i've suspected this for a while, but i don't really know how to start researching it; googling yields too much noise and propaganda. do you have any links/cites/summaries?
to be honest, i don't know, but let's look at the facts. i can tell you that i almost certainly wouldn't have been able to come up with google (and if i could have, it would have taken much, much longer than it took brin & page), and that google represents a true leap of technology. so, we've established that i don't attribute everything i can't do to luck.
on the other hand, facebook is more or less just a disney-fied version of livejournal and myspace. not that this is bad; it's obviously tremendously popular and maybe it should be, and i guess it's an innovation of sorts, but to me it's just not in the same class as google.
i can explain to you what's impressive about google. why don't you explain to me what's impressive about facebook?
since we're already veering off-topic and into emotional appeals: is zuckerberg really that "hardworking"? he did something pretty run-of-the-mill and got lucky being in the right place at the right time. this happens all the time, usually on a smaller scale, and probably shouldn't be "fixed," but hardworking? really? as opposed to everyone else?
since the gender emphasis seems to have been added exclusively by slashdot, you can stop weighting it, unless you think slashdot controls redhat management. i'm sure you'll find something else to fuel your paranoia though.
yeah, but there would be a few advantages to piggybacking on facebook, like automatic redundancy; when it comes to crooks, it's usually not what's possible, but what's easy. crooks are usually dumb; slashdotters don't get that.
but yeah, the primary reason is probably to put the kibosh on casual file-sharing, and i can't blame them for that.
that's true, but even then facebook will recompress your jpeg even if it's the "right" dimensions. they might even being do this expressly to defeat steganography (in addition to saving disk); research would be required. the standard steg algs can't survive a recompression, although should be doable in principle.
it depends what you mean by text data mining. yeah, you can grab keywords, and there are some simple clues about proximity of certain simple adjectives, and you can sort of associate certain vocabularies with income and spending habits, but the R^2 is pretty low. text mining is far, far away from "understanding the contents of every message." even google does a shoddy job; many of its text mining-based ads are silly and even insulting.
most of the marketing-juice comes from (surprise, surprise) the social network. facebook has trained people (maybe not you, but probably many of your "friends") to advertise themselves! if you're 1 hop away from 6 people who all explicitly "Like"d some expensive imported chocolate or coffee, that will probably tell me a whole lot more (marketing-wise) about you than any 100 of your messages, even if i had a human being reading every one of them, which text mining is nowhere near.
in most subjects i would agree with you, but i don't think i would want an imaginative doctor (at least not at the expense of a strong level of basic competence). some things damned well should be done by rote, based on centuries of hard-won experience.
some people do have to come up with the new stuff, but most doctors don't and shouldn't be trying.
i just don't understand this. it's not even an electronics project; it's sticking two totally arcane black boxes together. there's no particular logic or thinking or knowledge involved beyond the chip layout. it's just a tedious process involving luck, dexterity, and a very specialized skill that doesn't generalize much.
why the hell would anyone want to do this? it's certainly not curiosity, since that's generally about not doing things which a fucking robot can do 1000x better.
if you wanted to start a cottage industry repairing the sure-to-be ubiquitous raspberry pis (lol), that's a reason i guess, but you can just wait until it happens and you'll have plenty of junked parts to practice on.
well, of course you're not going to get what you want. you're going to get what optimizes the expected portfolios of large interests, which might coincide.
this is exactly why i wish the Rs would outgrow this childish denialism. if we're changing the environment, then let's try to do it optimally. if we're not changing the environment, then why the hell aren't we?! the chances of us being in the "best of all possible worlds" is almost zero, so why aren't the Rs doing what they do best and funneling massive public monies into private industry to change the environment? as a bonus, the old-school energy companies they've been buddying-up with have the best infrastructure to get on this project and become climate-control companies.
it's been a while, but i think that for PCR you need to know the flanking sequences. you don't need to know the "middle", but you need the primer to match the ends.
that said, it seems that even with just PCR you could do dimensionality reduction on the code sequence... i.e. if you try PCR with an arbitrary primer, and it amplifies significantly, then you have some information about the sequence. it's not like a crypto key, where it's all or nothing. i suspect that chip-based sequencing and some statistical algos could make quick work of this "code", but like i said it's been a while and the paper provided is... uh... mostly advertising.
Slashdot Mirror
half, huh? i didn't know there were that many people living on capital gains.
i've suspected this for a while, but i don't really know how to start researching it; googling yields too much noise and propaganda. do you have any links/cites/summaries?
yeah, you can get drunk on angostura bitters from the supermarket too, yet you aren't carded for it.
to be honest, i don't know, but let's look at the facts. i can tell you that i almost certainly wouldn't have been able to come up with google (and if i could have, it would have taken much, much longer than it took brin & page), and that google represents a true leap of technology. so, we've established that i don't attribute everything i can't do to luck.
on the other hand, facebook is more or less just a disney-fied version of livejournal and myspace. not that this is bad; it's obviously tremendously popular and maybe it should be, and i guess it's an innovation of sorts, but to me it's just not in the same class as google.
i can explain to you what's impressive about google. why don't you explain to me what's impressive about facebook?
since we're already veering off-topic and into emotional appeals: is zuckerberg really that "hardworking"? he did something pretty run-of-the-mill and got lucky being in the right place at the right time. this happens all the time, usually on a smaller scale, and probably shouldn't be "fixed," but hardworking? really? as opposed to everyone else?
since the gender emphasis seems to have been added exclusively by slashdot, you can stop weighting it, unless you think slashdot controls redhat management. i'm sure you'll find something else to fuel your paranoia though.
yeah, but there would be a few advantages to piggybacking on facebook, like automatic redundancy; when it comes to crooks, it's usually not what's possible, but what's easy. crooks are usually dumb; slashdotters don't get that.
but yeah, the primary reason is probably to put the kibosh on casual file-sharing, and i can't blame them for that.
that's true, but even then facebook will recompress your jpeg even if it's the "right" dimensions. they might even being do this expressly to defeat steganography (in addition to saving disk); research would be required. the standard steg algs can't survive a recompression, although should be doable in principle.
fair enough. nonetheless, it does keep away the more vulgar attempts. for example, no one bothered to develop facebookFS. ;-)
it depends what you mean by text data mining. yeah, you can grab keywords, and there are some simple clues about proximity of certain simple adjectives, and you can sort of associate certain vocabularies with income and spending habits, but the R^2 is pretty low. text mining is far, far away from "understanding the contents of every message." even google does a shoddy job; many of its text mining-based ads are silly and even insulting.
most of the marketing-juice comes from (surprise, surprise) the social network. facebook has trained people (maybe not you, but probably many of your "friends") to advertise themselves! if you're 1 hop away from 6 people who all explicitly "Like"d some expensive imported chocolate or coffee, that will probably tell me a whole lot more (marketing-wise) about you than any 100 of your messages, even if i had a human being reading every one of them, which text mining is nowhere near.
yes, but facebook rate limits messages also! with a coding scheme as sparse as that, you'd be lucky to send a kilobyte per hour.
or maybe they don't want people distributing binaries or running a number station on their service, for liability reasons.
not that i disagree necessarily; i just don't think facebook has very sophisticated text mining (yet).
in most subjects i would agree with you, but i don't think i would want an imaginative doctor (at least not at the expense of a strong level of basic competence). some things damned well should be done by rote, based on centuries of hard-won experience.
some people do have to come up with the new stuff, but most doctors don't and shouldn't be trying.
hm; arrogant and paranoid. yeah, you probably really are a doctor.
exactly. now, if he were named ULTRA-KILL! we'd know this wouldn't be a problem.
ah, yes. thanks.
some of these problems would be solved if they just priced the kit $0.50 over the unit.
huh? how?
i just don't understand this. it's not even an electronics project; it's sticking two totally arcane black boxes together. there's no particular logic or thinking or knowledge involved beyond the chip layout. it's just a tedious process involving luck, dexterity, and a very specialized skill that doesn't generalize much.
why the hell would anyone want to do this? it's certainly not curiosity, since that's generally about not doing things which a fucking robot can do 1000x better.
if you wanted to start a cottage industry repairing the sure-to-be ubiquitous raspberry pis (lol), that's a reason i guess, but you can just wait until it happens and you'll have plenty of junked parts to practice on.
well, of course you're not going to get what you want. you're going to get what optimizes the expected portfolios of large interests, which might coincide.
this is exactly why i wish the Rs would outgrow this childish denialism. if we're changing the environment, then let's try to do it optimally. if we're not changing the environment, then why the hell aren't we?! the chances of us being in the "best of all possible worlds" is almost zero, so why aren't the Rs doing what they do best and funneling massive public monies into private industry to change the environment? as a bonus, the old-school energy companies they've been buddying-up with have the best infrastructure to get on this project and become climate-control companies.
you mean X is reducible to Y.
it's been a while, but i think that for PCR you need to know the flanking sequences. you don't need to know the "middle", but you need the primer to match the ends.
that said, it seems that even with just PCR you could do dimensionality reduction on the code sequence... i.e. if you try PCR with an arbitrary primer, and it amplifies significantly, then you have some information about the sequence. it's not like a crypto key, where it's all or nothing. i suspect that chip-based sequencing and some statistical algos could make quick work of this "code", but like i said it's been a while and the paper provided is... uh... mostly advertising.
and even #2 is pretty shaky, apparently.
Is this to be an empathy test? Capillary dilation of the so-called blush response? Fluctuation of the pupil, involuntary dilation of the iris?