Any life would not be on the surface!
on
Europan Life In Doubt
·
· Score: 4, Interesting
Not only doesn't ionizing radiation penetrate solids and fluids very far, but ionizing radiation generates high-energy chemical compounds that any life could potentially use as a food source. It just can't be directly on the surface--but you wouldn't want to be there anyway, since you'd tend to evaporate into space and/or freeze.
Anyway, the ionizing radiation has been known
about for years and is one of the major reasons why scientists thought there might be life!
Possibly...but the description was hopelessly bungled the first time around. For instance, the scaling was called "exponential" and then stated to be N^N, which is superexponential. (N^N=e^(N ln N).)
In any case, simply observing that there are 2^N possible groups doesn't mean that any significant fraction of them come into being. It's the number that actually come into being that are important, and those are limited by how much time people have and human social dynamics.
Advertisers have been tricking our brains for a long time in order to sell products.
Why not juxtapose desirable women and famous athletes with...carbonated sugar water? Yes, that makes a lot of sense. Except our brains are wired to associate things that appear together....
I recently bought a Butterfinger Bar. 10% more free, it said, with a bright green stripe on the end of the bar suggesting just how much more you were getting! Wow, that was another good bite, yum! Except when I measured the stripe, it was 20% of the length of the bar....
I bought a banana too. There was a sticker advertising low-calorie sweeteners on it. That was just weird.
Anyway, point is, advertisers implicitly lie, mislead, deceive all over the place--just not explicitly, because explicit lies are illegal. If they want to spend a bunch of money on fMRI studies so they can be a bit more successful, I say, great! It won't make a huge difference to advertising (and if it does, we might outlaw misleading implications).
But if they dump enough money into fMRI research, they might bring the cost of the machines down, and that would really help with *useful* research into the functioning of the brain. That would be a nice fringe benefit.
This is *exactly* what you'd expect to see. You have fields with pollen right next to each other, try to kill off one side with Roundup, and gee, no wonder the plants who were cross-polinated and picked up the resistance gene started to survive!
However, this doesn't extrapolate to any giant evil nasty consequences. If you want to be able to kill your corn with Roundup, don't give corn a herbicide resistance gene (duh).
However, herbicide resistance genes are kind of stupid, because there *is* some low rate of horizontal gene transfer, and eventually a weed will pick it up too, and then your herbicide won't even be killing weeds. This doesn't show that GM is evil, just that people can use it in shortsighted ways.
Money = flexibility. You have nine months. If flexibility and adventure are important to you, save some now. Whether you end up in the Peace Corps or whatever, it will help give you room to breathe.
For anyone who wants equations to go along with the descriptive posts on event horizons and Schwarzschild radius, said radius is given by
r = 2GM/c^2
where G = 6.67e-11 m^3/s^2*kg (the gravitational constant) and c = 3e8 m/s (the speed of light, of course). Plug in 3 million sun-masses (the sun weighs 2e30 kg), and you have
r = 8.9e9 m = 5.5 million miles = 0.06AU
So unfortunately, the event horizon isn't three times as big as the solar system. The earth's orbit is 1AU (that's how the unit is defined). The event horizon barely stretches past the surface of the sun (7e8 meters)!
UCSD's acceptable use policy doesn't reserve the right of the university to remove sites based on content alone. (For copyright, safety, legality, etc., they do--but not to edit content.) See
UCSD's
web policy.
The letter specifically says that the UCSD policy that has been violated is the prohibition against breaking federal law (i.e. VI.H.i). That makes it a First Amendment issue--they are essentially claiming that this link is illegal. The First Amendment suggests that the link would count as protected speech.
The letter *could* have cited VI.C.ii and said, "We're getting complaints about this site and don't want to deal with it. This is an avoidable incremental cost, so stop it." This would be less directly a First Amendment issue, but it would still directly be a free speech issue. The policy would then be, "You can't say anything objectionable, because someone might object, and we don't want to hear about it."
Besides, since when have private ISPs been required to have fewer restrictions on speech than a public university? Maybe if ISPs were considered "common carriers", but so far they've been decent at dodging that label.
University provides network, doesn't own machine
on
That Link Is Illegal
·
· Score: 2, Informative
A bunch of people seem to be under the impression that UCSD actually owns the machine that burn is hosted on. As far as I know, this is not correct; burn is on a student-hosted machine, but uses the university network for internet access.
(That'll teach me to hit "submit" before "preview".)
No parity, interesting complementarity
on
Parity Code And DNA
·
· Score: 5, Interesting
As the previous poster pointed out, there isn't any parity to speak of in DNA. I have no idea why they're trying to make the comparison.
Hydrogen bonding is a much higher-fidelity error correcting system than parity checking. With parity checking, you can catch any single error (the number of 1's changes), but not any pair. With hydrogen bond donors and acceptors, you have to have an exact match at all positions. Any number of errors ruins the complementarity.
For what it's worth, the base pairing system is quite elegant: due to size constraints, purines (A,G) and pyrimidines (C,T) must pair with each other. Using the article's notation of 0=H-bond acceptor, 1=donor, the four nucleotides are
A = 10_
C = 100
G = 011
T = 101 where _ is simply a hole. The best pairwise complementarity is (evidently) AT and CG.
It's a chemically implemented RAID-1 system, not a parity check.
Well, not without changing the Earth's orbit or rotational velocity. And we can't do that yet.
As long as we inherently have 365ish days per year, we're not going to have a usable, truly metric system.
The 60/60/24 thing is kind of silly, but when the natural world works on 29ish (lunar period) and 365ish, you're kind of screwed to begin with. There are also aspects of our biology that might not work too well with multiples of 10 (e.g. a 10-day week would probably be too long between breaks).
So I'm all for metric, but until/unless we move predominantly into space, I'd stick with our quirky Earth-based time system.
2.8" x 4.5" x 0.7" ~600MHz Transmeta processor 600x800 reflective color touchscreen Stylus input with handwriting recognition Virtual keyboard input (watches finger position) Integrated cell phone with pull-away wireless mini-handset 3G cell/internet connectivity 8h battery life
Basically, a miniature piece of real computing hardware, not a toy. We can almost do this now; in a couple years it shouldn't be a problem. I don't see anyone really trying for it yet, though.
Data mining has nothing to do with buying other people's personal information without their permission, except that this is one way to get data. Personal information is already mined to death anyway; that's not where the hot new work is going to be.
Data mining is, at its heart, the use of mathematical techniques (self-organizing maps, independent component analysis, etc.) to extract patterns or key features from an overwhelming mass of largely irrelevant data. You can use it to predict customer buying patterns (old hat), to look for patterns in gene expression indicative of cancer (new hat), and to find relevant changes in fMRI-measured brain activity in substance abusers (hat-still-being-made).
So I view data mining in the same light as, say, encryption technology. Just because some big corporations want to encrypt their content to make you experience it in exactly the method they desire at exactly the time and place and with exactly the equipment that they specify, doesn't mean that encryption is of dubious ethical value. It means that big corporations act in ways that are ethically dubious.
Likewise with data mining.
False advertising
on
Enigma
·
· Score: 1, Redundant
That is absolutely the worst introductory paragraph for a review that I have ever seen. I am not fond of the DMCA either, but don't you think that the opening paragraph describing the review should have at least some resemblance to what the movie is actually about?
I am very disappointed in Slashdot (timothy in particular) for using that first paragraph as the summary on the main page. If you're going to publish the review at all, for heaven's sake, take the time to write two sentences describing the review, or excerpt a useful section, instead of using the useless and misleading first paragraph! Here, I'll do it for you:
Peter Waner reviews Enigma, writing "Enigma" is about good codebreakers -- the mathematicians and clerks of Great Britain's Bletchley Park who helped the Allied cause during World War II by breaking the German coding machine known as "Enigma." (Read on for the rest of
Peter's review.)
I've been writing collaborative fiction since about 1986 on the Island of Kesmai CompuServe message boards. It's commonplace among the online RPGing community (which wouldn't exist without collaborative fiction, though admittedly it's of widely varying quality). Most gaming sites now have associated bulletin boards (thanks to ezboard), and collaborative fiction is common among them. Sites for some games even let you build a hierarchy of pages to emulate a virtual world described and contributed to in text (e.g. the Shadowland fan-site for the Shadowrun game). And then there are MUXes....
If the literary/artistic community is only noticing this now, they are way behind the curve.
As other posters have mentioned, using C# seems a strange choice. It may be a good choice for the server-side, if you want to restrict yourself to only running the server on Windows boxes. However, for the client, using C# is a good way to dramatically reduce your customer base without gaining anything significant. Fortunately, the client and server don't need to be written in the same language.
There are a few considerations that you should keep in mind. First, I personally am an fan of rolling your own messaging system over open sockets. Use simple encryption to verify that the client is who it is supposed to be--don't pass plain text strings around. Make sure that the server doesn't trust the client to do anything sensible. (Lost packets, hackers, etc...you have no idea what is going to come in.) If you can get by with sending updates no more often than, say, once a second, have the server limit the amount of updates you're sending.
Also, as other posters have mentioned, you want to try to avoid N^2 behavior as much as possible. The server will have to inform everyone of what everyone else is doing--there is no way around that--but whenever there is a case where a change needs to be propagated, think about whether you can't hold off on propagating the change for a little while to see if other data comes in. If you want fast massively multiplayer interaction, this is a must.
Also, don't be afraid to send predictive information to the clients. For instance, most MMORPGs send not just the position of NPCs and PCs but also their velocity so the client can properly update the position without having to interact with the server. Once the velocity vector changes, the server sends an update to the client (sometimes revising history as to where the NPC was). This is a good strategy in general to offload work safely to the client--just make sure that the server checks all requests that the client makes, in case the client is working with inaccurate information and says the user can do something that they really can't do!
Re:Guilt By Association, don't buy it
on
Monsanto and PCBs
·
· Score: 5, Insightful
The main line of pro-GM arguments is that we can trust the science and the corporations.
That is only the line of argument to people who are unwilling to spend the time and effort necessary to examine the science--and I am not aware of too many people who argue that we can trust the corporations! Rather, it is people who understand the science who should keep an eye on what the corporations are doing.
For instance, GM crops that
* Allow massive pesticide use
* Do not produce fertile seed
* Massively overexpress the natural BT toxin
are all really stupid ideas in the long term, since they, respectively,
* Increase toxic residues in food, runoff, etc.
* Lead to a catestrophic situation (no crop) instead of a bad one (crop from poor stock) if for any reason the seed cannot be obtained next year
* Rapidly generate resistance to a substance that could otherwise be safely used for decades
On the other hand, GM crops that
* Increase the nutritional value of the crop
* Increase yield (all other things being equal)
* Increase natural resistance to disease (but not by having the plant make tons of one particular toxin)
are all really useful, for hopefully obvious reasons.
The sane answer is: pay attention to what corporations are doing, and (try to) call them when they do something stupid. If you don't have the background to decipher their claims yourself, find someone who can. But the bottom line is that GM crops are not inherently bad; just that a few of the simplest, greediest, short-sighted implementations by corporations are.
Publishing information about exploits is valuable, and creating code that can demonstrate an exploit is valuable, and releasing it if the company isn't going to fix the problem really fast is valuable, and releasing an exploit that helps people test to make sure they're no longer exploitable is valuable.
But in this case--due to threat of release or not--AOL claims to be fixing the problem very rapidly, and the exploit code is next to useless in fixing the problem. The only reason I can see, in this case, to release the code is if that is the only way that the people who found it would feel like they got enough attention/recognition to make it worth releasing the code rather than exploiting it themselves.
Next time, I hope they wait a little longer before releasing the code. Or if not, hopefully it won't get reported here until then. Otherwise it just gives ammunition to the "obscurity" folks who want to show how dangerous information is.
They have apparently forgotten the Franklin / Xircom / etc. "Rex". It's the size of a PCMCIA
card. It isn't as powerful as the handhelds,
but it sure seems like a PDA to me.
Also, when you say, "I'm not a scientist, I know
nothing about the technical issues involved,"
don't you think that knowing might help you judge
whether the problem with BSE has any bearing on
GM crops? Especially given that you're
postulating that genetic modifications could make
it "a million times worse"?
(Hint: even though the mechanism of action of BSE was surprising, the fact that feeding animals to each other could assist the spread of diseases should not have been surprising.)
There's often a sense that we really don't understand what might happen if we genetically modify crops--and that really, anything might happen.
Realistically, though, it's not that hard to figure out where the problems are likely to be. We are modifying organisms by modifying their DNA; it's not like this has never happened before. There are millions of different species (all with different DNA) containing quadrillions of individuals (all with different DNA) full of mutations, chromosomal rearrangements, gene transfer, and so on. Just by looking around at the world as it is now, we can get a pretty good idea of the parameters we're dealing with.
So, basically, most of the "Who knows what will happen!!" arguments are about as cogent as the "Linux is evil"-type FUD. Most people may not know what will happen, but experts in molecular biology, ecology, evolution, and so on, probably have a very good idea.
The real problem with GM crops isn't that we might accidentally create some world-conquering monster bug/plant/etc.. Evolution's tried that already. It generally doesn't work. So what are the problems?
One set of problems involves making resistant plants so you can blast everything else to death. It's the GM technology that makes this possible, but the GMness isn't the problem: it's the strategy of destroying everything but your-favorite-monoculture-crop. Related is releasing an organism in an inappropriate place to try to control a problem. This isn't at all a new problem to GM crops.
Another set of problems involves being really, really shortsighted, like with BT corn. BT is nice because it degrades rapidly and is only toxic to certain classes of insects. So we'll produce BT in all our crop plants and (through natural mechanisms) select for resistant insects. In five or ten years, BT will be utterly useless. Now doesn't that sound like a good idea!
I think allergies are really overblown as a problem. Yes, maybe a fish gene inserted into wheat might affect someone who, fantastically unluckily, happened to be allergic to just that gene product. But those people would quickly learn that they are allergic to fish and wheat and eat accordingly. No big deal. And perhaps some really common antigens (e.g. in peanuts) can be GMed out of the crops. That could be really helpful.
Also, fears of rampant spreading of unusual genes really confuses me. The reason we've got all these big fancy labs is that genes don't spread
well outside of the organisms they are in. That's why we have to modify them in the first
place! So how, exactly, would they escape?
Anyway, we should of course be careful, but between good background knowledge and some common sense, GM isn't that scary.
I strongly urge programmers to avoid releasing binaries before their code will compile. Doing so is liable to result in a flood of unhelpful bug reports, such as "The ****ing thing won't run!" and "You said this was the binary, but you gave me a 0-byte file!!".
Slashdot Mirror
Maybe they didn't write it.
Not only doesn't ionizing radiation penetrate solids and fluids very far, but ionizing radiation generates high-energy chemical compounds that any life could potentially use as a food source. It just can't be directly on the surface--but you wouldn't want to be there anyway, since you'd tend to evaporate into space and/or freeze.
Anyway, the ionizing radiation has been known about for years and is one of the major reasons why scientists thought there might be life!
Possibly...but the description was hopelessly bungled the first time around. For instance, the scaling was called "exponential" and then stated to be N^N, which is superexponential. (N^N=e^(N ln N).)
In any case, simply observing that there are 2^N possible groups doesn't mean that any significant fraction of them come into being. It's the number that actually come into being that are important, and those are limited by how much time people have and human social dynamics.
Advertisers have been tricking our brains for a long time in order to sell products.
Why not juxtapose desirable women and famous athletes with...carbonated sugar water? Yes, that makes a lot of sense. Except our brains are wired to associate things that appear together....
I recently bought a Butterfinger Bar. 10% more free, it said, with a bright green stripe on the end of the bar suggesting just how much more you were getting! Wow, that was another good bite, yum! Except when I measured the stripe, it was 20% of the length of the bar....
I bought a banana too. There was a sticker advertising low-calorie sweeteners on it. That was just weird.
Anyway, point is, advertisers implicitly lie, mislead, deceive all over the place--just not explicitly, because explicit lies are illegal. If they want to spend a bunch of money on fMRI studies so they can be a bit more successful, I say, great! It won't make a huge difference to advertising (and if it does, we might outlaw misleading implications).
But if they dump enough money into fMRI research, they might bring the cost of the machines down, and that would really help with *useful* research into the functioning of the brain. That would be a nice fringe benefit.
This is *exactly* what you'd expect to see. You have fields with pollen right next to each other, try to kill off one side with Roundup, and gee, no wonder the plants who were cross-polinated and picked up the resistance gene started to survive!
However, this doesn't extrapolate to any giant evil nasty consequences. If you want to be able to kill your corn with Roundup, don't give corn a herbicide resistance gene (duh).
However, herbicide resistance genes are kind of stupid, because there *is* some low rate of horizontal gene transfer, and eventually a weed will pick it up too, and then your herbicide won't even be killing weeds. This doesn't show that GM is evil, just that people can use it in shortsighted ways.
Gee, just like everything else.
Money = flexibility. You have nine months. If flexibility and adventure are important to you, save some now. Whether you end up in the Peace Corps or whatever, it will help give you room to breathe.
- r = 2GM/c^2
where G = 6.67e-11 m^3/s^2*kg (the gravitational constant) and c = 3e8 m/s (the speed of light, of course). Plug in 3 million sun-masses (the sun weighs 2e30 kg), and you have- r = 8.9e9 m = 5.5 million miles = 0.06AU
So unfortunately, the event horizon isn't three times as big as the solar system. The earth's orbit is 1AU (that's how the unit is defined). The event horizon barely stretches past the surface of the sun (7e8 meters)!So much for that idea!
UCSD's acceptable use policy doesn't reserve the right of the university to remove sites based on content alone. (For copyright, safety, legality, etc., they do--but not to edit content.) See UCSD's web policy.
The letter specifically says that the UCSD policy that has been violated is the prohibition against breaking federal law (i.e. VI.H.i). That makes it a First Amendment issue--they are essentially claiming that this link is illegal. The First Amendment suggests that the link would count as protected speech.
The letter *could* have cited VI.C.ii and said, "We're getting complaints about this site and don't want to deal with it. This is an avoidable incremental cost, so stop it." This would be less directly a First Amendment issue, but it would still directly be a free speech issue. The policy would then be, "You can't say anything objectionable, because someone might object, and we don't want to hear about it."
Besides, since when have private ISPs been required to have fewer restrictions on speech than a public university? Maybe if ISPs were considered "common carriers", but so far they've been decent at dodging that label.
A bunch of people seem to be under the impression that UCSD actually owns the machine that burn is hosted on. As far as I know, this is not correct; burn is on a student-hosted machine, but uses the university network for internet access.
As I recall, UCSD does not own the box. It's a
private machine hooked up to the university's
network.
(That'll teach me to hit "submit" before "preview".)
As the previous poster pointed out, there isn't any parity to speak of in DNA. I have no idea why they're trying to make the comparison.
Hydrogen bonding is a much higher-fidelity error correcting system than parity checking. With parity checking, you can catch any single error (the number of 1's changes), but not any pair. With hydrogen bond donors and acceptors, you have to have an exact match at all positions. Any number of errors ruins the complementarity.
For what it's worth, the base pairing system is quite elegant: due to size constraints, purines (A,G) and pyrimidines (C,T) must pair with each other. Using the article's notation of 0=H-bond acceptor, 1=donor, the four nucleotides are
A = 10_
C = 100
G = 011
T = 101
where _ is simply a hole. The best pairwise complementarity is (evidently) AT and CG.
It's a chemically implemented RAID-1 system, not a parity check.
Well, not without changing the Earth's orbit or
rotational velocity. And we can't do that yet.
As long as we inherently have 365ish days per year, we're not going to have a usable, truly metric system.
The 60/60/24 thing is kind of silly, but when the natural world works on 29ish (lunar period) and 365ish, you're kind of screwed to begin with. There are also aspects of our biology that might not work too well with multiples of 10 (e.g. a 10-day week would probably be too long between breaks).
So I'm all for metric, but until/unless we move predominantly into space, I'd stick with our quirky Earth-based time system.
2.8" x 4.5" x 0.7"
~600MHz Transmeta processor
600x800 reflective color touchscreen
Stylus input with handwriting recognition
Virtual keyboard input (watches finger position)
Integrated cell phone with pull-away wireless mini-handset
3G cell/internet connectivity
8h battery life
Basically, a miniature piece of real computing hardware, not a toy. We can almost do this now; in a couple years it shouldn't be a problem. I don't see anyone really trying for it yet, though.
They should just ship the units with a sticker that says:
This product has been designed and manufactured using the same level of rigor and testing as software packages from major vendors.
Data mining has nothing to do with buying other people's personal information without their permission, except that this is one way to get data. Personal information is already mined to death anyway; that's not where the hot new work is going to be.
Data mining is, at its heart, the use of mathematical techniques (self-organizing maps, independent component analysis, etc.) to extract patterns or key features from an overwhelming mass of largely irrelevant data. You can use it to predict customer buying patterns (old hat), to look for patterns in gene expression indicative of cancer (new hat), and to find relevant changes in fMRI-measured brain activity in substance abusers (hat-still-being-made).
So I view data mining in the same light as, say, encryption technology. Just because some big corporations want to encrypt their content to make you experience it in exactly the method they desire at exactly the time and place and with exactly the equipment that they specify, doesn't mean that encryption is of dubious ethical value. It means that big corporations act in ways that are ethically dubious.
Likewise with data mining.
That is absolutely the worst introductory paragraph for a review that I have ever seen. I am not fond of the DMCA either, but don't you think that the opening paragraph describing the review should have at least some resemblance to what the movie is actually about?
I am very disappointed in Slashdot (timothy in particular) for using that first paragraph as the summary on the main page. If you're going to publish the review at all, for heaven's sake, take the time to write two sentences describing the review, or excerpt a useful section, instead of using the useless and misleading first paragraph! Here, I'll do it for you:
Peter Waner reviews Enigma, writing "Enigma" is about good codebreakers -- the mathematicians and clerks of Great Britain's Bletchley Park who helped the Allied cause during World War II by breaking the German coding machine known as "Enigma." (Read on for the rest of Peter's review.)
I've been writing collaborative fiction since about 1986 on the Island of Kesmai CompuServe message boards. It's commonplace among the online RPGing community (which wouldn't exist without collaborative fiction, though admittedly it's of widely varying quality). Most gaming sites now have associated bulletin boards (thanks to ezboard), and collaborative fiction is common among them. Sites for some games even let you build a hierarchy of pages to emulate a virtual world described and contributed to in text (e.g. the Shadowland fan-site for the Shadowrun game). And then there are MUXes....
If the literary/artistic community is only noticing this now, they are way behind the curve.
As other posters have mentioned, using C# seems a strange choice. It may be a good choice for the server-side, if you want to restrict yourself to only running the server on Windows boxes. However, for the client, using C# is a good way to dramatically reduce your customer base without gaining anything significant. Fortunately, the client and server don't need to be written in the same language.
There are a few considerations that you should keep in mind. First, I personally am an fan of rolling your own messaging system over open sockets. Use simple encryption to verify that the client is who it is supposed to be--don't pass plain text strings around. Make sure that the server doesn't trust the client to do anything sensible. (Lost packets, hackers, etc...you have no idea what is going to come in.) If you can get by with sending updates no more often than, say, once a second, have the server limit the amount of updates you're sending.
Also, as other posters have mentioned, you want to try to avoid N^2 behavior as much as possible. The server will have to inform everyone of what everyone else is doing--there is no way around that--but whenever there is a case where a change needs to be propagated, think about whether you can't hold off on propagating the change for a little while to see if other data comes in. If you want fast massively multiplayer interaction, this is a must.
Also, don't be afraid to send predictive information to the clients. For instance, most MMORPGs send not just the position of NPCs and PCs but also their velocity so the client can properly update the position without having to interact with the server. Once the velocity vector changes, the server sends an update to the client (sometimes revising history as to where the NPC was). This is a good strategy in general to offload work safely to the client--just make sure that the server checks all requests that the client makes, in case the client is working with inaccurate information and says the user can do something that they really can't do!
The main line of pro-GM arguments is that we can trust the science and the corporations.
That is only the line of argument to people who are unwilling to spend the time and effort necessary to examine the science--and I am not aware of too many people who argue that we can trust the corporations! Rather, it is people who understand the science who should keep an eye on what the corporations are doing.
For instance, GM crops that
* Allow massive pesticide use
* Do not produce fertile seed
* Massively overexpress the natural BT toxin
are all really stupid ideas in the long term, since they, respectively,
* Increase toxic residues in food, runoff, etc.
* Lead to a catestrophic situation (no crop) instead of a bad one (crop from poor stock) if for any reason the seed cannot be obtained next year
* Rapidly generate resistance to a substance that could otherwise be safely used for decades
On the other hand, GM crops that
* Increase the nutritional value of the crop
* Increase yield (all other things being equal)
* Increase natural resistance to disease (but not by having the plant make tons of one particular toxin)
are all really useful, for hopefully obvious reasons.
The sane answer is: pay attention to what corporations are doing, and (try to) call them when they do something stupid. If you don't have the background to decipher their claims yourself, find someone who can. But the bottom line is that GM crops are not inherently bad; just that a few of the simplest, greediest, short-sighted implementations by corporations are.
Publishing information about exploits is valuable, and creating code that can demonstrate an exploit is valuable, and releasing it if the company isn't going to fix the problem really fast is valuable, and releasing an exploit that helps people test to make sure they're no longer exploitable is valuable.
But in this case--due to threat of release or not--AOL claims to be fixing the problem very rapidly, and the exploit code is next to useless in fixing the problem. The only reason I can see, in this case, to release the code is if that is the only way that the people who found it would feel like they got enough attention/recognition to make it worth releasing the code rather than exploiting it themselves.
Next time, I hope they wait a little longer before releasing the code. Or if not, hopefully it won't get reported here until then. Otherwise it just gives ammunition to the "obscurity" folks who want to show how dangerous information is.
They have apparently forgotten the Franklin / Xircom / etc. "Rex". It's the size of a PCMCIA
card. It isn't as powerful as the handhelds,
but it sure seems like a PDA to me.
Also, when you say, "I'm not a scientist, I know nothing about the technical issues involved," don't you think that knowing might help you judge whether the problem with BSE has any bearing on GM crops? Especially given that you're postulating that genetic modifications could make it "a million times worse"?
(Hint: even though the mechanism of action of BSE was surprising, the fact that feeding animals to each other could assist the spread of diseases should not have been surprising.)
Realistically, though, it's not that hard to figure out where the problems are likely to be. We are modifying organisms by modifying their DNA; it's not like this has never happened before. There are millions of different species (all with different DNA) containing quadrillions of individuals (all with different DNA) full of mutations, chromosomal rearrangements, gene transfer, and so on. Just by looking around at the world as it is now, we can get a pretty good idea of the parameters we're dealing with.
So, basically, most of the "Who knows what will happen!!" arguments are about as cogent as the "Linux is evil"-type FUD. Most people may not know what will happen, but experts in molecular biology, ecology, evolution, and so on, probably have a very good idea.
The real problem with GM crops isn't that we might accidentally create some world-conquering monster bug/plant/etc.. Evolution's tried that already. It generally doesn't work. So what are the problems?
One set of problems involves making resistant plants so you can blast everything else to death. It's the GM technology that makes this possible, but the GMness isn't the problem: it's the strategy of destroying everything but your-favorite-monoculture-crop. Related is releasing an organism in an inappropriate place to try to control a problem. This isn't at all a new problem to GM crops.
Another set of problems involves being really, really shortsighted, like with BT corn. BT is nice because it degrades rapidly and is only toxic to certain classes of insects. So we'll produce BT in all our crop plants and (through natural mechanisms) select for resistant insects. In five or ten years, BT will be utterly useless. Now doesn't that sound like a good idea!
I think allergies are really overblown as a problem. Yes, maybe a fish gene inserted into wheat might affect someone who, fantastically unluckily, happened to be allergic to just that gene product. But those people would quickly learn that they are allergic to fish and wheat and eat accordingly. No big deal. And perhaps some really common antigens (e.g. in peanuts) can be GMed out of the crops. That could be really helpful.
Also, fears of rampant spreading of unusual genes really confuses me. The reason we've got all these big fancy labs is that genes don't spread well outside of the organisms they are in. That's why we have to modify them in the first place! So how, exactly, would they escape?
Anyway, we should of course be careful, but between good background knowledge and some common sense, GM isn't that scary.
I strongly urge programmers to avoid releasing binaries before their code will compile. Doing so is liable to result in a flood of unhelpful bug reports, such as "The ****ing thing won't run!" and "You said this was the binary, but you gave me a 0-byte file!!".