A few weeks back, I got curious about something: what was the total karma of the AC? I asked Rob. It was, on December 17th, 1999, 1972 points. (Obviously, since the AC doesn't moderate, this is just a sum of moderation.) This means that overall, we've considered ACs to be posting some pretty useful stuff.
When that number goes negative, then it might be time to think about changing the AC system.
I think I see what's going on here. It looks like the judge issued an injunction against VCR and Ferret, but not against Ripper. Hence, Real decides to claim victory on the best-two-of-three principle, while Streambox claims victory because they can still sell Ripper. Notice that they only talked about "continuing development" on VCR and Ferret, not about releasing them.
-shrug- It's a start, since this yanks away some of Real's content control.
I think you could could do the same thing to grab speech without speeking that you could do to grab arm movments without moving your arm.. just watch little abortive movements of the vocal cords.
Subvocalization, yes. You are correct that this may be a way to grab speech signals (I dare not say more because I'm thinking of researching in this area), but you still wouldn't get the vocal tones. Also, if you're grabbing directly from small muscle movements, the technology will not be useful for helping persons (like Hawking) with motor disorders. Since the brainjack is most likely to arise out of research for the disabled, I think it unlikely that we'll end up going this route.
You might not care about spech that much though once you got the arm/finger movements working since a 2D symbole buffer is a much better means of communication (and your langauge can adapt to it the same way it adapts to writing instead of speeking).
In which case, you once again don't really need direct neural interface; this sort of thing is the principle behind the various chorded keyboards and "sign languages" that are being experimented with to enable wearables.
Actually, your face and mouth might be the best way to get info to the computer. There are LOTS of seperatly controlled mussles which you could be trained to move individually.
That's certainly true. On the other hand, you'd look pretty weird, especially trying to compute while talking.
Connecting close to the mussle level solves all these problems since you can always choose notto talk.. unless yuou talk to yourself a lot..:)
Many people can be seen to move their lips when they read. I know that I have a tendency to subvocalize when I'm typing messages. (Of course, I also talk to myself.)
Example: I supposet there might be way to stimulate a section of knowledge to make you recolection of it better, but we could do something like this by having a series of "note cards" which recap the importent theorems or something.
-shrug- We've got that now to one degree or another; my text editor lets me have a "clip book" of common language motifs for all the markup and programming languages I know. I could probably write similar reminders for any new language, or any extensions I make to my existing languages. If you're talking about something that provides better memory by giving people an external store... well, in theory it's possible, but it's a long way off, because we haven't a clue what format the brain uses to encode memories, let alone the signals to access those memories.
It just sems that we don't know that much about our though processes execpt for the parts that work via langauge.
We don't know much about our thought processes, period. The 1990s were supposedly the "Decade of the Brain". We made some progress. The advent of fMRI helped a great deal, as did advances in microbiology and biochemistry. The Genome Project may also help. However, there's a huge amount of work remaining to be done, just as the HGP has sequences, but no clue what those sequences do.
At the intro levels of cognitive science, AI, and robotics, everything looks neat. Then you get inside and discover how little we actually understand. It can get a bit discouraging at times, but it also gives you some job security.
Hrm... The first time I read that statement, I thought you were saying that couldn't just do that with a computer. The second time I read it I thought you were saying that you could do it with a computer, but not with the brain.
I believe that doing it to the brain is potentially impossible, or at least very very hard. Doing such a thing to a computer is simpler, since computers are simpler devices, but I still don't think it'd be easy.
In any event, you probably could solder more stuff on to a computer motherboard if you really needed to (the memory interface would be a good place to start...)
There's two problems with that analogy. First, the memory interface is just that: an interface. If you've got an open slot, then you've got something more than the brain has. (Yes, the brain has memory, and it's got to be stored somewhere. Maybe someday we'll be able to read/write it directly, but that's even further off than doing image capture from the visual system.)
The second problem is the one you'd have to overcome to try and do the computer task. Let's say you do solder something in between one of the SIMMs/DIMMs and the microprocessor. Every time you send a signal to your device, the memory also sees it. Every time your device writes something to the lines, memory also sees it. Every time you fetch from or write to that memory, your device sees the signal. In all cases, something will be trying to react to a signal it wasn't meant to see. It seems unlikely that you could get this to work even with very clever software, unless the device's purpose was simply to modify the actions of the memory chip.
If that were true, we would all still be single celled organisms, I'd bet. I don't know why our brains would need to be so advanced as to be able to figure out things like relativity, integral calc or the structure of the atom.
I said "superfluous". The ability of our brain to handle abstract thought is clearly not superfluous; it's allowed us to survive and proliferate very nicely. Advanced physics will eventually get us the hell off this planet before it becomes uninhabitable, and therefore the ability to understand it is directly useful for the continuation of the species. However, if there were a population of neurons that wasn't doing anything (like the fallacy that people are only using 10% of their brain), it would be very rapidly selected against.
But what about "sidebanding" the data? I mean, when someone plays quake they don't have to think about there fingers or the monitor in front of them there just "there". In a Sci-Fi Novel I'm writing people interface with computers by 'co-opting' unnecessary nerves in the spinal cord. Would something like this work?
I'm not making the connection between the two. I went and read the parts of your story you've posted. You made reference to "unconnected nerves" --- AFAIK, there ain't no such animal. Part of the biology of neurons is that if they don't receive regular simulation (both electrical and chemical), they die.
You raise an interesting point, though. A lot of this die-off occurs in childhood. If you did one of the neonatal implants and plugged it into some sensory cortex and used it to stimulate and secrete growth factors, you could create an extra population of sense neurons specifically devoted to that implant. The problem is, it's still sharing a bus with the existing sense data. We have no idea whether or not the brain will be able to handle the extra data. (If you used it for something related to that sense, like seeing UV or smelling new compounds, it could possibly be integrated properly.) To test it, though, we'd need to be doing potentially harmful experiments on perfectly normal kids. That's unlikely. It's also a rather limited use of implants anyway. You don't need an implant to see UV or IR; all you need are the right kind of lenses. I think brain implants are very cool, but they shouldn't be used if not necessary; the body never responds perfectly to being messed with.
How much do they really know about the phonological loop? How diffrent is the voice in the back of my head from the voice that is about to come out?
The answer in the cobwebs of my mind is that there is some linkage, but it's not total. That is, when you're hearing your thoughts in your head or articulating via typing, the hearing/speech zones do become activated, but to a lesser degree than when you actually perform those tasks.
Actually, when one thinks about it, the processes of typing and writing are pretty neat; it's conversion to a completely different output form, and yet my brain knows that speech and text have very similar informational content.
If the process of thinking about moving an arm or saing something really dose triger activity like the actual movement then maybe we could interact with a computer via these abortive movements. This would be the killer app. for brain implants since people could use images and audio in everyday communication.
There is the possibility of doing what you suggest, though we need to improve our tech a bit first. Using it to move a mouse or select symbols is a bit more realistic than having the computer snag entire images and sounds out of your head, though. (And remember, when the sound is in your head, it doesn't have the tones of your voice attached to it yet; those come when it's actually articulated. Thus, if you coded it to audio, the computer would have to do voice synthesis for you. (Stephen Hawking, obviously, would love such a thing. I doubt we'll be able to give it to him before he dies.))
Also, realize the potential problems of doing this. If there's an implant recording your thoughts and sending the data stream to an Internet-connected computer... you think you've got Big Brother looking over you now? Just wait.
Idea: the phase space of the ``possitions of the human body'' is MCUH larger then 3 dimentional. It might be possible to take a mussle group and program a computer to respond to the movements of those mussles as if they were moving an object in a higher dimentional space, then the user might gain some intuition about things which they could apply to solving some open problem.
This is an interesting idea. I don't think it needs direct neural interface, though. Why not just use a control glove or other existing haptic interface? If you need a really large number of DFs, use one of the motion-capture systems they have for animation tasks, like the Flock of Birds.
Damn straight. My kid's not going to have holes punched in his head either. And that's a big hurdle for this sort of research to clear. Experiments on the human brain are exclusively done to volunteers, usually those who have no real alternative to this sort of surgery. And let's face it, how many people want to have their brain poked at?
Actually, I should amend my original comment. I would not volunteer a child of mine for such an experiment given the current state of the tech. However, if I knew that my child would be blind from birth, and if such implants were highly successful in adult humans, and if neonatal implants had already been tested and proven in other species, then I'd be willing to go ahead with it.
As you say, volunteers are a hurdle. However, there are plenty of people with various kinds of sensory and motor disabilities out there. Implantation of electrodes isn't pleasant, but it's usually not life-threatening. Given the tradeoff, I think there'll be sufficient volunteers to establish clinical usefulness once this field takes off (which seems likely to happen Real Soon Now).
Basically, what you were saying was "The brain has never been able to use anymore then 5 senses." That's true, but until we hook more stuff up to it, we'll never know.
Right, but that's like saying "My PC which has no expansion slots has never been able to use anything more than the default hardware, but until I crack the case and start soldering stuff onto the traces on the motherboard, I'll never know." It's not that simple. For one thing, where would you plug in new inputs? I'm in neuroscience (admittedly only at the beginning grad level), and I can't think of a place.
There's also the possibility of ether "growing" new lobes for new things or "emulating" the lobes in hardware. Or we could just plug new senses into the visual cortex or something (would it be another type of sight then?)
Well, first off, you can't really say that a given function exists in a given lobe, because they're shipping signals all over the place. Secondly, even if you grow a new set of tissue to plug your new port into, you still need to teach the brain to integrate that signal, and that's a process which is not really understood at all. To some degree, we'd have to solve the problem of where consciousness is.
Your comment about plugging in new things to the visual pathway (or another sense pathway) makes the most sense, and I think that this is how most "new senses" will end up. If we could actually get massive arrays of microelectrodes (and the software to configure and drive them), then it'd be possible to overlay stuff on the retinotopic maps in realtime, just like CBS does to their video feed. (Yes, we will cover your significant other's body with ads for our new porno site. Ain't technology wonderful?)
Anyway, the human brain can do *a lot* more then is evolutionarily needed. I'm sure that it could be augmented somewhat by technology.
It depends on how one parses your statement. Yes, the brain can think of a whole lot of stuff. On the other hand, there's evidence that suggests that it basically handles all that stuff using a few limited pathways. It all mostly comes down to efficient pattern-matching anyway. However, that does not imply that the brain has a lot of extra circuitry lying around which one can just co-opt at will. Evolution is not kind to superfluous stuff.
I just looked back and realized that they said he had this implanted in 1978 -- Does that seem like an awful long time ago (for this sort of thing) to anyone else?
Not really. Ever since we realized the brain was electric, people have been stimulating it like mad and seeing what they can put in and take out. The fact that it's taken us this long to get this far should tell you something about how hard a problem it is. (Consider how much trouble we still have with the problem of computer vision. It's getting better, but it's nowhere near a solved problem, IMHO.)
If they could do a 10x10 pixel image back then, what are they capable of now?
Well, speaking as someone who's sort of part of they... a 10x10 pixel image. That's why this is news --- after two decades of trying, computer tech has finally gotten to the point where we can give Jerry useful input. However, the science of brain electrodes hasn't advanced that much. They're more durable now and often thinner, but in practice, we still probably wouldn't be able to sink nearly enough into Jerry's visual cortex to convey a complete visual image. (On the other hand, there is the example of the cat. However, that cat was never expected to have long-term survival.) We may be jacking in within our lifetimes, but it's not going to be as soon as you hope.
Alik Widge MD/PhD Program University of Pittsburgh/Carnegie Mellon
Makes you wonder if, as a baby, you had some strange thing (IR port, GPS, radio) wired into your brain just after birth, would you learn how to use it, just as you learn how to stand up, talk, and focus your eyes?
My training is nowhere near complete, but my initial guess would be no. The brain does have dynamic connection capacity, but that capacity has always been set to deal with the five senses, all the way down the evolutionary tree. The brain has all the input ports connected to senses already, and as far as we know there's no place God left for us to install new peripherals. Thus, any kind of neural-interface tech is likely to work using the existing sense inputs.
Now, that doesn't mean there can't be some kind of extra output added, as was the case with that guy down in Georgia. It is likely that people will eventually be able to control computers through thought. However, chances are that the output from the computer will still come back through the same sense lines. (It's possible, I suppose, that someone might figure out exactly where the "phonological loop" of short-term memory is (that's the part you're using when you hear your own voice in your head), decode the representations of all known phonemes, and start injecting thoughts in via electrodes. However, that sort of capability is in the very far future.)
As for putting something into a kid so they'd be naturally adapted to it... might not work as well as you'd think. Yes, kids' brains are more adaptive. However, kids' brains are also still developing, and thus really sensitive to disruption. A little bit of extra electrical stimulation at the wrong time can seriously fuck things up. I personally would not volunteer a child of mine for such experiments, even if I'd designed the device myself.
Reread the analysis. The guidelines are mainly intended to address expressed sequence tags, or ESTs. These are fairly small bits of DNA which let you quickly identify areas of the genome. To use a somewhat-sloppy analogy, they're like regexps for the genome.
Now, imagine that I spend a few months compiling a whole bunch of useful regexps. I think I'd be within my rights to want to profit from all that work. Now, I can't patent a regexp, because in most cases it falls under obviousness. However, DNA tags are a lot less obvious, and therefore require significant effort to find and construct. Therefore, some companies claim they should be able to patent the ones they discover so that they can make money selling reagents to biologists.
ESTs, to the best of my knowledge, cannot be considered genes. They may code for a few amino acids, but as far as I know none of them codes anything near a complete enzyme.
Also, in response to the "they'll patent the cure for Alzheimer's" and "they'll patent the diabetes gene" worries: these are multifactorial diseases. It's not just one gene at the heart of them. Furthermore, it's not just one mutation causing all the cases. There are both genetic and environmental triggering factors.
Now, it's certainly true that a company might get a patent on a given kind of gene therapy or a "healing" DNA sequence that they've constructed. IMHO, this is fair. Such a sequence would be essentially equivalent to a drug, and anyone who figures out a cure for a disease deserves some profit.
Nobody is going to patent your genes and then sue you for infringement-through-living. Or rather, they might do that, but there's no way a court will even grant them the preliminary injunction, let alone damages. (If you are a genetically-engineered perfect child using "upgraded" genes that the company invented, *then* you might be in trouble.) As far as medical therapies go, gene patents are probably going to just perpetuate the system of drug patents.
Finally, a semi-offtopic response to someone who said that the genome is open-source but encrypted: I don't think it's encrypted at all. God's just not very good at commenting his code.
Furthermore has anyone even _heard_ of someone that bought something because of SPAM?
Yes. My father. Twice in the past month. First, he was spammed by one of the telcos offering a low rate. (I think it was GTE, but I'm not certain.) Problem is, it's a legitimate company, not some fly-by-night, and it really was a better rate than we have. He's seriously considering the switch.
Just this weekend, we got spammed about DSL. He wants DSL. They're offering a rate that's cheaper than our current two-analog-line bill from Hell Atlantic. I think he won't end up getting it simply because we live in an area where DSL is not yet available, but otherwise, that'd be another successful spam.
The only reason for him to not take such offers is conscientious objection. However, like most Americans, my father doesn't give a rat's ass about the way a few geeks want the Internet to be; he just wants BetterFasterCheaperNOW. Worst part is, he's an engineer. He does have some understanding of why certain things are problem (he's been subject to things like the One-Click patent in his own field), but considers it foolish to pass up a deal just because you dislike a company's business practice.
If John Wanker receives a "FREE TEEN Pr0n!!!!!" message in his inbox, he probably clicks it and racks up another banner count for the spammer. He might even sign up for the pay service.
Spam is used because spam works. It's got marginal costs that are almost zero, to the point where a 1/10,000 response rate to your ad will more than pay for the cost of the mailing. It may digust and annoy us, but like so many things, it'll exist as long as it's profitable.
P.S. I used hacker intentionally here so please no "use cracker" posts. It is a lost cause the rest of the nation has already adopted the new vocabulary. Besides its just a word why do you care?
It may be just a word, but it's our word. (Relatively speaking; I wasn't even born yet when the first hackers were being named.) Moreover, since words are the method of communicating thought, words shape that thought. This is one of the reasons why Newspeak was implemented in _1984_.
Consider the original definition of "hacker", a person who makes furniture using an axe. Obviously, there's a sense of doing a lot of bashing around with crude tools. However, the final product of this, when done properly, is useful and often aesthetically pleasing. (For a modern-day example, consider those who make wood sculpture with chainsaws.) In many ways, I think this captures the way a "true" hacker works.
Now consider "cracker". The best I can come up with is "A tasty hard bread-like thing" or "Someone who goes around breaking things." There's different mental associations with these words, and I personally think crackers should be stuck with the label that fits their activities. (This is why the guys from l0pht and CoDC get to call themselves hackers as far as I'm concerned.)
Furthermore, your claim that the nation has adopted the new meaning is not quite true. Over the past year, an increasing number of stories referenced on Slashdot have been using "systems cracker" instead of "hacker" when referring to someone involved in digital intrusion. As has been pointed out in the AOL/TW threads, most people will swallow whatever the media feeds them. It appears that media outlets are slowly learning to distinguish, and that will easily be absorbed by media consumers.
Go thou, read _Snow Crash_, _1984_, and _Brave New World_, and then watch _Brazil_. You'll understand that there's very little which is "just a word".
I think the problem with alot of the.coms is they are run by computer geeks who've never been to business school and have only read a couple books on the subject.
Unless you've got some stats to back this up, my experience contradicts your statement. From what I've heard (and I admit to having no backup other than a few articles in the newspaper), most new dotcom people are not geeks. At least part of this is because so many of the new companies are based around business models that involve things like "Enter our contest to win a free plastic trinket by giving us a sample of your DNA and the email addresses of everyone in your department!" or "You must have IE 9.0 and enable every single feature in it to use this site." Such things seem to be inherently annoying and disgusting to the geeks I've met. (Of course, it varies.)
Many business school profs are currently complaining about the dropout rate. Students start school, get some random idea for a new thing to sell over the web, and drop out to start the company and grab the domain before someone else does. It seems to me that this is where a good chunk of the expansion in e-business is coming from. (The exceptions would be sites that actually feature new technology; Google is one such. Most new commercial sites aren't featuring any new algorithms, though; they seem to just be variations on the catalog, fact-database, and auction themes.)
Also, a completely random thought that just popped into my head. The current buzzword for e-commerce companies is "dotcoms". Why didn't we end up with the word "dotcommerce" as well?
The only change we are asking for, is that those who believe they are qualified to not go that route, be allowed to not go that route.
I understand this. However, if such permission were enacted, I think the majority of people would end up trying to self-prescribe. They might be trying to save the cost of a doctor's visit, they might have an inflated opinion of their own intelligence, or they might get swayed by corporate propaganda.
For evidence, consider the current craze for herbs and nutritional supplements. None of it has to be approved by the FDA to treat anything. It can all be bought without prescription. The result is that people are wolfing down tons of pills that in many cases aren't likely to have more than a placebo effect. (This is not always true, but it is often true.) There have even been cases where people died because they neglected standard medicine in favor of alternative medicine and herbal supplements.
The above practices, although potentially useless, also appear to be free of dangerous side-effects. The currently regulated medications are not nearly so benign. I count myself as libertarian, but this seems like a case where the average citizen needs to be protected from the power of advertising.
Not true. How does being apointed make a person less greedy? Its true it makes them possibly more suited to the job (having been selected not by popularity contest but by selection based on qualifications) but...once there... them dead green presidents look mighty sweet no matter how you got your job.
True, they are still vulnerable to bribes. However, bribery is not the main form by which money influences government, at least not at the federal level. Most of the interfering money, at least as far as I've seen (and I spent six months working for two Congressmen, three of them helping to manage a campaign), comes in the form of campaign contributions. Campaign contributions aren't really relevant to an appointed official, for obvious reasons.
Of course, if you're looking at bribery within the system, you might also look at all the free food and teeny promotional gizmos the drug company reps give to the doctors...
Um, Alik?? Prescription medications are *already* regulated in every state. You haven't explained why we need federal control.
IMHO, the Feds have to be invoked in this case because we're not dealing with transactions within one state. Most e-pharmacies are trying to sell to the entire nation at once, which suggests the need for national-level regulation. If nothing else, it clearly brings in the interstate commerce authorities.
p.s. yes, sometimes private entities fail. However, those that fail go out of business, and their owners fall into disrepute. When government fails to properly regulate, does it ever go out of business? Do any government employees suffer?
I haven't seen TRUSTe go out of business...
Aside from that, you make a fair point. Government is poor at noticing when something is broken. However, I still claim that the government is likely to do a better job than a private entity; see next post.
I'll admit to being a bit biased, being in med school and all that. However, I think the points that some other posters have made need to be reinforced. It's true that in an ideal libertarian society, there'd be no need for any kind of prescriptions; you could just find a pharmacy and buy what you wanted. The problem is that this ideal depends on having an educated citizenry. Issues of intelligence aside, in order to be educated about a medication, you need valid information. How likely do you think it is that the pharmaceutical industry, if left unregulated, would be willing to provide negative information about their products? The chemical structures of drugs would probably be locked up as "trade secrets". This is a fairly clear case of abuse of power. (Remember the days when people went around in wagons selling their Patented Cure-All Tonics?) Hence, the Food and Drug Act, which has had a net benefit for the health of Americans.
It's true that people should take responsibility for their actions. However, there's a limit to how much self-education you can expect someone to do. It's not reasonable to expect every person to keep perfectly up-to-date on biochemistry, consumer electronics, economics, and every other field of study which might affect his or her life. The volume of information out there is just too huge; this is why the pundits say we're in the Information Age. Therefore, like a good little capitalist economy, we specialize in various fields of knowledge. The problem comes when we can no longer trust a given person to have or use the knowledge their role implies. Without the ability to trust individuals or corporations directly, we have to fall back to the trusted-third-party model.
One poster suggested that instead of government, we rely on corporate entities (profit or nonprofit) whose job is to certify credentials. That sometimes works, but not always --- witness the recent problems with TRUSTe. In this case, the Federal Government doesn't have anything to gain by playing favorites, and therefore it can probably be trusted. (The FDA, IIRC, is made up of appointed officials, not elected ones, making it a bit harder to corrupt with money.) Therefore, the Feds might not be the optimal solution, but they strike me as better than any proposed alternative.
Alik Widge MD/PhD Program University of Pittsburgh/Carnegie Mellon University
A reading suggestion in the same vein
on
Gates of Fire
·
· Score: 1
Those of you who enjoy comic books may wish to check out Frank Miller's _300_, published last year as a 5-issue miniseries and recently reprinted as a single volume. It's another telling of the same battle, but with some very nice visuals. (Not that there's anything wrong with traditional books where you have to make up the visuals, of course.)
You may argue whether Katz has successfully made himself well-informed or whether he's right in his opinions, but you can't say that because he doesn't have a PhD (or he isn't going for one) that he has no right to contribute. On the other hand, if you're just saying he's ignorant and should get informed before venturing an opinion, fair enough. I'm not sure that you were, though.
I felt that I was saying more of the latter. (The silly credentials were more to illustrate the "at least I'm trying to understand" viewpoint.) I can understand how some might see it as more of the former, though.
I just get a bit annoyed at the current trend of people making rules about things they haven't taken the time to study. It ain't as simple as the newspapers make it out to be.
But this is the problem. This discussion shouldn't be in closed circle..geneticists, biologists, clergyman and rabbis..It should include nitwits like me. When you start talking about creating life, or altering life -- sometimes obviously worthwhile things to do, as in eradicating disease -- everybody should be invited to the table for a huge discussion. I reject the patronizing suggestion that only trained ph'd's are smart enuf to to get into this discussion.
Justify that statement. Why should you be included in the discussion? What can you contribute? "I don't understand what you're doing, but I have an opinion anyway." Everyone has an opinion on something. Those comments and emails you're getting are telling you something --- they're saying that you have your basic facts and assumptions wrong, which means you're indulging in lousy journalism.
We wouldn't want a judge who didn't understand law, and many people here are annoyed by software companies run by people who couldn't write "Hello, World". Why, then, should decisions about science be made by people who don't understand the science? An opinion which is formed without being drawn from facts/evidence is fairly worthless, IMHO.
This certainly makes an argument that we need more people who understand science in conjunction with other fields such as ethics, law, and journalism. However, I don't think people are entitled to make uninformed decisions just because they feel scared of the unknown.
Alik Widge MD/PhD Program University of Pittsburgh/Carnegie Mellon University
Would the answers be differnt is we implanted the devices in utero?
Disclaimer: I am a medical student, but not a doctor. This is speculation, not advice. (The same goes for that opinion about craneotomy.)
If we had the devices in right as the brain was growing, yes, I think it would be different. Your brain starts out very neuron-heavy, followed by massive die-off as the neurons compete for access to sensory data. Neurons which don't get stimulated die. So, if you just have some object providing a bit more sensory data, in theory some neurons would contact it and be kept alive, thus providing your brain with an extra I/O port, as it were.
This leaves two big problems. First, you have no idea where that extra port is connected to, unless you have very good control over which neurons make the contact. So, if you implant a brainjack, you can't know in advance what sense it will plug into without some fiddly biochemical manipulations (like the trophic factors used here, but even more so). Imagine trying to do taste-based computing.
The other problem is that you're talking about working with a fetus. Development is a really tricky process. Small disturbances at the wrong time can lead to really nasty birth defects (or they can lead to absolutely nothing). If you're doing as mentioned above and having the device secrete developmental factors, then things get really hairy. It's like trying to patch a Microsoft application as it's compiling without having any access to the source --- all you know is what you can infer from looking at core dumps and random crashes. I personally wouldn't care to try it just yet.
Alik Widge MD/PhD Program University of Pittsburgh/Carnegie Mellon
While his device may require extremely invasive proceedures to implant (anyone know what a Craneotomy is?), it doesn't mean that we can't figure an easier way to get it in there, as we make things smaller.
A craneotomy is the generic name for a procedure which makes a hole in the skull. It's usually something to avoid if possible. (I'll be surprised if genuine cyberpunk-style interfaces are possible without something invasive.)
Alik Widge MD/PhD Program University of Pittsburgh/Carnegie Mellon
I won't buy into this article until I hear about artificial limbs and eyes that work just like real limbs and eyes. Until then, I hold firm in my faith that the nervous system cannot be duplicated with wires.
Actually, WRT to those two apps, you don't need this article. Limbs and eyes do not require any kind of direct brain interface, IMHO. If someone's lost an arm, you ought to attach the interface right at the end of the nerve in the stump. If they're missing an eye, plug into the optic nerve at the back of the orbit. Even a para/quadriplegic doesn't need this; there's work being done on small devices that would essentially serve as "bridges" across breaks in the spinal cord. This, IMHO, is more useful in its capacity to get people to realize that cybernetics is completely possible within most of our lifetimes (including, most likely, SuperGuyAl's).
Alik Widge MD/PhD Program University of Pittsburgh/Carnegie Mellon
This is a field I'm trying to get into, so I've paid some attention to the kinds of neural-interface electronics which already exist. The ones which, IMHO, have the best chance to work require the nerve to be severed and allowed to regrow through the implant so that single axons can be recorded independently. Thing is, when you sever a nerve, it never comes back quite perfectly. To put in such an electrode and then remove it a few months later (as Warwick had to do with his previous chip) is going to be two nerve traumas. Even if he uses a different system than the one I'm thinking of, it's still got to be next to the nerve in order to control the whole arm, and that still leaves plenty of potential for nerve trauma.
I'm neither a doctor nor a trained bioengineer, so I could be wrong, but it seems like Warwick is setting himself up for minor loss of limb function at the very least. I certainly don't consider this technology mature enough for this kind of trial. (Of course, the BBC article doesn't say when he's going to do this experiment --- could be a couple years from now, though they make it sound like tomorrow.)
Re:Heisenburg vs. possibility of nanotech
on
Nanosystems
·
· Score: 2
In fact, we already have nanotech assemblers and their ilk, and have had them for well over two billion years; they're called enzymes. We have enzymes all over the place, unzipping our DNA apart and reassembling it into two copies by finding matching proteins and bonding them to the other half of the stepladder. Basically, a nanomachine can be thought of as an artificial enzyme.
That's true, but there's also one teeny little problem, which is the very different behavior of biological vs. engineered systems. Biology is a lot like that spaghetti code many of us wrote when we started out.
To use your example: DNA replication. It is true that one enzyme unzips the DNA and that there's another (polymerase) which brings in nucleotides that glom on. However, it is a very common event (it's happened to you a couple times while you're reading this) that a copy error is made --- a base gets left out, or the wrong one gets added. This isn't a machine --- it can't perfectly repeat the steps every time. So, there are other enzymes which check the DNA, try to detect mutations, and then try to fix them or delete the strand. If the bad copy gets expressed into protein, there are other enzymes which check the protein and send it off for destruction if it's bad. If the bad protein still makes it into cellular function, your immune system can try to recognize the cell as defective and kill it. If none of that works... time to call your primary care provider.
Your body is making millions of errors at the molecular level every second. It's just that those errors get lost in the huge stream of stuff that manages to work out correctly. There's no machine I know of that works anything like a biological system; some of Rodney Brooks' subsumptionist architectures come close, but no human can build something of the same complexity as the machinery which makes even a single cell pathway work. There are things we can still learn from the biological model, but I'd be very surprised if we ended up doing nanotech using the same architecture. I'd be similarly surprised if we ended up doing AI by making a neural net that simulated the brain. IMHO, we need electronics and nanoengineering to do those things which a biological system can't do well --- the rest can be left up to the wetware.
Alik (Currently reviewing lipids for the next biochem exam...)
This patent exists because they can't determine ahead of time if there will be an error. This patent, inconjunction with one of their other patents, provides a method for them to muddle on in the usual case of no error and still have a means of rolling back in the less usual case of an error.
In conjunction with what someone else said about branches counting as exceptions, this is a start on the answer I'm looking for, but I'm not satisfied. Specifically, there's still a problem of infinite loops. There's no way, AFAIK, to prove that the translation of a given series of target instructions wouldn't send the host into an infinite non-branching loop. Therefore, the host will "muddle on" forever or until the user hits the reboot key. Worse yet, this looping would be totally dependent on the state of registers and memory, and thus is very likely to be irreproducible.
Someone else said that it was a matter of executing a few instructions to see if they screwed up, then committing the memory ops to permanent store, and continuing with that incremental process. However, how does one know at which point the "cause" of an exception or error occurred? It could be that the instruction you executed five minutes ago was the problem. Given that their checkpointing buffer must be finite, this can't possibly catch all translation errors. It's also likely that they can't tell in advance which ones won't be caught, and thus you're gambling every time you run a program. They're pretty good odds, but it's still gambling.
Slashdot Mirror
A few weeks back, I got curious about something: what was the total karma of the AC? I asked Rob. It was, on December 17th, 1999, 1972 points. (Obviously, since the AC doesn't moderate, this is just a sum of moderation.) This means that overall, we've considered ACs to be posting some pretty useful stuff.
When that number goes negative, then it might be time to think about changing the AC system.
Alik
I think I see what's going on here. It looks like the judge issued an injunction against VCR and Ferret, but not against Ripper. Hence, Real decides to claim victory on the best-two-of-three principle, while Streambox claims victory because they can still sell Ripper. Notice that they only talked about "continuing development" on VCR and Ferret, not about releasing them.
-shrug- It's a start, since this yanks away some of Real's content control.
Alik
I think you could could do the same thing to grab speech without speeking that you could do to grab arm movments without moving your arm.. just watch little abortive movements of the vocal cords.
:)
Subvocalization, yes. You are correct that this may be a way to grab speech signals (I dare not say more because I'm thinking of researching in this area), but you still wouldn't get the vocal tones. Also, if you're grabbing directly from small muscle movements, the technology will not be useful for helping persons (like Hawking) with motor disorders. Since the brainjack is most likely to arise out of research for the disabled, I think it unlikely that we'll end up going this route.
You might not care about spech that much though once you got the arm/finger movements working since a 2D symbole buffer is a much better means of communication (and your langauge can adapt to it the same way it adapts to writing instead of speeking).
In which case, you once again don't really need direct neural interface; this sort of thing is the principle behind the various chorded keyboards and "sign languages" that are being experimented with to enable wearables.
Actually, your face and mouth might be the best way to get info to the computer. There are LOTS of seperatly controlled mussles which you could be trained to move individually.
That's certainly true. On the other hand, you'd look pretty weird, especially trying to compute while talking.
Connecting close to the mussle level solves all these problems since you can always choose notto talk.. unless yuou talk to yourself a lot..
Many people can be seen to move their lips when they read. I know that I have a tendency to subvocalize when I'm typing messages. (Of course, I also talk to myself.)
Example: I supposet there might be way to stimulate a section of knowledge to make you recolection of it better, but we could do something like this by having a series of "note cards" which recap the importent theorems or something.
-shrug- We've got that now to one degree or another; my text editor lets me have a "clip book" of common language motifs for all the markup and programming languages I know. I could probably write similar reminders for any new language, or any extensions I make to my existing languages. If you're talking about something that provides better memory by giving people an external store... well, in theory it's possible, but it's a long way off, because we haven't a clue what format the brain uses to encode memories, let alone the signals to access those memories.
It just sems that we don't know that much about our though processes execpt for the parts that work via langauge.
We don't know much about our thought processes, period. The 1990s were supposedly the "Decade of the Brain". We made some progress. The advent of fMRI helped a great deal, as did advances in microbiology and biochemistry. The Genome Project may also help. However, there's a huge amount of work remaining to be done, just as the HGP has sequences, but no clue what those sequences do.
At the intro levels of cognitive science, AI, and robotics, everything looks neat. Then you get inside and discover how little we actually understand. It can get a bit discouraging at times, but it also gives you some job security.
Alik
Hrm... The first time I read that statement, I thought you were saying that couldn't just do that with a computer. The second time I read it I thought you were saying that you could do it with a computer, but not with the brain.
I believe that doing it to the brain is potentially impossible, or at least very very hard. Doing such a thing to a computer is simpler, since computers are simpler devices, but I still don't think it'd be easy.
In any event, you probably could solder more stuff on to a computer motherboard if you really needed to (the memory interface would be a good place to start...)
There's two problems with that analogy. First, the memory interface is just that: an interface. If you've got an open slot, then you've got something more than the brain has. (Yes, the brain has memory, and it's got to be stored somewhere. Maybe someday we'll be able to read/write it directly, but that's even further off than doing image capture from the visual system.)
The second problem is the one you'd have to overcome to try and do the computer task. Let's say you do solder something in between one of the SIMMs/DIMMs and the microprocessor. Every time you send a signal to your device, the memory also sees it. Every time your device writes something to the lines, memory also sees it. Every time you fetch from or write to that memory, your device sees the signal. In all cases, something will be trying to react to a signal it wasn't meant to see. It seems unlikely that you could get this to work even with very clever software, unless the device's purpose was simply to modify the actions of the memory chip.
If that were true, we would all still be single celled organisms, I'd bet. I don't know why our brains would need to be so advanced as to be able to figure out things like relativity, integral calc or the structure of the atom.
I said "superfluous". The ability of our brain to handle abstract thought is clearly not superfluous; it's allowed us to survive and proliferate very nicely. Advanced physics will eventually get us the hell off this planet before it becomes uninhabitable, and therefore the ability to understand it is directly useful for the continuation of the species. However, if there were a population of neurons that wasn't doing anything (like the fallacy that people are only using 10% of their brain), it would be very rapidly selected against.
But what about "sidebanding" the data? I mean, when someone plays quake they don't have to think about there fingers or the monitor in front of them there just "there". In a Sci-Fi Novel I'm writing people interface with computers by 'co-opting' unnecessary nerves in the spinal cord. Would something like this work?
I'm not making the connection between the two. I went and read the parts of your story you've posted. You made reference to "unconnected nerves" --- AFAIK, there ain't no such animal. Part of the biology of neurons is that if they don't receive regular simulation (both electrical and chemical), they die.
You raise an interesting point, though. A lot of this die-off occurs in childhood. If you did one of the neonatal implants and plugged it into some sensory cortex and used it to stimulate and secrete growth factors, you could create an extra population of sense neurons specifically devoted to that implant. The problem is, it's still sharing a bus with the existing sense data. We have no idea whether or not the brain will be able to handle the extra data. (If you used it for something related to that sense, like seeing UV or smelling new compounds, it could possibly be integrated properly.) To test it, though, we'd need to be doing potentially harmful experiments on perfectly normal kids. That's unlikely. It's also a rather limited use of implants anyway. You don't need an implant to see UV or IR; all you need are the right kind of lenses. I think brain implants are very cool, but they shouldn't be used if not necessary; the body never responds perfectly to being messed with.
Alik
How much do they really know about the phonological loop? How diffrent is the voice in the back of my head from the voice that is about to come out?
The answer in the cobwebs of my mind is that there is some linkage, but it's not total. That is, when you're hearing your thoughts in your head or articulating via typing, the hearing/speech zones do become activated, but to a lesser degree than when you actually perform those tasks.
Actually, when one thinks about it, the processes of typing and writing are pretty neat; it's conversion to a completely different output form, and yet my brain knows that speech and text have very similar informational content.
If the process of thinking about moving an arm or saing something really dose triger activity like the actual movement then maybe we could interact with a computer via these abortive movements. This would be the killer app. for brain implants since people could use images and audio in everyday communication.
There is the possibility of doing what you suggest, though we need to improve our tech a bit first. Using it to move a mouse or select symbols is a bit more realistic than having the computer snag entire images and sounds out of your head, though. (And remember, when the sound is in your head, it doesn't have the tones of your voice attached to it yet; those come when it's actually articulated. Thus, if you coded it to audio, the computer would have to do voice synthesis for you. (Stephen Hawking, obviously, would love such a thing. I doubt we'll be able to give it to him before he dies.))
Also, realize the potential problems of doing this. If there's an implant recording your thoughts and sending the data stream to an Internet-connected computer... you think you've got Big Brother looking over you now? Just wait.
Idea: the phase space of the ``possitions of the human body'' is MCUH larger then 3 dimentional. It might be possible to take a mussle group and program a computer to respond to the movements of those mussles as if they were moving an object in a higher dimentional space, then the user might gain some intuition about things which they could apply to solving some open problem.
This is an interesting idea. I don't think it needs direct neural interface, though. Why not just use a control glove or other existing haptic interface? If you need a really large number of DFs, use one of the motion-capture systems they have for animation tasks, like the Flock of Birds.
Alik
Damn straight. My kid's not going to have holes punched in his head either. And that's a big hurdle for this sort of research to clear. Experiments on the human brain are exclusively done to volunteers, usually those who have no real alternative to this sort of surgery. And let's face it, how many people want to have their brain poked at?
Actually, I should amend my original comment. I would not volunteer a child of mine for such an experiment given the current state of the tech. However, if I knew that my child would be blind from birth, and if such implants were highly successful in adult humans, and if neonatal implants had already been tested and proven in other species, then I'd be willing to go ahead with it.
As you say, volunteers are a hurdle. However, there are plenty of people with various kinds of sensory and motor disabilities out there. Implantation of electrodes isn't pleasant, but it's usually not life-threatening. Given the tradeoff, I think there'll be sufficient volunteers to establish clinical usefulness once this field takes off (which seems likely to happen Real Soon Now).
Alik
Basically, what you were saying was "The brain has never been able to use anymore then 5 senses." That's true, but until we hook more stuff up to it, we'll never know.
Right, but that's like saying "My PC which has no expansion slots has never been able to use anything more than the default hardware, but until I crack the case and start soldering stuff onto the traces on the motherboard, I'll never know." It's not that simple. For one thing, where would you plug in new inputs? I'm in neuroscience (admittedly only at the beginning grad level), and I can't think of a place.
There's also the possibility of ether "growing" new lobes for new things or "emulating" the lobes in hardware. Or we could just plug new senses into the visual cortex or something (would it be another type of sight then?)
Well, first off, you can't really say that a given function exists in a given lobe, because they're shipping signals all over the place. Secondly, even if you grow a new set of tissue to plug your new port into, you still need to teach the brain to integrate that signal, and that's a process which is not really understood at all. To some degree, we'd have to solve the problem of where consciousness is.
Your comment about plugging in new things to the visual pathway (or another sense pathway) makes the most sense, and I think that this is how most "new senses" will end up. If we could actually get massive arrays of microelectrodes (and the software to configure and drive them), then it'd be possible to overlay stuff on the retinotopic maps in realtime, just like CBS does to their video feed. (Yes, we will cover your significant other's body with ads for our new porno site. Ain't technology wonderful?)
Anyway, the human brain can do *a lot* more then is evolutionarily needed. I'm sure that it could be augmented somewhat by technology.
It depends on how one parses your statement. Yes, the brain can think of a whole lot of stuff. On the other hand, there's evidence that suggests that it basically handles all that stuff using a few limited pathways. It all mostly comes down to efficient pattern-matching anyway. However, that does not imply that the brain has a lot of extra circuitry lying around which one can just co-opt at will. Evolution is not kind to superfluous stuff.
Alik
I just looked back and realized that they said he had this implanted in 1978 -- Does that seem like an awful long time ago (for this sort of thing) to anyone else?
Not really. Ever since we realized the brain was electric, people have been stimulating it like mad and seeing what they can put in and take out. The fact that it's taken us this long to get this far should tell you something about how hard a problem it is. (Consider how much trouble we still have with the problem of computer vision. It's getting better, but it's nowhere near a solved problem, IMHO.)
If they could do a 10x10 pixel image back then, what are they capable of now?
Well, speaking as someone who's sort of part of they... a 10x10 pixel image. That's why this is news --- after two decades of trying, computer tech has finally gotten to the point where we can give Jerry useful input. However, the science of brain electrodes hasn't advanced that much. They're more durable now and often thinner, but in practice, we still probably wouldn't be able to sink nearly enough into Jerry's visual cortex to convey a complete visual image. (On the other hand, there is the example of the cat. However, that cat was never expected to have long-term survival.) We may be jacking in within our lifetimes, but it's not going to be as soon as you hope.
Alik Widge
MD/PhD Program
University of Pittsburgh/Carnegie Mellon
Makes you wonder if, as a baby, you had some strange thing (IR port, GPS, radio) wired into your brain just after birth, would you learn how to use it, just as you learn how to stand up, talk, and focus your eyes?
My training is nowhere near complete, but my initial guess would be no. The brain does have dynamic connection capacity, but that capacity has always been set to deal with the five senses, all the way down the evolutionary tree. The brain has all the input ports connected to senses already, and as far as we know there's no place God left for us to install new peripherals. Thus, any kind of neural-interface tech is likely to work using the existing sense inputs.
Now, that doesn't mean there can't be some kind of extra output added, as was the case with that guy down in Georgia. It is likely that people will eventually be able to control computers through thought. However, chances are that the output from the computer will still come back through the same sense lines. (It's possible, I suppose, that someone might figure out exactly where the "phonological loop" of short-term memory is (that's the part you're using when you hear your own voice in your head), decode the representations of all known phonemes, and start injecting thoughts in via electrodes. However, that sort of capability is in the very far future.)
As for putting something into a kid so they'd be naturally adapted to it... might not work as well as you'd think. Yes, kids' brains are more adaptive. However, kids' brains are also still developing, and thus really sensitive to disruption. A little bit of extra electrical stimulation at the wrong time can seriously fuck things up. I personally would not volunteer a child of mine for such experiments, even if I'd designed the device myself.
Alik
Reread the analysis. The guidelines are mainly intended to address expressed sequence tags, or ESTs. These are fairly small bits of DNA which let you quickly identify areas of the genome. To use a somewhat-sloppy analogy, they're like regexps for the genome.
Now, imagine that I spend a few months compiling a whole bunch of useful regexps. I think I'd be within my rights to want to profit from all that work. Now, I can't patent a regexp, because in most cases it falls under obviousness. However, DNA tags are a lot less obvious, and therefore require significant effort to find and construct. Therefore, some companies claim they should be able to patent the ones they discover so that they can make money selling reagents to biologists.
ESTs, to the best of my knowledge, cannot be considered genes. They may code for a few amino acids, but as far as I know none of them codes anything near a complete enzyme.
Also, in response to the "they'll patent the cure for Alzheimer's" and "they'll patent the diabetes gene" worries: these are multifactorial diseases. It's not just one gene at the heart of them. Furthermore, it's not just one mutation causing all the cases. There are both genetic and environmental triggering factors.
Now, it's certainly true that a company might get a patent on a given kind of gene therapy or a "healing" DNA sequence that they've constructed. IMHO, this is fair. Such a sequence would be essentially equivalent to a drug, and anyone who figures out a cure for a disease deserves some profit.
Nobody is going to patent your genes and then sue you for infringement-through-living. Or rather, they might do that, but there's no way a court will even grant them the preliminary injunction, let alone damages. (If you are a genetically-engineered perfect child using "upgraded" genes that the company invented, *then* you might be in trouble.) As far as medical therapies go, gene patents are probably going to just perpetuate the system of drug patents.
Finally, a semi-offtopic response to someone who said that the genome is open-source but encrypted: I don't think it's encrypted at all. God's just not very good at commenting his code.
Alik
Furthermore has anyone even _heard_ of someone that bought something because of SPAM?
Yes. My father. Twice in the past month. First, he was spammed by one of the telcos offering a low rate. (I think it was GTE, but I'm not certain.) Problem is, it's a legitimate company, not some fly-by-night, and it really was a better rate than we have. He's seriously considering the switch.
Just this weekend, we got spammed about DSL. He wants DSL. They're offering a rate that's cheaper than our current two-analog-line bill from Hell Atlantic. I think he won't end up getting it simply because we live in an area where DSL is not yet available, but otherwise, that'd be another successful spam.
The only reason for him to not take such offers is conscientious objection. However, like most Americans, my father doesn't give a rat's ass about the way a few geeks want the Internet to be; he just wants BetterFasterCheaperNOW. Worst part is, he's an engineer. He does have some understanding of why certain things are problem (he's been subject to things like the One-Click patent in his own field), but considers it foolish to pass up a deal just because you dislike a company's business practice.
If John Wanker receives a "FREE TEEN Pr0n!!!!!" message in his inbox, he probably clicks it and racks up another banner count for the spammer. He might even sign up for the pay service.
Spam is used because spam works. It's got marginal costs that are almost zero, to the point where a 1/10,000 response rate to your ad will more than pay for the cost of the mailing. It may digust and annoy us, but like so many things, it'll exist as long as it's profitable.
Alik
P.S. I used hacker intentionally here so please no "use cracker" posts. It is a lost cause the rest of the nation has already adopted the new vocabulary. Besides its just a word why do you care?
It may be just a word, but it's our word. (Relatively speaking; I wasn't even born yet when the first hackers were being named.) Moreover, since words are the method of communicating thought, words shape that thought. This is one of the reasons why Newspeak was implemented in _1984_.
Consider the original definition of "hacker", a person who makes furniture using an axe. Obviously, there's a sense of doing a lot of bashing around with crude tools. However, the final product of this, when done properly, is useful and often aesthetically pleasing. (For a modern-day example, consider those who make wood sculpture with chainsaws.) In many ways, I think this captures the way a "true" hacker works.
Now consider "cracker". The best I can come up with is "A tasty hard bread-like thing" or "Someone who goes around breaking things." There's different mental associations with these words, and I personally think crackers should be stuck with the label that fits their activities. (This is why the guys from l0pht and CoDC get to call themselves hackers as far as I'm concerned.)
Furthermore, your claim that the nation has adopted the new meaning is not quite true. Over the past year, an increasing number of stories referenced on Slashdot have been using "systems cracker" instead of "hacker" when referring to someone involved in digital intrusion. As has been pointed out in the AOL/TW threads, most people will swallow whatever the media feeds them. It appears that media outlets are slowly learning to distinguish, and that will easily be absorbed by media consumers.
Go thou, read _Snow Crash_, _1984_, and _Brave New World_, and then watch _Brazil_. You'll understand that there's very little which is "just a word".
Alik
I think the problem with alot of the .coms is they are run by computer geeks who've never been to business school and have only read a couple books on the subject.
Unless you've got some stats to back this up, my experience contradicts your statement. From what I've heard (and I admit to having no backup other than a few articles in the newspaper), most new dotcom people are not geeks. At least part of this is because so many of the new companies are based around business models that involve things like "Enter our contest to win a free plastic trinket by giving us a sample of your DNA and the email addresses of everyone in your department!" or "You must have IE 9.0 and enable every single feature in it to use this site." Such things seem to be inherently annoying and disgusting to the geeks I've met. (Of course, it varies.)
Many business school profs are currently complaining about the dropout rate. Students start school, get some random idea for a new thing to sell over the web, and drop out to start the company and grab the domain before someone else does. It seems to me that this is where a good chunk of the expansion in e-business is coming from. (The exceptions would be sites that actually feature new technology; Google is one such. Most new commercial sites aren't featuring any new algorithms, though; they seem to just be variations on the catalog, fact-database, and auction themes.)
Also, a completely random thought that just popped into my head. The current buzzword for e-commerce companies is "dotcoms". Why didn't we end up with the word "dotcommerce" as well?
Alik
The only change we are asking for, is that those
who believe they are qualified to not go that
route, be allowed to not go that route.
I understand this. However, if such permission were enacted, I think the majority of people would end up trying to self-prescribe. They might be trying to save the cost of a doctor's visit, they might have an inflated opinion of their own intelligence, or they might get swayed by corporate propaganda.
For evidence, consider the current craze for herbs and nutritional supplements. None of it has to be approved by the FDA to treat anything. It can all be bought without prescription. The result is that people are wolfing down tons of pills that in many cases aren't likely to have more than a placebo effect. (This is not always true, but it is often true.) There have even been cases where people died because they neglected standard medicine in favor of alternative medicine and herbal supplements.
The above practices, although potentially useless, also appear to be free of dangerous side-effects. The currently regulated medications are not nearly so benign. I count myself as libertarian, but this seems like a case where the average citizen needs to be protected from the power of advertising.
Not true. How does being apointed make a person
less greedy? Its true it makes them possibly
more suited to the job (having been selected
not by popularity contest but by selection based
on qualifications) but...once there... them
dead green presidents look mighty sweet no
matter how you got your job.
True, they are still vulnerable to bribes. However, bribery is not the main form by which money influences government, at least not at the federal level. Most of the interfering money, at least as far as I've seen (and I spent six months working for two Congressmen, three of them helping to manage a campaign), comes in the form of campaign contributions. Campaign contributions aren't really relevant to an appointed official, for obvious reasons.
Of course, if you're looking at bribery within the system, you might also look at all the free food and teeny promotional gizmos the drug company reps give to the doctors...
Alik
Um, Alik?? Prescription medications are *already* regulated in every state. You haven't explained why we need federal control.
IMHO, the Feds have to be invoked in this case because we're not dealing with transactions within one state. Most e-pharmacies are trying to sell to the entire nation at once, which suggests the need for national-level regulation. If nothing else, it clearly brings in the interstate commerce authorities.
p.s. yes, sometimes private entities fail. However, those that fail go out of business, and their owners fall into disrepute. When government fails to properly regulate, does it ever go out of business? Do any government employees suffer?
I haven't seen TRUSTe go out of business...
Aside from that, you make a fair point. Government is poor at noticing when something is broken. However, I still claim that the government is likely to do a better job than a private entity; see next post.
Alik
I'll admit to being a bit biased, being in med school and all that. However, I think the points that some other posters have made need to be reinforced. It's true that in an ideal libertarian society, there'd be no need for any kind of prescriptions; you could just find a pharmacy and buy what you wanted. The problem is that this ideal depends on having an educated citizenry. Issues of intelligence aside, in order to be educated about a medication, you need valid information. How likely do you think it is that the pharmaceutical industry, if left unregulated, would be willing to provide negative information about their products? The chemical structures of drugs would probably be locked up as "trade secrets". This is a fairly clear case of abuse of power. (Remember the days when people went around in wagons selling their Patented Cure-All Tonics?) Hence, the Food and Drug Act, which has had a net benefit for the health of Americans.
It's true that people should take responsibility for their actions. However, there's a limit to how much self-education you can expect someone to do. It's not reasonable to expect every person to keep perfectly up-to-date on biochemistry, consumer electronics, economics, and every other field of study which might affect his or her life. The volume of information out there is just too huge; this is why the pundits say we're in the Information Age. Therefore, like a good little capitalist economy, we specialize in various fields of knowledge. The problem comes when we can no longer trust a given person to have or use the knowledge their role implies. Without the ability to trust individuals or corporations directly, we have to fall back to the trusted-third-party model.
One poster suggested that instead of government, we rely on corporate entities (profit or nonprofit) whose job is to certify credentials. That sometimes works, but not always --- witness the recent problems with TRUSTe. In this case, the Federal Government doesn't have anything to gain by playing favorites, and therefore it can probably be trusted. (The FDA, IIRC, is made up of appointed officials, not elected ones, making it a bit harder to corrupt with money.) Therefore, the Feds might not be the optimal solution, but they strike me as better than any proposed alternative.
Alik Widge
MD/PhD Program
University of Pittsburgh/Carnegie Mellon University
Those of you who enjoy comic books may wish to check out Frank Miller's _300_, published
last year as a 5-issue miniseries and recently reprinted as a single volume. It's another telling of the same battle, but with some very nice visuals. (Not that there's anything wrong with traditional books where you have to make up the visuals, of course.)
You may argue whether Katz has successfully made himself well-informed or whether he's right in his opinions, but you can't say that because he doesn't have a PhD (or he isn't going for one) that he has no right to contribute.
On the other hand, if you're just saying he's ignorant and should get informed before venturing an opinion, fair enough. I'm not sure that you were, though.
I felt that I was saying more of the latter. (The silly credentials were more to illustrate the "at least I'm trying to understand" viewpoint.) I can understand how some might see it as more of the former, though.
I just get a bit annoyed at the current trend of people making rules about things they haven't taken the time to study. It ain't as simple as the newspapers make it out to be.
Alik
But this is the problem. This discussion shouldn't be in closed circle..geneticists, biologists, clergyman and rabbis..It should include nitwits like me. When you start talking about creating life, or altering life -- sometimes obviously worthwhile things to do, as in eradicating disease -- everybody should be invited to the table for a huge discussion. I reject the patronizing suggestion that only trained ph'd's are smart enuf to to get into this discussion.
Justify that statement. Why should you be included in the discussion? What can you contribute? "I don't understand what you're doing, but I have an opinion anyway." Everyone has an opinion on something. Those comments and emails you're getting are telling you something --- they're saying that you have your basic facts and assumptions wrong, which means you're indulging in lousy journalism.
We wouldn't want a judge who didn't understand law, and many people here are annoyed by software companies run by people who couldn't write "Hello, World". Why, then, should decisions about science be made by people who don't understand the science? An opinion which is formed without being drawn from facts/evidence is fairly worthless, IMHO.
This certainly makes an argument that we need more people who understand science in conjunction with other fields such as ethics, law, and journalism. However, I don't think people are entitled to make uninformed decisions just because they feel scared of the unknown.
Alik Widge
MD/PhD Program
University of Pittsburgh/Carnegie Mellon University
Would the answers be differnt is we implanted the devices in utero?
Disclaimer: I am a medical student, but not a doctor. This is speculation, not advice. (The same goes for that opinion about craneotomy.)
If we had the devices in right as the brain was growing, yes, I think it would be different. Your brain starts out very neuron-heavy, followed by massive die-off as the neurons compete for access to sensory data. Neurons which don't get stimulated die. So, if you just have some object providing a bit more sensory data, in theory some neurons would contact it and be kept alive, thus providing your brain with an extra I/O port, as it were.
This leaves two big problems. First, you have no idea where that extra port is connected to, unless you have very good control over which neurons make the contact. So, if you implant a brainjack, you can't know in advance what sense it will plug into without some fiddly biochemical manipulations (like the trophic factors used here, but even more so). Imagine trying to do taste-based computing.
The other problem is that you're talking about working with a fetus. Development is a really tricky process. Small disturbances at the wrong time can lead to really nasty birth defects (or they can lead to absolutely nothing). If you're doing as mentioned above and having the device secrete developmental factors, then things get really hairy. It's like trying to patch a Microsoft application as it's compiling without having any access to the source --- all you know is what you can infer from looking at core dumps and random crashes. I personally wouldn't care to try it just yet.
Alik Widge
MD/PhD Program
University of Pittsburgh/Carnegie Mellon
While his device may require extremely invasive proceedures to implant (anyone know what a Craneotomy is?), it doesn't mean that we can't figure an easier way to get it in there, as we make things smaller.
A craneotomy is the generic name for a procedure which makes a hole in the skull. It's usually something to avoid if possible. (I'll be surprised if genuine cyberpunk-style interfaces are possible without something invasive.)
Alik Widge
MD/PhD Program
University of Pittsburgh/Carnegie Mellon
I won't buy into this article until I hear about artificial limbs and eyes that work just like real limbs and eyes. Until then, I hold firm in my faith that the nervous system cannot be duplicated with wires.
Actually, WRT to those two apps, you don't need this article. Limbs and eyes do not require any kind of direct brain interface, IMHO. If someone's lost an arm, you ought to attach the interface right at the end of the nerve in the stump. If they're missing an eye, plug into the optic nerve at the back of the orbit. Even a para/quadriplegic doesn't need this; there's work being done on small devices that would essentially serve as "bridges" across breaks in the spinal cord. This, IMHO, is more useful in its capacity to get people to realize that cybernetics is completely possible within most of our lifetimes (including, most likely, SuperGuyAl's).
Alik Widge
MD/PhD Program
University of Pittsburgh/Carnegie Mellon
This is a field I'm trying to get into, so I've paid some attention to the kinds of neural-interface electronics which already exist. The ones which, IMHO, have the best chance to work require the nerve to be severed and allowed to regrow through the implant so that single axons can be recorded independently. Thing is, when you sever a nerve, it never comes back quite perfectly. To put in such an electrode and then remove it a few months later (as Warwick had to do with his previous chip) is going to be two nerve traumas. Even if he uses a different system than the one I'm thinking of, it's still got to be next to the nerve in order to control the whole arm, and that still leaves plenty of potential for nerve trauma.
I'm neither a doctor nor a trained bioengineer, so I could be wrong, but it seems like Warwick is setting himself up for minor loss of limb function at the very least. I certainly don't consider this technology mature enough for this kind of trial. (Of course, the BBC article doesn't say when he's going to do this experiment --- could be a couple years from now, though they make it sound like tomorrow.)
That's true, but there's also one teeny little problem, which is the very different behavior of biological vs. engineered systems. Biology is a lot like that spaghetti code many of us wrote when we started out.
To use your example: DNA replication. It is true that one enzyme unzips the DNA and that there's another (polymerase) which brings in nucleotides that glom on. However, it is a very common event (it's happened to you a couple times while you're reading this) that a copy error is made --- a base gets left out, or the wrong one gets added. This isn't a machine --- it can't perfectly repeat the steps every time. So, there are other enzymes which check the DNA, try to detect mutations, and then try to fix them or delete the strand. If the bad copy gets expressed into protein, there are other enzymes which check the protein and send it off for destruction if it's bad. If the bad protein still makes it into cellular function, your immune system can try to recognize the cell as defective and kill it. If none of that works... time to call your primary care provider.
Your body is making millions of errors at the molecular level every second. It's just that those errors get lost in the huge stream of stuff that manages to work out correctly. There's no machine I know of that works anything like a biological system; some of Rodney Brooks' subsumptionist architectures come close, but no human can build something of the same complexity as the machinery which makes even a single cell pathway work. There are things we can still learn from the biological model, but I'd be very surprised if we ended up doing nanotech using the same architecture. I'd be similarly surprised if we ended up doing AI by making a neural net that simulated the brain. IMHO, we need electronics and nanoengineering to do those things which a biological system can't do well --- the rest can be left up to the wetware.
Alik
(Currently reviewing lipids for the next biochem exam...)
In conjunction with what someone else said about branches counting as exceptions, this is a start on the answer I'm looking for, but I'm not satisfied. Specifically, there's still a problem of infinite loops. There's no way, AFAIK, to prove that the translation of a given series of target instructions wouldn't send the host into an infinite non-branching loop. Therefore, the host will "muddle on" forever or until the user hits the reboot key. Worse yet, this looping would be totally dependent on the state of registers and memory, and thus is very likely to be irreproducible.
Someone else said that it was a matter of executing a few instructions to see if they screwed up, then committing the memory ops to permanent store, and continuing with that incremental process. However, how does one know at which point the "cause" of an exception or error occurred? It could be that the instruction you executed five minutes ago was the problem. Given that their checkpointing buffer must be finite, this can't possibly catch all translation errors. It's also likely that they can't tell in advance which ones won't be caught, and thus you're gambling every time you run a program. They're pretty good odds, but it's still gambling.