Forgot your password?
typodupeerror

Comment: Re:Bluffing? (Score 1) 693

by atmtarzy (#34271894) Attached to: 200 Students Admit Cheating After Professor's Online Rant

Because keeping network logs specific enough to definitively determine cheating is totally not invasion of privacy.

...

Maybe if the document were shared via public Google Docs or so, and the university logged a mapping of university accounts (presumably students have to have one to access the university's network) -> URIs, then you're still on the violation side of privacy protection. Or maybe if the university somehow can read a student's university email without the student's consent or knowledge without it being considered a privacy violation, but even then you're only catching the kids dumb enough to use the university's mail service for cheating. Also consider that students have no selective power over which emails their account receives, so you'd have to check to see if the email was read, and even then you wouldn't know if the student "read" it just to delete it, or actually read it, and used the material to cheat.

Comment: Re:Why E.coli? (Score 5, Informative) 86

by atmtarzy (#34263610) Attached to: Problem-Solving Bacteria Crack Sudoku

Cultivated strains (e.g. E. coli K12) are well-adapted to the laboratory environment, and, unlike wild type strains, have lost their ability to thrive in the intestine. Many lab strains lose their ability to form biofilms.[70][71] These features protect wild type strains from antibodies and other chemical attacks, but require a large expenditure of energy and material resources.

Basically the E. coli K12 gets totally owned by our immune system, as in before it has a chance to cause much damage, as in it doesn't make us sick, as in it is not "associated with human disease". In an abstract sense, saying, "K12 is 'associated with human disease' because O157:H7 is (and probably others are) associated with human disease," is very much like saying, "garden snakes/<insert relatively harmless snake> are associated with human death because black mambos/king cobras/<insert other deadly snake> is associated with human death." More colloquially put, "OMG it's a snake! It's going to kill me!" and "OMG it's E. coli! It's going to make me sick!" have the same logical flaws.

Comment: Re:no shocker (Score 1) 551

by atmtarzy (#33763238) Attached to: Animal Farms Are Pumping Up Superbugs

What do you do when a portion of the bacterial population evolves to defend against A in a way that protects it against B, and part against C?

If I understand you right, you might encounter a situation in which, say, half the cows have a variant of an infection which is A-resistant (requiring B-treatment) and the other half have a B-resistant variant (requiring A-treatment). I would agree this would be difficult to handle in a way that gives every cow the same antibiotics. However, if the antibiotics are designed correctly, then there's really only one or two remotely probable ways to evolve from being susceptible to A, and those ways can be treated by B, C, D, etc. which in turn only have a few probable ways to evolve away from, and eventually you'll be able to loop those evolutions back in on themselves.

I think your point would still stand though, because the evolution of the diseases in different cows would happen at different rates (though I imagine those rates won't differ by very much relative to how often antibiotics are given, so perhaps this is a moot point). Some cow might have a slowly-evolving disease still close to A-state but another cow might have a quickly-evolving disease closer to B-state, at which point it's difficult to supply just one medication to the entire herd. This might be countered with a triangular "circle", so that the disease evolves from A->B->C->A. It's unlikely that a disease would still be barely past A-state in one cow, but another cow would be entirely past C-state and ready to move back into A-state.

Obviously doing this is extremely precarious, as it's pretty easy to muff up the timing of drug treatments and have cows with diseases in all sorts of evolutionary states, which would make it significantly easier for a superbug to evolve. Treating each cow on a case-by-case basis would work, but that's more expensive.

what about a bacteria that is specifically designed to...

The HIV would evolve to differentiate between the 'tar pit' bacteria and the cells it typically infects. If it were somehow impossible to make that differentiation, the body itself wouldn't be able to tell the difference, and wouldn't be able to effectively regulate white blood cell levels while the treatment is taking place, which would basically cause AIDS in the person you're trying to treat for HIV. After a while, the levels of HIV in the system would hopefully decline as the 'tar pit' bacteria clean them up, but I don't think the HIV levels would ever go lower than a certain equilibrium point with the white blood cells and 'tar pit' bacteria, so the person's white blood cell count would never be able to return to normal. I have no idea if the difference between not-quite-normal and normal is enough to care about though. I think inducing AIDS to treat HIV would completely defeat the purpose though, even if the AIDS was only temporary.

Some sort of hybrid between the 'tar pit' bacteria and the circular evolution technique might be able to force HIV levels to go low without nuking the body's white blood cell count though. Maybe then the body will have enough time to develop its own vaccination against HIV? That'd be pretty damn cool.

Comment: Re:no shocker (Score 1) 551

by atmtarzy (#33761678) Attached to: Animal Farms Are Pumping Up Superbugs

AFAIK, current HIV medications are given to counter the evolution of HIV. Instead of simply stopping evolution, they get HIV to "evolve" in a circle. eg. present antiviral A where HIV tends to evolve to adapt to A with some mechanism, then after a while, switch to antiviral B that kills HIV through that mechanism, causing the HIV to evolve back to its former state.

It's more complicated than that (eg, say the HIV evolves a mechanism that can counter A and B, to which you would respond with a C that causes the HIV to evolve back to something A works well against), but the general idea of making HIV evolve in circles is the important part.

If the antibiotics used in cows and such were designed to work this way, then this whole superbug issue would be significantly less important. It's probably a lot more expensive (in the short term), so you'd need the FDA to step in and start requiring drug treatments that function by the circular evolution method. obviously in the long term, using antibiotics against superbugs will be extremely difficult and probably even less cost-effective than the drugs that cause circular evolution (which might not work when the superbug is super-resistant to every super-awesome super-antibiotic we've thrown at it).

Comment: Re:Don't just slap some tape on it. Fix it. (Score 1) 102

by atmtarzy (#33694800) Attached to: Deodorant Sought to Save New Zealand's Native Birds

Unless this odor is a significant component of mating (to the point where birds with odor won't mate with birds without), removing it won't create a new species, at least by the (objective) biological definition of a species. It simply forces an adaptation onto an existing species. Preliminary testing on small populations should be done to ensure the adaptation actually benefits the species so that no unintended side-effects surprise us later. So unless this situation is more complex (eg reproductive rates are drastically lower in birds without the odor, or the predators have an alternative method to identify the birds), then the birds with this new adaptation would quickly grow back to near the population level they were at before the predators entered their environment.

If the odor is used in mating, there isn't much we can do without going really hardcore, and somehow remove the role the odor plays in the reproductive/mating process. Also, if odor is a requirement in mating, then preliminary testing should identify it and we'll realize the idea is a bust before we have any significant negative impact on the population.

If there are multiple methods for predators to identify the birds, then we'd have to research how to remove all of them, and hope all of them play no major role in the mating process. From there, it's just cost/benefit analysis to decide whether it's worth the time/money to research adaptations for all those methods, which should be in the hands of whoever is deciding whether to grant the funding for the research.

Comment: Don't just slap some tape on it. Fix it. (Score 1) 102

by atmtarzy (#33693048) Attached to: Deodorant Sought to Save New Zealand's Native Birds

So we're going to remove any incentive for these birds to evolve by deodorizing them? It's highly unlikely we'll be able to deodorize every bird all the time, so there will be birds that we miss. Only the birds we miss, and miss often, will be under pressure to evolve. This is a significantly smaller population. Evolution only works well on large populations. So this species will be significantly less likely to evolve, which means the most likely scenario is humans deodorizing these birds for the rest of time, or until we come up with some other solution.

I say we spend the money researching which genes create the odors, find a way to disable them, steal some eggs, modify the genetics of those eggs to disable the odor-creating genes, and then let natural selection work its magic.

That's all assuming investing all that money is even worth keeping this bird species around.

Comment: Re:Nothing will happen (Score 2, Interesting) 360

by atmtarzy (#29346147) Attached to: Lawsuit Claims WGA is Spyware

Then maybe they shouldn't break the law.

I think if the guys with the power to make decisions at MS could chose between making $100M and killing the global economy, or not, they'd take the $100M, quickly get it put into their Swiss bank account, and retire in Switzerland while the rest of the world goes to hell. Maybe I'm being cynical, but if MS (or any other overly-huge corporation, like say AIG pre-recession) were to just disappear from the global economy, it'd be like ripping a kidney out of your body. You just might survive, but it sure is not going to be pleasant. If we handle the situation differently, and slowly kill MS off (by essentially shutting down everything but say support and whatever people are necessary to keep the systems relatively secure), the rest of the world will be much more able to adequately adjust.

Comment: Re:Plausible denability? (Score 1) 45

by atmtarzy (#29151025) Attached to: Scientists Find Way To Combat Forged DNA

They would need some other way of proving you were at the scene to say you were. You can't argue that just because the one bit of fake DNA they found was trying to be yours, you couldn't have been at the crime scene. At least that sounds like the logical solution. Best I can do is hope that's how the legal system actually sees it.

Real Programmers don't write in FORTRAN. FORTRAN is for pipe stress freaks and crystallography weenies. FORTRAN is for wimp engineers who wear white socks.

Working...