The worms are as likely to become resistant to bacillus toxins if they are on the plant as they are if they are in the plant. This isn`t an issue. In fact, they`re more likely to become resistant with the organic growers applying it to the plants, as they have to use so much more of it. Actually, intermittent massive doses are harder to develop a resistance for than constant low doses. A gene that confers partial resistance is a constant selective advantage when the dose is low, but has less effect in a massive die-off. It's a similar principle to that of multi-drug therapy for AIDS. Moreover, recent studies have shown negative health effects to some organic workers who are allergic to BT. Since BT is all washed off by the time anything hits the shelves, that's only a minor concern currently; but plants with a BT gene are suffused with low levels of the stuff, so consumers have to worry. For an effect like this, "safety testing" is a joke. It would take decades for us to realize that BT (for instance) was helping to cause later-life diabetes in the 5% of infants allergic to it.
Microsoft Passport is the first example of a "megaservice" available for businesses from Microsoft. Megaservices are a new breed of service that extend the Web's capabilities by directly linking applications, services and devices with one another over the Internet. Passport provides online merchants with customizable and easy-to-implement e-commerce services that improve their customers' experience by streamlining purchasing and simplifying registration across a network of Web sites and services.
I wonder what's next... they'll keep your windows registry remotely so software vendors can check for compatibility? Within-site search engines? Web-advertising bundling and sales?
This is exciting, but all the specifics will have to change before you get a practical application of this tech.
1. Manufacturability: These guys have connected 3 molecules of porphyrin to "conducting arms" and to a fourth "custom modified" porphyrin with a buckyball on it. That's a whole lot of custom reactions. Even if you could run these reactions in a vat instead of a test tube, there have to be at least a few of them where the maximum theoretical yield is pretty low.
2. Efficiency: as the article explained, real photosynthesis has to hand the electrons off many times before it can get useful work out of them. Things may get easier when you want electrons out the end instead of ATP, but this is still only the first step. And even in this step, you've already lost a lot of the energy. The reason a buckyball ion is so stable is that the charge distributes over 60 atoms of carbon. The electron is pretty happy there - it doesn't have a whole lot of oomph left to power your [wearable beowulf cluster].
Sure, this reaction may help us understand the chemistry of modified photosyntheses, but in the long run, I'd bet that the first green photocells will crib a lot more from life than just one part of one molecule. In other words, the pure chemists have to start talking to the genetic engineers a lot more than these ones have.
Someone had to say it. Where I come from, when people get a little older and suddenly stop being so revolutionary, there's a word for that. And whether or not you actually think it's a bad thing to "sell out" like that, it sure as hell is pretty normal. Not something that should be analyzed as if it's a shocking development.
The fact that this critic admits he hasn't read the "newer" writers relevant to his point (including, coincidentally, all three of the women he mentions as potentially within the scope of his analysis. Well, 3/4 if you count Pat Cadigan, who he briefly mentions) also calls his conclusions into question. As does his questioning of the term "class" because of the vast social mobility within the US (actually, the gaps between haves and have-nots have been widening steadily over the last 20 years, AND percentile-wise social mobility has slowed even after you correct for the age-distribution effect of stagnant boomers.)
(oops, cut myself off there, in the middle of my disclaimer.)
I was saying that these are complicated ideas, and in order to summarize I've simplified a little.
Feel their pain, too...
on
The Cat Cam
·
· Score: 2
In a related development, neuroscientists found that if they simultaneously stuck sharp electrodes under the skin of a human and of a cat, the sensations produced were remarkably similar.
Seriously - I find this to be a very disturbing study. I'm not going to cop a holier-than-thou attitude here; I myself have done painful neurological experiments on living rats for no better reason than to satisfy a course requirement. And scientifically, this study is tricky work, well done. However, you really have to question the motives here. This is the ultimate in gee-whiz graphics, and by posting it on slashdot for a bunch of non-neuroscientists to ooh and ahh at, we're really buying into that.
Donna Haraway has written about the importance of vision in science (Before you dismiss her as just another of those postmodern feminist theorists who of course don't know what they're talking about, like Randy's overstereotyped ex in Necronomicon, you should read her "Cyborg Manifesto"). She's arguing that no picture, not even natural vision, is as direct and honest as it seems to be. As science strives (rightly) for true objectivity, it tends towards several kinds of false objectivity. Pictures are one such method. They seem natural - it's easy for the layman to say "Oh, that's what a cat sees," rather than "That's a picture constructed by scientists as an educated guess about what a cat sees". This is not as bad as the false objectivity that says "Who are you to challenge my objectivity? You're not white and male, so whatever you say is obviously biased.", but it's a step in the wrong direction.
The way to fight this kind of false objectivity is to appreciate what went into making the image. When you see something, appreciate that it's actually being projected upside down on your retina, multiplied by three color/response curves, broken down into shapes and movement, and so on; the mechanisms of your vision bias you to pay attention to some attributes and unavoidably ignore others. And when you see a picture, appreciate how constructed that is. In this case, that means that the picture came to you through a very unhappy cat, into some electrodes, through some statistical software, and through the web. Anyone who can look at these pictures without feeling some real suffering in sympathy with that cat is not truly objective enough to be a scientist.
The saving grace of the open source community is that there are no deep pockets to sue. You either sue to damage a competitor or to line your pockets. The open source community is too amorphous to be seriously damaged by a lawsuit - who do you slap the restraining order on? - and too nonmonetary to line someone's pockets. Even if you sued Red Hat, you couldn't touch their major revenues (support and VC) except as punitive damages... and the chances of getting a sane judge would just be too great to risk going so far into uncharted legal ground.
The crucial distinction made in this article is between "excessive inertia" and "reasonable inertia". The question is, as an already-skilled qwerty typist, would you be selfishly happy if the rest of the world made the switch? Obviously, the generations of unborn children would be happy, because Dvorak is acknowledged to be better. But the inertia can still be "reasonable" if the average skilled typist's investment in relearning is not repaid by a significant margin. Inertia's only "excessive" if everyone wishes they could switch (as opposed to wishing that they'd made the right choice the first time) but is just waiting for the other guy to do it first.
Obviously, this article argues largely from a lack of evidence, and so doesn't prove that the inertia's not excessive. But its thesis is at least plausible; I know a lot of people who would not be happy if the world switched to Dvorak. Can you imagine how your average user would scream if suddenly they had to learn how to type all over again? And it's not just that they don't know what's good for them. If you're never going to be an excellent typist, the hassle of relearning the layout probably won't ever repay itself.
Visio is most famous in the Seattle area for their bus-side ads looking for programmers. They scream: 43 guitar players 13 marathon runners 0 pocket protectors I was always tempted to go for an interview wearing a pocket protector... but the fact remains that I was tempted to go for an interview.
I think the Seattle location is definitely a factor in this purchase. And that's a bad sign for the engineering-job ecosystem up here. As soon as a Seattle-area software company gets successful, M$ makes them an offer they can't refuse. Whether or not you hate microsoft in particular, it's sad that yet another unique culture is being "embraced". That's one option fewer for the Puget Sound software engineer.
I'll be wearing my pocket protector to work tomorrow in mourning.
Who needs a hardware add-on for calculator capability? As long as the heaviest-duty thing you want is static 2d graphs, there's no need for a math coprocessor. There are already over 100 calculator apps available for palmOS, including Mathgraph and Hgraph for graphing, and the old standby RPN for programmability and script availability.
Unfortunately, given the variety of phone systems and phone handsets, (AFAIK) no product which creates audio DTMF tones to dial a phone works very reliably. Since most non-slashdotters would rather not have a feature than have one which works 90% or less of the time ("This thing's a piece of junk!"), don't expect either Palm or Handspring to bend over backwards to give you this capability. Both companies are too influenced by Jeff "Don't Always Listen to what the User Thinks they Want" Hawkins, and probably rightly so.
If the speaker gets good enough to do this in software, that's another story. I doubt that the cheapo Visor has brought much improvement in sound quality, though (as opposed to volume, which is a far more pressing customer concern...).
From day 1 of handspring, Jeff and Donna have been promising binary compatibility with existing PalmOS apps, and that is still on the menu for the visor. Of course, there will be the odd app that does something unorthodox directly with registers or OS globals (remember, no true memory protection in PalmOS) that will break, but there were already a couple of these when Palm switched from the 68328 in palm III to the 328EZ in V and IIIx.
The greater worry is whether the hotsync and cradles will be compatible. I doubt that it would be flat-out incompatible, but I can imagine many minor headaches when your partner tries to sync their new visor to the same machine you use to sync your Palm V.
With current tech, color sucks batteries, hard. You have to have a backlight running the whole time it's on, and a broader-spectrum one than just the old indiglo-style.
You don't notice it so much on a WinCE device - the overpowered processors in that court already drain your batteries out fast unless you have a bulky heavy beefy battery - but if the only other battery drain were a measly Dragonball 68328 or '328EZ (PalmOS's power), you'd really notice your battery life in the toilet.
The reason the original palm was so good was Jeff Hawkins's ability to say an emphatic "no" to user wish-list checkboxes like this, and hold fast to the essentials - form factor, usability, battery life. Don't expect Handspring to embrace color until it's way more mature.
True, we'd probably want an upper limit on the time and space resources of the programs...
What do you mean "we", white man?:)
Since it's Commander Taco that would do all this, I think we should remember that the idea is to make his life easier. If karma et al are *not* the primitives, what are? I gave my example based on full-text search ("Nothing from someone who's replied to 'hemos sux'") to show two things. First, how ultimately useless it would be. Second, how un-indexable and thus computationally explosive that kind of query power would be.
Now, if we could just use distributed.net to turn all slashdot readers' machines into a giant beowulf cluster... then use *that* to do turing-complete searches on the message-space (encrypted searchably, of course, to protect anonymity)...
Obviously you don't want anything that's actually turing-complete, because you want to be sure that people's filtering scripts always terminate rather than hanging the server (impossible by definition for turing-complete languages).
It's true, sometimes the best answer to the problem is to go general. However, without a certain amount of problem-specific hacking, people's filter scripts won't have any useful data to work with. Without such primitives as "moderated up", "moderated down", and "adjective:funny", what will your script look for? "Show me all the posts from people that have never replied to posts with the words 'hemos sux' in them"?
So, in light of the "more general" philosophy, let me restate ideas 1 and 2: maintain information on the adjectives people have acquired through past moderation and make that info available as a primitive to the filter scripts.
It's a little unclear what the consequences for poor moderation are. If it's just bad karma, that's a pretty weak motivator. How will someone even know that they're getting karma-dinged for unfair moderation? And assuming they continue to make good posts, what is to stop them from continuing their reign of poor moderation?
If the goal is to motivate good moderation, you have to put people's "unfair moderation" rating right in their face (on the main page; most people probably never check users.pl). And if the goal is to prevent poor moderation, there should be a threshhold where they never moderate again, no matter how much good karma they get from posts (Say, if "unfair" mods ever exceed "fair" ones by 3 or more...). IMO, you should do both.
I don't mind reading the stuff that has been explicitly moderated up as "funny". However, there are a few folks who posted a lot of hair-trigger joke responses. That's great; sometimes they score a direct hit. But when they accumulate enough karma to get a default score of 2, all those jokes start to become annoying.
So, what's the solution? More feeping creatures. The question is, which ones feep the least?
-People's karma comes with the adjective that is the "mode" of all their moderation adjectives. Their default-moderation-level posts also acquire this adjective. In your preferences, you can explicitly give a -1 to posts with certain adjectives ("I don't *want* to be amused, dammit!"). Feeps pretty hard, because it requires additions to the already-huge preference screen.
-You can't attain Enlightenment (enough positive karma for a default upgrade) without a mix of adjectives. Someone with 20 "funny" posts wouldn't make it, but someone with 17 "funnies", 2 "interesting", and an "informative" would. Also highly feepish from an implementation POV, but at least pretty transparent to the user.
-Only a certain number of posts a day get your default moderation bonus. If you're posting more than once or twice a day, chances are you're not putting excessive thought/new information into each post. This idea is by far the easiest to implement and the cleanest. It would probably solve the problem, even though it doesn't address it directly.
This looks like a newton-sized device. Unless you're Joe Young, it probably won't fit in your palm. And yes; it does make a difference. Something that fits in a pocket can be with you, always, whenever you're awake. Something that fits in a fanny pack can be back at your desk... let me go check.
"Until and unless new algorithms are found, there are not enough atoms in the universe to factor 4096 bit keys before the universe collapses back into the next big bang."
Using von-neumann computing, this is true. However, there is a quantum computing algorithm to factor large numbers (you take a quantum superposition of all possible numbers, raise a test integer to that power modulo the number you want to factor, do a fast fourier transform on the result, and add one to the periodicity that jumps out of the noise. There's a good chance that you have a factor, and if not, you can try again.). I believe this algorithm has actually been run on an avagadro sample of quantum computers (otherwise known as molecules in an NMR machine) to factor the number 6. Of course, even if QC obeys Moore's law starting now (and I suspect that QC's doubling constant will be longer than 18 months), there are still a good 20 years or so before it gets to the point of factoring 4096-bit keys, and you can't distribute quantum computations to hurry that up any.
Anyway, the eventual number of atoms needed is theoretically 1 per bit in your computer. The algorithm requires three full-width registers, and any reasonable implementation of QC will take at least triple redundancy and error correction, so that's 3*3*4096=36864 atoms. I would be happy to provide this quantity to anyone who asks.
I don't think manned is an option. A 10-mile magnetosphere would probably mess up your passengers at ground zero.
<TROLL VERACITY=TRUE>Besides, given current technology, manned space is a waste of money. That's why the NASA budget cuts are doubly criminal - cutting basic science, but NOT the space station.</TROLL>
What's with/.? Just because Katz has a lot of his own opinions to share with us doesn't mean he shouldn't include a link to the actual policy statement. Is this, or is this not, a web-based news source?
If the above link has already been posted, please moderate the first post containing it upwards.
To keep your window count from going totally out of control, you can keep 2-3 windows open with part of the pane visible at all times. Then, you can drag links you want to visit to the oldest window. Sort of a circular buffer protocol.
However much you do or do not trust the FBI, this is still bad news. If satellite phone companies are willing to give wiretap capabilities to a "legitimate" government in one case, why not other cases? And there are clearly governments out there that have even less regard for life than the FBI has for privacy.
In fact, Iridium originally excited me because it is essentially a democratic technology. It can give information the power to cross borders unsupervised by authorities. Even at 2400 baud, it could send a whole lot of newspaper articles into the hinterlands of China or out of the jungles of Chiapas. If the corporation controlling it is going to start collaborating with governments, I guess that potential will be squashed.
Slashdot Mirror
The worms are as likely to become resistant to bacillus toxins if they are on the plant as they are if they are in the plant. This isn`t an issue. In fact, they`re more likely to become resistant with the organic growers applying it to the plants, as they have to use so much more of it. Actually, intermittent massive doses are harder to develop a resistance for than constant low doses. A gene that confers partial resistance is a constant selective advantage when the dose is low, but has less effect in a massive die-off. It's a similar principle to that of multi-drug therapy for AIDS. Moreover, recent studies have shown negative health effects to some organic workers who are allergic to BT. Since BT is all washed off by the time anything hits the shelves, that's only a minor concern currently; but plants with a BT gene are suffused with low levels of the stuff, so consumers have to worry. For an effect like this, "safety testing" is a joke. It would take decades for us to realize that BT (for instance) was helping to cause later-life diabetes in the 5% of infants allergic to it.
- Microsoft Passport is the first example of a "megaservice" available for businesses from Microsoft. Megaservices are a new breed of service that extend the Web's capabilities by directly linking applications, services and devices with one another over the Internet. Passport provides online merchants with customizable and easy-to-implement e-commerce services that improve their customers' experience by streamlining purchasing and simplifying registration across a network of Web sites and services.
I wonder what's next... they'll keep your windows registry remotely so software vendors can check for compatibility? Within-site search engines? Web-advertising bundling and sales?Here's another URL to the press release.
This is exciting, but all the specifics will have to change before you get a practical application of this tech.
1. Manufacturability: These guys have connected 3 molecules of porphyrin to "conducting arms" and to a fourth "custom modified" porphyrin with a buckyball on it. That's a whole lot of custom reactions. Even if you could run these reactions in a vat instead of a test tube, there have to be at least a few of them where the maximum theoretical yield is pretty low.
2. Efficiency: as the article explained, real photosynthesis has to hand the electrons off many times before it can get useful work out of them. Things may get easier when you want electrons out the end instead of ATP, but this is still only the first step. And even in this step, you've already lost a lot of the energy. The reason a buckyball ion is so stable is that the charge distributes over 60 atoms of carbon. The electron is pretty happy there - it doesn't have a whole lot of oomph left to power your [wearable beowulf cluster].
Sure, this reaction may help us understand the chemistry of modified photosyntheses, but in the long run, I'd bet that the first green photocells will crib a lot more from life than just one part of one molecule. In other words, the pure chemists have to start talking to the genetic engineers a lot more than these ones have.
Someone had to say it. Where I come from, when people get a little older and suddenly stop being so revolutionary, there's a word for that. And whether or not you actually think it's a bad thing to "sell out" like that, it sure as hell is pretty normal. Not something that should be analyzed as if it's a shocking development.
The fact that this critic admits he hasn't read the "newer" writers relevant to his point (including, coincidentally, all three of the women he mentions as potentially within the scope of his analysis. Well, 3/4 if you count Pat Cadigan, who he briefly mentions) also calls his conclusions into question. As does his questioning of the term "class" because of the vast social mobility within the US (actually, the gaps between haves and have-nots have been widening steadily over the last 20 years, AND percentile-wise social mobility has slowed even after you correct for the age-distribution effect of stagnant boomers.)
(oops, cut myself off there, in the middle of my disclaimer.)
I was saying that these are complicated ideas, and in order to summarize I've simplified a little.
In a related development, neuroscientists found that if they simultaneously stuck sharp electrodes under the skin of a human and of a cat, the sensations produced were remarkably similar.
Seriously - I find this to be a very disturbing study. I'm not going to cop a holier-than-thou attitude here; I myself have done painful neurological experiments on living rats for no better reason than to satisfy a course requirement. And scientifically, this study is tricky work, well done. However, you really have to question the motives here. This is the ultimate in gee-whiz graphics, and by posting it on slashdot for a bunch of non-neuroscientists to ooh and ahh at, we're really buying into that.
Donna Haraway has written about the importance of vision in science (Before you dismiss her as just another of those postmodern feminist theorists who of course don't know what they're talking about, like Randy's overstereotyped ex in Necronomicon, you should read her
"Cyborg Manifesto"). She's arguing that no picture, not even natural vision, is as direct and honest as it seems to be. As science strives (rightly) for true objectivity, it tends towards several kinds of false objectivity. Pictures are one such method. They seem natural - it's easy for the layman to say "Oh, that's what a cat sees," rather than "That's a picture constructed by scientists as an educated guess about what a cat sees". This is not as bad as the false objectivity that says "Who are you to challenge my objectivity? You're not white and male, so whatever you say is obviously biased.", but it's a step in the wrong direction.
The way to fight this kind of false objectivity is to appreciate what went into making the image. When you see something, appreciate that it's actually being projected upside down on your retina, multiplied by three color/response curves, broken down into shapes and movement, and so on; the mechanisms of your vision bias you to pay attention to some attributes and unavoidably ignore others. And when you see a picture, appreciate how constructed that is. In this case, that means that the picture came to you through a very unhappy cat, into some electrodes, through some statistical software, and through the web. Anyone who can look at these pictures without feeling some real suffering in sympathy with that cat is not truly objective enough to be a scientist.
[Of course, most
The saving grace of the open source community is that there are no deep pockets to sue. You either sue to damage a competitor or to line your pockets. The open source community is too amorphous to be seriously damaged by a lawsuit - who do you slap the restraining order on? - and too nonmonetary to line someone's pockets. Even if you sued Red Hat, you couldn't touch their major revenues (support and VC) except as punitive damages... and the chances of getting a sane judge would just be too great to risk going so far into uncharted legal ground.
The crucial distinction made in this article is between "excessive inertia" and "reasonable inertia". The question is, as an already-skilled qwerty typist, would you be selfishly happy if the rest of the world made the switch? Obviously, the generations of unborn children would be happy, because Dvorak is acknowledged to be better. But the inertia can still be "reasonable" if the average skilled typist's investment in relearning is not repaid by a significant margin. Inertia's only "excessive" if everyone wishes they could switch (as opposed to wishing that they'd made the right choice the first time) but is just waiting for the other guy to do it first.
Obviously, this article argues largely from a lack of evidence, and so doesn't prove that the inertia's not excessive. But its thesis is at least plausible; I know a lot of people who would not be happy if the world switched to Dvorak. Can you imagine how your average user would scream if suddenly they had to learn how to type all over again? And it's not just that they don't know what's good for them. If you're never going to be an excellent typist, the hassle of relearning the layout probably won't ever repay itself.
Visio is most famous in the Seattle area for their bus-side ads looking for programmers. They scream:
43 guitar players
13 marathon runners
0 pocket protectors
I was always tempted to go for an interview wearing a pocket protector... but the fact remains that I was tempted to go for an interview.
I think the Seattle location is definitely a factor in this purchase. And that's a bad sign for the engineering-job ecosystem up here. As soon as a Seattle-area software company gets successful, M$ makes them an offer they can't refuse. Whether or not you hate microsoft in particular, it's sad that yet another unique culture is being "embraced". That's one option fewer for the Puget Sound software engineer.
I'll be wearing my pocket protector to work tomorrow in mourning.
Who needs a hardware add-on for calculator capability? As long as the heaviest-duty thing you want is static 2d graphs, there's no need for a math coprocessor. There are already over 100 calculator apps available for palmOS, including Mathgraph and Hgraph for graphing, and the old standby RPN for programmability and script availability.
Unfortunately, given the variety of phone systems and phone handsets, (AFAIK) no product which creates audio DTMF tones to dial a phone works very reliably. Since most non-slashdotters would rather not have a feature than have one which works 90% or less of the time ("This thing's a piece of junk!"), don't expect either Palm or Handspring to bend over backwards to give you this capability. Both companies are too influenced by Jeff "Don't Always Listen to what the User Thinks they Want" Hawkins, and probably rightly so.
If the speaker gets good enough to do this in software, that's another story. I doubt that the cheapo Visor has brought much improvement in sound quality, though (as opposed to volume, which is a far more pressing customer concern...).
From day 1 of handspring, Jeff and Donna have been promising binary compatibility with existing PalmOS apps, and that is still on the menu for the visor. Of course, there will be the odd app that does something unorthodox directly with registers or OS globals (remember, no true memory protection in PalmOS) that will break, but there were already a couple of these when Palm switched from the 68328 in palm III to the 328EZ in V and IIIx.
The greater worry is whether the hotsync and cradles will be compatible. I doubt that it would be flat-out incompatible, but I can imagine many minor headaches when your partner tries to sync their new visor to the same machine you use to sync your Palm V.
With current tech, color sucks batteries, hard. You have to have a backlight running the whole time it's on, and a broader-spectrum one than just the old indiglo-style.
You don't notice it so much on a WinCE device - the overpowered processors in that court already drain your batteries out fast unless you have a bulky heavy beefy battery - but if the only other battery drain were a measly Dragonball 68328 or '328EZ (PalmOS's power), you'd really notice your battery life in the toilet.
The reason the original palm was so good was Jeff Hawkins's ability to say an emphatic "no" to user wish-list checkboxes like this, and hold fast to the essentials - form factor, usability, battery life. Don't expect Handspring to embrace color until it's way more mature.
True, we'd probably want an upper limit on the time and space resources of the programs...
:)
What do you mean "we", white man?
Since it's Commander Taco that would do all this, I think we should remember that the idea is to make his life easier. If karma et al are *not* the primitives, what are? I gave my example based on full-text search ("Nothing from someone who's replied to 'hemos sux'") to show two things. First, how ultimately useless it would be. Second, how un-indexable and thus computationally explosive that kind of query power would be.
Now, if we could just use distributed.net to turn all slashdot readers' machines into a giant beowulf cluster... then use *that* to do turing-complete searches on the message-space (encrypted searchably, of course, to protect anonymity)...
Obviously you don't want anything that's actually turing-complete, because you want to be sure that people's filtering scripts always terminate rather than hanging the server (impossible by definition for turing-complete languages).
It's true, sometimes the best answer to the problem is to go general. However, without a certain amount of problem-specific hacking, people's filter scripts won't have any useful data to work with. Without such primitives as "moderated up", "moderated down", and "adjective:funny", what will your script look for? "Show me all the posts from people that have never replied to posts with the words 'hemos sux' in them"?
So, in light of the "more general" philosophy, let me restate ideas 1 and 2: maintain information on the adjectives people have acquired through past moderation and make that info available as a primitive to the filter scripts.
It's a little unclear what the consequences for poor moderation are. If it's just bad karma, that's a pretty weak motivator. How will someone even know that they're getting karma-dinged for unfair moderation? And assuming they continue to make good posts, what is to stop them from continuing their reign of poor moderation?
If the goal is to motivate good moderation, you have to put people's "unfair moderation" rating right in their face (on the main page; most people probably never check users.pl). And if the goal is to prevent poor moderation, there should be a threshhold where they never moderate again, no matter how much good karma they get from posts (Say, if "unfair" mods ever exceed "fair" ones by 3 or more...). IMO, you should do both.
I don't mind reading the stuff that has been explicitly moderated up as "funny". However, there are a few folks who posted a lot of hair-trigger joke responses. That's great; sometimes they score a direct hit. But when they accumulate enough karma to get a default score of 2, all those jokes start to become annoying.
So, what's the solution? More feeping creatures. The question is, which ones feep the least?
-People's karma comes with the adjective that is the "mode" of all their moderation adjectives. Their default-moderation-level posts also acquire this adjective. In your preferences, you can explicitly give a -1 to posts with certain adjectives ("I don't *want* to be amused, dammit!"). Feeps pretty hard, because it requires additions to the already-huge preference screen.
-You can't attain Enlightenment (enough positive karma for a default upgrade) without a mix of adjectives. Someone with 20 "funny" posts wouldn't make it, but someone with 17 "funnies", 2 "interesting", and an "informative" would. Also highly feepish from an implementation POV, but at least pretty transparent to the user.
-Only a certain number of posts a day get your default moderation bonus. If you're posting more than once or twice a day, chances are you're not putting excessive thought/new information into each post. This idea is by far the easiest to implement and the cleanest. It would probably solve the problem, even though it doesn't address it directly.
This looks like a newton-sized device. Unless you're Joe Young, it probably won't fit in your palm. And yes; it does make a difference. Something that fits in a pocket can be with you, always, whenever you're awake. Something that fits in a fanny pack can be back at your desk... let me go check.
"Until and unless new algorithms are found, there are not enough atoms in the universe to factor 4096 bit keys before the universe collapses back into the next big bang."
Using von-neumann computing, this is true. However, there is a quantum computing algorithm to factor large numbers (you take a quantum superposition of all possible numbers, raise a test integer to that power modulo the number you want to factor, do a fast fourier transform on the result, and add one to the periodicity that jumps out of the noise. There's a good chance that you have a factor, and if not, you can try again.). I believe this algorithm has actually been run on an avagadro sample of quantum computers (otherwise known as molecules in an NMR machine) to factor the number 6. Of course, even if QC obeys Moore's law starting now (and I suspect that QC's doubling constant will be longer than 18 months), there are still a good 20 years or so before it gets to the point of factoring 4096-bit keys, and you can't distribute quantum computations to hurry that up any.
Anyway, the eventual number of atoms needed is theoretically 1 per bit in your computer. The algorithm requires three full-width registers, and any reasonable implementation of QC will take at least triple redundancy and error correction, so that's 3*3*4096=36864 atoms. I would be happy to provide this quantity to anyone who asks.
I don't think manned is an option. A 10-mile magnetosphere would probably mess up your passengers at ground zero.
<TROLL VERACITY=TRUE>Besides, given current technology, manned space is a waste of money. That's why the NASA budget cuts are doubly criminal - cutting basic science, but NOT the space station.</TROLL>
What's with /.? Just because Katz has a lot of his own opinions to share with us doesn't mean he shouldn't include a link to the actual policy statement. Is this, or is this not, a web-based news source?
If the above link has already been posted, please moderate the first post containing it upwards.
To keep your window count from going totally out of control, you can keep 2-3 windows open with part of the pane visible at all times. Then, you can drag links you want to visit to the oldest window. Sort of a circular buffer protocol.
(wow, we're way off topic, eh?)
However much you do or do not trust the FBI, this is still bad news. If satellite phone companies are willing to give wiretap capabilities to a "legitimate" government in one case, why not other cases? And there are clearly governments out there that have even less regard for life than the FBI has for privacy.
In fact, Iridium originally excited me because it is essentially a democratic technology. It can give information the power to cross borders unsupervised by authorities. Even at 2400 baud, it could send a whole lot of newspaper articles into the hinterlands of China or out of the jungles of Chiapas. If the corporation controlling it is going to start collaborating with governments, I guess that potential will be squashed.
Palm aces size AND battery life.