requires some sort of installation. I have no PCI slots left, no ISA slots to begin with (!). All three of my memory slots (it's a cheap-ass motherboard, okay?) have 64 meg DIMMS-- I'm happy. I have a bigass hard drive, more than I'll ever need. Software has to be *for* *Linux*, but you'll likely *never* get that right. In other words, buy me a keyboard. PS/2 (yeah, yeah, quit yer griping, I told you it was a cheapass mobo). That's $15 for a decent one (no ergonomic stuff); not bad.
But please, don't get me anything. Call me instead. Come on over, we'll drink egg nog and work on something together (screenplays to pseudorandom number generators-- we can spend time together doing it). That's what a good friend is, and that's what I'd do for you.
One reason that people use stream ciphers is often a speed factor-- it's much faster to simply XOR the data stream with the keystream from a stream cipher than it is to actually involve the plaintext in the permutation. This is especially true if you implement the stream cipher on, say, a separate chip.
Given the fact that public-key ciphers are generally slower than symmetric ciphers, the main reason to use a stream cipher is pretty much defeated. Why not just use a pulic-key cryptosystem to send the key, then send the rest of the message encrypted with a conventional stream cipher?
Also, cryptologists prefer to work on things that have practical qualities (well, *most* do). Perhaps a public-key stream cipher hasn't been designed because there is very little practical advantage to doing so. Or maybe some of the major cryptologists out there haven't even considered the implications.
As part of your thesis, perhaps you could give some reasons and legitimate uses for a stream cipher. Are there any protocols where this would be useful? Are there applications that would be greatly helped by this? Could the design of a public-key stream cipher give new insights into the mathematical workings of other public-key ciphers such as ElGamal and RSA? Can you show that there is a legitimate reason to use a PKSC instead of key exchange, followed by a conventional stream cipher?
I think it'd be interesting to see a paper on this subject. If you do include a section on this topic as part of your thesis, I think people would be *very* interested in seeing it. Go for it!
Best of luck!
The way Gore said it-- "...be worth *two* trillion dollars..."-- makes it sound like Gates is worth *one* trillion right now.
Dan Quayle may need some spelling lessons, but Gore needs a lesson in basic mathematics-- perhaps he could use a refresher course on scientific notation... Or counting, since CTRL-ALT-DEL is a *three* key combination, last time I checked...
Then again, he said that the whole "creating the Internet" thing was due to a "lack of sleep". Perhaps he'll finally fess up and attribute these screw ups to "lack of clue".
"The purposes for the Open Benchmark were:
To confirm that Mindcraft's previous testing was unbiased and representative of Linux's performance."
It seems to me that their main intent was to validate their results. If the specific intent is to validate previous results, then I will guarantee that bias exists. Will Mindcraft publish results that validate the suspicions that they were biased from the start? No-- that would totally destroy the shreds of credibility they have left.
But I'm not going to totally dismiss the results, either. They followed a definite process, and we can't simply ignore what they say because we don't like it. I think that Microsoft *does* have some areas where it beats Linux-- we need to work on these. Mindcraft is only bringing it to the public eye.
But if Mindcraft is bring problems with Linux into the public eye, that means that developers can show why Linux *is* better. Bug fixes are going to pour in. Changes will be made, enhancements created, performance tweaks and changes will make their way into the kernel and into Apache. Performance will improve. The developers will gain more experience, allowing them to enhance their creations even more.
Damn, it's almost like they're doing us a favor!
From http://microsoft.aynrand.org
"...Microsoft, whose trailblazing Windows is the operating system for most personal computers today..."
I am totally against the DoJ on this issue. I'm not a libertarian (as far as I know), but I do tend to think that the best way to make a marketplace better is to make it more free. I also believe that lawsuits like that of the DoJ are symptomatic of a society which wishes to take no responsibility-- nobody was *forced* to enter into a contract with Microsoft, and nobody was *forced* to use Internet Explorer. I'm not going to stay quiet on these issues-- I hold my freedom too dear to not say anything.
But no matter *what* I do, I will *never* support the people over at microsoft.aynrand.org. They spend as much time putting Gates up on a pedestal as they do trashing the DoJ. They seem to think that, if they support Microsoft's right to do business, then they must show Microsoft to be an upstanding and wonderful company. This isn't true-- Microsoft isn't a good company; their software sucks and their prices are inflated.
Put it in perspective: if KKK members were to be banned from demonstrating or exercising their rights to free speech, I would be upset. But that doesn't mean I would put Klan members up on a pedestal and say that they are pillars of society-- that would be bullshit!
Perhaps a better paper to read is this one, which was written (surprisingly) by Eric S. Raymond-- yes, that's right, the same guy who wrote "The Cathedral and the Bazaar". I don't totally agree with this one, either, but at least it doesn't try to make Bill Gates into some sort of demigod while trashing the DoJ until it looks like pure evil.
Besides that, what does product quality have to do with defending the company? You should fight for the freedom of *any* entity whose freedom is being taken away unjustly-- even those that say despicable things or make atrocious products. It doesn't matter-- you should be willing to stand up for freedom.
Okay, I have problems with antitrust law-- I don't consider them right. But I also won't hold up Bill Gates as a visionary who creates "trailblazing" operating systems (http://microsoft.aynrand.org -- ugh!). But I'm going to put aside the moral and political issues for the moment and say this:
I really wanted to see Linux kick Microsoft's ass. I wanted to see Linux, NetBSD, OpenBSD, FreeBSD, and other open source alternatives (even GNU/Hurd) take the place of Windows. And I think it would have happened.
Let's take a look at the following: Western Union was in the business of communications. Messages were telegraphs sent via morse code over copper wires. Along came this goldurned telephone thingy-- it couldn't *possibly* compete with telegraphs! It was an interesting toy, but it had no practical value, and nobody wanted to use it. But telephones are now so commonplace that you have very few situations in which you need to use Western Union. So now WU isn't so big. Why? Because they forgot-- they were in the communications industry, not the telegraph industry. They forgot what business they were in.
Or how about railroads? They're still a big method of transportation. It's still profitable to be in the railroad biz. But not as profitable as it used to be. Why? Because the railroad companies thought that they were in the railroad business-- not the transportation industry. Along came trucks. Trucks were extremely regulated in the beginning, due in part to the legislation and lobbying done by the railroads. But trucking won out-- it's a more common mode of transportation today. The railroads lost because they forgot what business they were in.
Now look at Microsoft. They're in the software business, right? Well, that's the business they *could* be in. But they're in the Windows business, not software. And this Linux thing, this Open Source stuff-- it's the equivalent of trucking to railroads, of telephones to telegraphs. While railroads and telegraphs are still used to a small extent today, telephones and trucks are more common.
Microsoft was on its way to doing massive damage to itself. It wasn't going to go away completely, but damned if it wasn't going to suffer like every other company that forgets its origins.
But now, if the appeals exhaust and the DoJ has its way, Microsoft is going to be dismantled by the government. And that's a lot less exciting to watch. It's like watching all sorts of character development in a tragedy-- you know the hero's tragic flaw, and you *see* how his downfall will come-- only to have the hero killed by a bit player. It really *sucks*.
Oh well... enough of my ranting. I'm not trying to advocate a political, social, or economic viewpoint here, I'm just saying that I'm going to be much less amused at Microsoft's slow death because of this. Such is life, I suppose...
Just my $0.02
Just not so severely. Let me give you an example.
I was in second grade when the art teacher asked us to draw the best scarecrow we could think of. It was near Halloween, and there were a number of scary movies out. I had also just received a Ghostbusters toy for my birthday-- a Mummy. Being the weird kid I was, my thought processes were like so:
The purpose of a scare crow is to scare birds away. If you need something to scare away the birds, they obviously are not very easily scared. If it's hard to scare these particular birds away, then the best scarecrow will be the *scariest* scarecrow. There isn't anything scarier than a mummy, so a scarecrow that looks like a mummy would be the best scarecrow possible. QED (Well, I didn't think "QED", it seemed irrefutable to me at the time).
I then proceeded to draw what looked like a mummy with a big stick up his ass stuck out in the plains of Nebraska. My art teacher thought it was a great picture, and put it in my "portfolio" (students were *required* to enter at least two drawings into the art contest each year, both out of his or her "portfolio"). That January, I only had two drawings in my portfolio, and both of them were entered (by default).
The reaction was astounding. I was told that my drawing was "unacceptable" and my parents were told that it was disturbing. I refused to change it (they tried to make me), and my parents wouldn't force me (they told the school to fuck off-- God bless 'em, I couldn't have asked for better!). So, while I was absent from class one day with the flu, the teacher had another student draw *over* my mummy to turn it into a normal scarecrow. The school counselor was then called in, and (without informing my parents) I was told to meet with Ms. Corfield, the disturbingly cheery bitch who passed for a school counselor. To make a long story short, my parents found out and hell broke loose-- I never had to meet with the counselor again!
So we know that schools have been overreacting since before Columbine. I'm not posting it out of any "pity me, I was opressed" sentiment, but to show that schools have historically reacted without thinking through the consequences. It has always been that way.
The heart of the issue is the extent to which the school reacted. Sending a letter home to the parents would most likely be "okay" in this situation-- even though the situation doesn't (IMHO) warrant a letter home, the letter wouldn't wind up on the kid's permanent criminal record. I guess you could call a letter home a "socially acceptable reaction" (SAR). Arresting the kid is a "socially unnacceptable reaction" (SUR).
While I don't think we can stop SARs, I think that we need to at least make schools realize that we won't allow them to continue this pattern of SURs. Showing one school's stupid reactions to a harmless (if ill-written) story is a good story. Maybe a few stories like this will make schools think twice before they have kids arrested on nonexistent grounds.
Any other suggestions on stopping SURs?
I thought that the vote was 234 AGAINST the bill, with 122 FOR it. Since it needs a two-thirds majority to pass, 122 *would* be about 1/2 the proper amount (the real number of votes would be 118 or 119). As it is, 234 + 122 = 356, and two thirds of that is 237.3, which we'll round up to get 238. So, actually, it needed four more votes to pass, as a number of people have pointed out-- thanks!
So, based on my (mis) understanding of the article, I came to a mathematically correct conclusion based upon a faulty assumption-- I don't need more math, I need more common sense!
No, a digital signature is *not* just "piece of data". You seem to be confusing the idea of regular signatures with digital signatures.
A regular signature is static-- you put it on one document the same way you put it on another (well, with minor variations, but the overall whole of the thing stays the same). It is difficult for somebody else to imitate it, due to those idiosynchrasies.
However, a digital signature is *not* static. If you digitally sign a document, I cannot create another document and simply "cut and paste" the signature of the other document onto mine-- it would be immediately recognizable as bogus. The reason is that a digital signature *changes* with the contents of the document being signed and with the person signing it. If you and I sign the same document, the signatures are different and unique to each of us. If you sign two different documents, the signatures are different-- and unique to *you*.
If you want to learn more, I'd recommend the original RSA paper, at http://theory.lcs.mit.edu/~rivest/rsapaper.ps , as well as a copy of the DSS (Digital Signature Standard), available from NIST at http://csrc.nist.gov/fips/fips1861.pdf , as well as Bruce Schneier's "Applied Cryptography".
Okay, I think that the article made a little bit more out of the whole CAL thing that is really there. The pricing scheme still sucks, but it's not quite as bad as the article makes it out to be.
One thing, though: $219 for an upgrade from Win9x to Win2K? I get the feeling that this is a bit too much for even Microsoft to be getting away with. And Microsoft knows it.
I figure it this way: By making it look like Windows 2000 is meant only for serious, high-end computing, Microsoft fosters the image that only the technological "in-crowd" will have it (ha!). People will consider Win2K to be a status symbol. Then, blammo, within a couple of months, the price drops from $220 back down to good ol' $89.99. Well, for $89.99, who *wouldn't* want the latest software?
I guess it would be sort of akin to the people who put the latest kernel on their machine, just for the bragging rights-- not because it offers them anything new that they need or want. Except Linux is free anyway, so it doesn't matter too much.
Look out for the bait-and-switch... Microsoft has used it effectively before!
We actually have multiple naming systems. We name hardware development's computers after classic cartoons (Snidely, Dudley, Rocky, Bullwinkle). We name software's computers after people that the softare guys have crushes on (Pam_Lee, Jodi_Foster [that dude worries me...], and my own machine, Barbara_Walters...). When we hook up the machines we're developing to the network, we call 'em all Fred-- Fred1, Fred2, etc. It's boring, but you don't have to worry about accidentally logging onto one of the development machines and screwing it up royally...
Of course, at home, it's a different story. Shakespeare and the Simpsons-- Shakespeare for Linux, Simpsons for NT. Apu, Verona, Moe, and Benvolio. And, no, the NT machines *aren't* mine...
Oh well... I gotta go... Barbara's acting up again... she's such a naughty girl...
Hm... I think Linux has great potential in the area of "information appliances". Linux has a number of distinct advantages:
It's portable. Linux will most likely run on any of the systems that Gateway comes out with, as long as they don't ship Windows-only hardware with the boxes. Any portability issues can be resolved quickly, in-house, for the simple reason that the companies will have the source code.
It's cheaper. How can you undercut free? I don't see any companies trying to *pay* OEMs to use their OS. Though BeOS is supposedly free to OEMs that will use it on their products...
It's more stable. Hey, I know. Sometimes Linux crashes, but if these computers are just "information appliances", whose sole purpose in life is Web surfing and e-mail, then it wouldn't be too tough to fine-tune the system until it becomes pretty damn hard to make it go down.
Linux handles networks a hell of a lot better. If your ISP uses some sort of weird system for authentication, Linux probably supports it a lot more easily than Windows. And if, for some ungodly reason, you need to use non-standard protocols to communicate with somebody, you don't need $300 3rd-party add-ons to make the two systems talk.
Linux is just as user-friendly (if you do it right). Gnome, KDE, and a number of other projects make for interfaces that either match Windows (sometimes by looking just like it!), or even surpass it (KDE, WindowMaker, Enlightenment). Once people get past the "dear god, it's not like everything else I've ever fucking used" factor, it will feel like second nature.
I think Linux has quite a potential in the market. Funny that the article didn't mention that.
I've said before: all technology has two sides.
Face recognition has really cool applications, such as more intuitive and user-friendly interfaces. It also is perceived as a threat due ot the fact that you could (in theory) be recognized just about anywhere that digital video cameras are.
The Internet has allowed us to communicate and exchange ideas in a quicker, more efficient manner than ever before. People wonder about whether geeks are starting to show symptoms "similar to those of autism".
The Mosaic-2000 issue cuts two ways. If it's written well, and used well, it could be *very* helpful in identifying children who *do* need help-- and that's definitely something good. When either one of these conditions is taken away, though, it is likely that kids are going to suffer.
The real reason that geeks react so strongly is because of an inherent belief that exposing kids to more potential pressure is bad. Geeks seem to hold this belief very dear, perhaps because of hideous memories of hellish high school experiences. But geeks aren't the only ones.
All kids are going through hell during their teenage years. They're learning to act independently of their parents. They are worrying about college or work. They are in a pressure cooker as they physically, emotionally, mentally, and often academically mature. It's a tough time. And a system like Mosaic, when not used properly, puts *all* these kids at risk.
So the question is: is the chance of "healthy" kids suffering bad enough to overshadow the chance of "unhealthy" kids getting help?
Some say yes-- any chance of suffering is large enough to cancel out the benefits of the latter. Others say no-- we can minimize the risk of the former.
Since the system is being deployed, I think that it's almost a moot point. Apparently, enough people say that the potential for bad does not outweigh the potential for good. For those who disagree, we've entered the realm of damage control-- how do we minimize the chance of a particular kid being forced to suffer unjustly?
Well, for starters, there's education. Let the administrators know that the software is *not* a piece of screening software to be used for labelling and pigeonholing (is that a word?) students. Secondly, we need to verify the validity of the software. Now, I doubt Mosaic-2000 is gonna be opensourced, but we can at least perform periodic statistical tests. If the numbers fall outside an agreed-upon "acceptable" range, then the software is scrapped until the numbers become "acceptable" again.
Listen, I'm not saying I'm a supporter of this software. I wouldn't want it deployed if I could stop it. But it's happening. So we need to do our best to reduce the chances of the software's misuse. Maybe we can at least maximize the good while we minimize the bad.
You have stated, time and again, that while picking a good cryptographic algorithm with an adequate key length is important to security, it is only one link in the chain. There are numerous examples of this, including the attacks on Netscape's PRNG's and attacks against smart cards that measure power consumption, timing, etc. to determine the key. Any one of these methods can effectively render the rest of the system useless.
Now for the question: what do you think is the most overlooked aspect of designing a secure system? For example, PRNGs, ineffective key management, mismanaged trust, bad authentication, etc... What can people writing software do (aside from peer review, which is a *must*) to reduce the risks of common problems?
Thanks!
Note: Statements below are not neccessarily my opinions-- they're just illustrations.
The Internet has allowed us to communicate more and more with other people, through forums like Slashdot.
Questions are now raised over privacy, censorship, sexually explicit or violent material, and the social impacts that the Internet has on geeks (are we all hermits?)
Guns allowed people a method to protect their homes against criminals.
Guns have also allowed criminals the ability to use deadly force while committing crimes.
Radio brought into existence an entirely new area of journalism, allowing information to be spread nearly instantaneously through the airwaves.
The same journalists also brought close-ups of grieving families in the post-Columbine hysteria, inciting a nation towards a series of knee-jerk reactions.
I can see face recognition as a cool area of study. It has some cool applications-- sitting down in front of a computer and having it "recognize" you would be cool.
But, as with guns, radio, and the Internet, there's a downside. The technology can be used for things that people don't like-- including the invasion of privacy.
Anyway, just my $0.02
In the downward direction.
I had searched the patent database for something simple and ridiculous to have a patent on. I skimmed over the entry that was labeled (quite simply) "Wheel barrow". I skimmed a little too quickly, and I screwed up...
My apologies. I don't wish to be perpetuating the flow of incorrect information, and I was *seriously* out of line with that one-- moderators take note, the information *is not accurate*. I will *definitely* be more careful from now on...
Thanks to the folks who pointed that out!
Because the idea has been around longer than the patent. For example, the wheel barrow has been patented as recently as 1990 (see http://patent.womplex.ibm.com/details?pn=US0492130 5__ ). Can the gentleman who patented it go to all the companies who manufacture wheel barrows and tell them that the wheel barrow is his intellectual property? No-- the wheel barrow has been in use for hundreds of years. It would never hold up in court.
The same situation applies for "retrieval and organization" of data from a database. Hell, that would mean that ATM's are older than terminals-- technically, using vi to view a textfile via a telnet session is retrieving an organized piece of information from a database.
This *is* slightly different, however, from the LZW/gif compression thing. The patent existed *before* it was used for gifs-- so the company (even though they sat on the patent) can still enforce it. Just one of the idiosynchrasies (sp?) of patents...
But, then again, IHNLE (I have no legal experience).
Somebody pointed out that Unisys used a bait-and-switch technique; they didn't start enforcing their patent until it had already become an established standard and there were *TONS* of programs that used the LZW compression, etc...
By allowing these companies to do this, did Unisys temporarily abandon their patent? I know it's possible to lose trademarks and copyrights for stuff like this, but what about patents?
I can't really see a reasonable defense-- it would be kind of hard to claim that Unisys never knew about Netscape, Gimp, Microangelo, and a million other programs...
So my question is this: Can it be contested that Unisys abandoned their patent, and therefore gave up (some of) their rights to it?
When I worked for a company that was contracted for government work, I was informed that my job was to "make the TAMO Chip work".
When I asked what a "TAMO Chip" was, I was surprised by the answer. The engineers had chosen a very cheap, incapable, and badly designed microcontroller to handle a horrendously complex task, and I was to write the firmware. "TAMO" stood for "Then A Miracle Occurs".
Fascinating stuff, government contracting...
The group is called "Geeks Into The Streets", and the homepage is at http://linux.umbc.edu/ You can follow the links from there.
If we can keep people from treating the internet as an electronic babysitter, I don't think bringing internet access to kids will be much of a problem-- who knows, if we hook 'em up with Linux boxes, we might raise an entire generation of SysAdmins!
Geeks Into The Streets!
.sig-- it's buggy!)
http://linux.umbc.edu/gits/
Right now, it's only in the Baltimore, MD area. That's where the UMBC LUG (which is basically sponsoring the project) is located.
Currently, they're working on the Agape House-- an inner-city Baltimore after school club. Anybody have some other projects they want to suggest?
The ironic part is, a lot of groups seem to think that by giving kids a copy of MS Word, they are doing them a favor by making them learn the most well-known word processor. But when you give a kid Linux, he or she will learn System Administration, Programming, User Management, Networking, and Configuration.
Anyway, check out Geeks Into The Streets. It's a pretty neat bunch of guys, IMHO.
(Ignore the
Your point is?
Lots of companies are ruthless. "Playing nice" may make the world a better place overall, but that's not what business is about. Business is about money.
Being able to make money any which way you can is not something that counts against Microsoft. It's something we should look up to. If the same tactics were to be used against Microsoft, the company doing so would be called "creative" or at most, "aggressive".
And what's wrong with being ruthless? When somebody is being ruthless, he is being self-protective at the expense of others. Just like a cornered dog biting at anybody who comes too close. Is this something we should be upset over? No-- it's natural.
So what if MS writes buggy software? Nobody *has* to use MS, just like nobody *has* to use Linux. The whole "pre-installation monopoly" bit is bull. Most people would buy Windows anyway, even if it *weren't* pre-installed.
And FUD? It's not like your rant about the MS monopoly isn't just as bad.
***************************
******SLIGHT SPOILERS******
***************************
The movie was... decent. It was spooky. I was definitely edgy while watching it-- kinda nervous, but not scared.
The movie was, I'll admit, absolutely predictable. It followed a standard cookie-cutter "hey-let's-make-a-horror-flick" template. But let's be fair-- all scary movies do that.
The ending sucked. It just gave no closure. And not only was it the most predictable part of the movie, it was the least well-done. Of course you expect to see Mike standing in the corner just before you hear a "THUD" and see the camera fall to the floor.
I will say, though, that the movie was better than most horror flicks. Why? Because no matter how unbelievable the characters (there were a few times when they became really campy), I somehow felt a connection to them that was stronger than with any other movie.
Is the connection there because of the first-person perspective? Maybe, maybe not. Is it because of the acting? A little. Is it because of the fact that there isn't a shitload of eerie pipe organ music and a bunch of guys with weapons? Perhaps.
I can't recommend this movie to anybody but a few people who really like horror flicks anyway. The film makes a single, novel innovation. After that, it's just a horror flick.
I'd like to see some more films of this type, but BWP is sort of a one-time only deal. Which sucks-- this movie could have been great. It just needed a little more work-- maybe the same guys who came up with the idea of doing it all in first person should have done a little bit more of the same with their script.
I was discussing this issue with a friend of mine not too long ago.
One interesting argument is this: Consider a movie theatre owner. At the front door, he charges admission (or makes people watch advertisements, or whatever). However, his theatre has a side door that is always unlocked. The theatre owner does have the right to post signs telling people not to come in through the side door-- it may not prevent people from coming in, but he can at least prosecute those who do.
This is all well and good for theatre owners.
However, there is a major distinction between theatres and the Web, and the distinction is this: the Web is a public forum.
When placing a work in a public forum, you are the one responsible for restricting access to that work. Plain and simple.
I understand UP's position, but it's their responsibility to restrict access to the site, not the responsibility of everybody else on the web.
Best wishes to MovieList-- stick with it, this is an important issue!
After all, it's only natural to be fashionable.