I would give it all of 7 days before that paid for data became available for free. Someone, somewhere, would buy the data for $2 per month and load it up for others to have free of charge. It would be a daily torrent that you could pull, or a streamed RSS feed, a static layout site with a downloadable screen scraper, or any one of a dozen other ways I can think.
Absolutely. The thing that prevents this from happening is that the data are all localized for a given market. My channels 3, 4, 5, 18, 32, 55, whatever, are not the same as yours. Heck, they probably aren't the same as the ones two towns over for obscure reasons due to arcane infrastructure. Time zone differences. State differences. Markets with different numbers of cable providers. Broadcast. It's a huge, wide data set, of which each of us would want only one thin slice. Each user cares about the channel lineup and scheduling for the signals at their end of the cable and no other, so creating a torrent of any single slice doesn't necessarily serve many people.
Like many others, I'm a little surprised that they aren't moving to a subscription model. Clearly they know better what their available resources are, and what they are and are not capable of handling, but it seems like a missed opportunity to walk away from a situation where their servers are getting hammered, and start charging a small fee. Many, many of the MythTV users would happily pay a few dollars a month to have a steady stream of information. Sounds like it could be a million dollar annual income right there, and that's got to be very hard to walk away from for any company.
As for commercial abuse, if they know it's happening, they presumably are taking steps to quash it as well, without much luck. Probably like playing whack-a-mole.
Add to the list of benefits: if the computers are deemed stolen, the local District Attorney's office should have evidence after the audit to prosecute someone, even if that's the former head of the school board.
Um, even in 480 BC (well before the 2nd century BC), professional armies were whupping the heck out of amateur armies through superior training and tactics.
The HP Sojourn (a rebadged Mistubushi Pedion) explored this corner of the design space in the very late 1990s, just about 10 years ago. I've owned a few of them. Fantastically thin at 0.72 inches (just 0.02 inches thicker than the claimed thickness of the new Intel device), with a then state-of-the-art 2 or 6 GB 9 mm disk drive and Pentium II 233 MMX processor with 64 MB of main memory. Very nice display, too. 2 PCMCIA slots, one USB 1.1 port, but no network interface. Excellent support under Linux including sleep and hibernation modes. To achieve such a slim form factor, the keyboard was chicklet-style (not unlike the HP calculators) which really didn't appeal to the power user. It was marketed to executives with the even-then astronomical pricetag of USD 6k. I bought my first one used at USD 1.5k (they were really bad at holding their value); currently you see them on eBay for under USD 200.
These are really, really thin. With full-sized keyboards and 12.1 inch displays. Slimmer than many padfolios. I never understood why that part of the design space wasn't more fully populated, as it's such an obvious (to me) win to have a really light, really thin, computer with a full-sized display and keyboard. Perhaps we're coming back to it.
No, you're wrong. On the screen font you happen to be using, under the OS you happen to be using, on the browser you happen to be using, using the rendering engine you happen to be using, it might or might not support kerning. So, you might or might not see the effect.
To a great extent, the resolutions available for computer screens sufficient to even think about kerning have only recently become available to the mainstream. Kerning is a subtle, but important, effet that most screen fonts are designed to not require (because of the limited rendering resolution).
You can, however, easily tell the difference between a properly hinted font where the hints have been correctly used and one where things are just wrong by printing out some text on a decent printer (ie, nearly anything manufactured in the last 2 years). If you typeset, for example "VA" on a kerned font, and very very carefully compare where the "V" ends and the "A" begins, you'll see they overlap just a hair. As in 1/72nd of an inch. The difference between a properly typeset font and one that's lacking kerning is the difference between a beautifully drawn pen-and-ink illustration and something hacked together with Powerpoint.
From TFA: The salespeople assured her that the C304NR was "Linux ready."
If true, this is an implied warranty for fitness for a particular purpose (which is, in some states, a right that cannot be taken away from you even if the written warranty excludes it, and even if you sign a document saying you give up such rights), and the original place of purchase is liable for deceptive business practices if, indeed, HP's warranty states you cannot load another operating system on the machine.
Disk space is ridiculously inexpensive these days -- a good 500 GB hard drive will set you back under USD 140 when on sale. You didn't specify how large your collection is in terms of uncompressed space, but let's say 20,000 tracks represents about 1,300 albums on CD (15 tracks per CD). That's about 800 GB (600 MB per CD). Four 500 GB drives (two for data, two for backup... you *were* going to have an on-line backup, right?) and you're set.
As other replies on this thread have observed, lossless compression recovers about 30%. Is your time and worry converting to a lossless format worthwhile? Tagging, yes, but compression for your reference collection? To me, the answer isn't clear, and as disk prices continue to plummet, the answer will sometime soon become, "don't bother compressing."
I read TFA. It states that a forensic analyist went through at least one of eight computers that the defendant is suspected of using. The one in question is from her home, if I recall. If the report is accurate (probably close enough) and complete (who knows), neither her ISP nor Google were involved in determining what the person using that computer was searching for.
While a fresh-faced junior in college, I took a tour of Europe. One day, I left my camera in what I thought was a secure place. It was stolen, and along with it went all of my exposed film. Today, while film is essentially dead, the spectre of catastrophic loss looms even larger because it's even more convenient than before to store all of your information in one place, like your laptop.
Therefore, my advice is, no matter what devices you plan to bring with you, take pains to have multiple copies of your data (photos, songs, whatever) and keep them physically separated to reduce the likelihood that both would get stolen at the same time. I use a laptop drive in an external USB enclosure as backup storage when traveling. Professional photographers have a mantra-like phrase, "protect the take," that means to guard the captured data.
IAAA. I am an academician. My entire career has been in academia. I spend time in my real office and get paid real money, with which I pay very real bills. Your world is somehow more real than mine?
I, too, like the extra hours of daylight in the evening, but this attitude that academia isn't relevant needs to go. Or, maybe, you'd rather live without antibiotics and nearly every other medical advance, telecommunications, computers, and most of the trappings of modern society? You use Linux, right? Where did that come from? You use email? The Web? Google? Academia isn't some parasite on society: it's where the seeds of the future grow.
Who do you thinks going to get the credit when they finally invent fission? The person who thought it up or those that actually made it?
I'm not sure you've written what you meant to here.
First, fission, that is, the breaking of atomic nuclei into separate clusters of particles with a concomitant release of energy, has been around on Earth for billions of years, as it's a natural phenomenon (see, eg, http://www.ocrwm.doe.gov/factsheets/doeymp0010.sht ml). Doing it in a controlled way to extract energy or explosive power, well the idea is something like 100 years old, and it was brought to fruition about 60 years ago. We remember the names of both sets of people (eg, Einstein, Bohr, Oppenhiemer, Fermi, Rutherford, Teller, etc.).
Second, you probably mean fusion. That, too, is really old, as it's a natural phenomenon. Powers the Sun, you see. Now, perhaps you meant a controlled fusion reaction in order to extract energy, or explosive power. The latter has already been done, and we remember both sets of names (see the list above). Perhaps, you really meant controlled fusion reactions to extract electrical power. That's still being worked on, but saying that it will be, "finally invented," is, well, an odd turn of phrase. It *has* been invented. The ideas (see that same list above) are all pretty old. The devil, it turns out, is in the details, and the details are really, really hard. So hard and so expensive that the current best hope, ITER, requires cooperation from seven countries (if you lump all of the EU into one country). We don't know who will be running the show when ITER, or some successor, finally creates a sustainable energy-positive burn, but I hope we remember their name as well as we remember the names above.
Clean burning fuel has nothing to do with C02, but everything to do with nitrogen and sulfer compounds, often call NOx and SOx ("x" because the number of oxygen atoms varies depending on the species). Those two classes of compounds are responsible for smog, acid rain, and, in part, the ozone layer depletion. Given the choice between burning, say, coal, which produces an excess of NOx and SOx, and methane which produce only traces of same when properly combusted, I'll take the methane, thank-you-very-much.
While this is a wonderful recognition of some fantastic work, the Slashdot editors should bear in mind that science and mathematics are not the same thing. To call solving the Poincarre Conjecture a breakthrough in Science (breakthrough of the year, no less!) is disrespectful to both scientists and mathematicians.
There have been some breakthroughs in Mathematics that were simultaneously notable in Science (solving the 4 Colors Problem, for example, the first time a computer was used to experimentally and exhaustively validate the results of a theorem), but these are rare. To the limits of my mathematical knowledge, this was not one of them, despite it being remarkable work.
[My candidate for the best microcomputer instruction set from the programmer's POV -- hands down, the MC6809]
Amen, brother! While I haven't been coding for quite as long as you (for me, it was 1976 when I started), I've used a hefty number of instruction sets and designed a handful myself. The 6809 was always my favorite. I still have a well-worn copy of the 6800 instruction set manual in my library; so clear, so beautiful. This was back when instruction set design was based purely on merit (what is the best way to compute?) rather than market forces (what is the best way to run MS applications?).
You must be working in a fairly non-commerical, non-medical arena...
IAAMR (I am a medical researcher), and let me tell you, the funding picture isn't pretty here, either. At a recent meeting of 25,000 of the world's neuroscientists in Atlanta, the NIH (National Institutes of Health) tried to present the current funding situation in a positive light, but no matter how hard they tried, "bleak" is the word that best describes it. I was in the audience along with a few thousand other researchers.
It's not that great proposals are getting funded and good ones no longer are -- it's that many if not most of the great ones are now not getting funded. The current success rate for new applications to the NIH is approximately 10%, and that's if you use the best interpretation of the NIH's own numbers. The less favorable interpretation has the funding rate at 7%. That is, seven percent of the submitted projects get monetary awards of some kind. My project, to make a Six Million Dollar Man / Jeordi LaForge style artificial vision device, has been among the unlucky 93% that did not get an award; you would think restoring sight to the blind would be fundable, huh?
There is, to first approximation, no research money available from the US Federal Government for new projects. Not even for sure-fire medical devices that have exceptionally good PR potential. I agree with the parent poster (who disagrees with the grandparent) in observing that this fellow's not securing funding from the US government is relatively meaningless because of the dire situation for non-weapons-research money available from the Feds.
I have been making the following argument for years (check the Slashdot archives for evidence of that assertion):
1. Diamond, it turns out, is an excellent semiconductor substrate. High band-gap, excellent thermal tolerance. It has the same crystal structure as silicon.
2. The semiconductor industry is able to spend billions of dollars in R&D. They have already done so to perfecting silicon crystals that are far, far more pure and far, far larger than any other single crystals. We're talking seven 9s purity single crystal ingots that are too big for one person to lift. I have a silicon boule head on my desk that is much larger than the Hope diamond, and much purer, that I bought for USD 25.
3. Once diamond has been developed as a semiconductor substrate, the jewelery market for expensive diamonds will no longer exist, except for the niche market of natural diamonds, because of the ready availability of diamonds that are far superior to what can be mined. (And if inclusions is what you want, colors, etc., that can be added back in. You could use beam epitaxy to draw designs, in color even.)
Within a few feet of the reader is probably a few carats' worth of what started out as ultra-pure single crystal silicon. One can currently buy blank silicon wafers on the surplus market for a few dollars. If that were diamond instead, diamond jewlery wouldn't have much value at all. De Beers might be powerful, but I'd put my money on Intel, AMD, and their bretheren.
if you work at Microsoft you'll have a life outside of work. If you work at Google, then work will be your life.
I visited the Google campus two weekends ago. On a Saturday. I counted only three working employees (in the Pirate group) other than the contractors who were setting up something in the main auditorium: the whole place was cavernously empty. The corporate culture is that life outside of Google is first, working at Google second. When it's time for work, everyone's there. When it's time to go home, people enjoy the rest of their life. And this makes for some very loyal employees.
I don't know about the Microsoft corporate culture, but the one at Google is definitely not what your professor described in the least.
You also forgot to add that exact film processing conditions (concentrations, age of the reagents, temperatures, humidity, etc.) can affect the image tremendously in often unreproducible ways. Digital imaging reduces many of these uncontrolled variables.
Electronic voting machines smell like a scam to me.
Mechanical voting machines work just fine and are comparatively simple, easy to verify, and last a long, long time. The major complaint against them is that they require maintenance, and that the parts are no longer available. For the millions of dollars of development and equipment costs to program and manufacture electronic voting machines that will surely last only a few years before being declared obsolete, a new set of dies could be designed and struck to make any spare parts necessary for another 50 years of mechanical voting using the same machines we have now (that get used, what, once per year at most for a few hours?). If we were in fact concerned with getting accurate, reliable counts, it seems to me we should be more concerned with proper maintenance and custodial care of our mechanical voting machines.
Higher acuity vision (like the proverbial baseball players with 20/10 or 20/6) have much better resolution in their foveas, but I don't know what the exact numbers would be. Also note that it is definitely possible to have your optometrist correct your eyes to much better than 20/20 -- for most people -- but that you probably would find it uncomfortable.
There are so many different variables involved in determining acuity that it's very very hard to nail one specific value down.
Acuity depends on the task, on the luminance contrast level, on the chrominance contast level, on the illumination level, whether you're using one eye or two, how much time you're allowed to view the test, how well your lenses focus on your retinae, and, probably, how much coffee you've had, how much you slept the night before, and what the phase of the moon is.
OK, kidding about the last one.
If you measure acuity with gratings, you get a very different number than if you measure it with split lines (vernier acuity), for example. I've heard people talking about different acuities for vertical vs. horizontal vs. non-cardinal gratings. It's pretty sloppy, thus I've no doubt you'd be able to find many different reports with better or worse findings than 3 minutes of arc. But, for 20/20 vision, the equivalent resolution would be about 3 minutes of arc.
IAAVN (I Am A Visual Neuroscientist). The answer to your question is, "no." The article pointed to claiming 15 million pixels specifically states the pixels are variable resolution. The photosensors in the central part of human (and primate) vision are packed at a much, much higher resolution than those at the periphery. The standard resolution in central vision for people with 20/20 vision is about 3 minutes of arc; at 3 degrees away from the fovea, this drops to 1/2 that figure; and at only 20 degrees eccentric (about two fist widths held at arm length), it's at 1/10. (If you've never heard that vision is variable resolution, try this trick: open a book or newspaper and stare at a single word in the middle of a paragraph; then, without moving your eyes, see how far to the left, right, up and down, you can read. You will find that the limits are astonishingly narrow. Evenly sampled high resolution vision is a powerful illusion based on the extreme resolution we have in the central part of vision, the ability to move our eyes, and some incredible circuitry in our brains.)
More importantly, saying you have N by M pixels alone doesn't give visual resolution, it gives object resolution: it is not possible to resolve individual pixels in an 8x10 photo printed at VGA resolution held 10 meters away, despite the relatively low resolution of the image. It is necessary to know not only the resolution of the image but the viewing distance as well to be able to say if the combination approaches the limits of human vision.
MIT, Stanford and UC Berkley have been doing this for years. Years. Those are the three I personally know about, and where those heavy-hitters go, others are sure to follow, so it's likely there are many more. (A little-known, short-lived institution called Ars Digita University was the first to do it; MIT was the first major player.) So why not go and study their systems? You and your employer are not treading virgin territory, so, before designing your system, it would be prudent to understand the paths other smart people and institutions have already taken when facing the same challenges.
Slashdot Mirror
I would give it all of 7 days before that paid for data became available for free. Someone, somewhere, would buy the data for $2 per month and load it up for others to have free of charge. It would be a daily torrent that you could pull, or a streamed RSS feed, a static layout site with a downloadable screen scraper, or any one of a dozen other ways I can think.
Absolutely. The thing that prevents this from happening is that the data are all localized for a given market. My channels 3, 4, 5, 18, 32, 55, whatever, are not the same as yours. Heck, they probably aren't the same as the ones two towns over for obscure reasons due to arcane infrastructure. Time zone differences. State differences. Markets with different numbers of cable providers. Broadcast. It's a huge, wide data set, of which each of us would want only one thin slice. Each user cares about the channel lineup and scheduling for the signals at their end of the cable and no other, so creating a torrent of any single slice doesn't necessarily serve many people.
Like many others, I'm a little surprised that they aren't moving to a subscription model. Clearly they know better what their available resources are, and what they are and are not capable of handling, but it seems like a missed opportunity to walk away from a situation where their servers are getting hammered, and start charging a small fee. Many, many of the MythTV users would happily pay a few dollars a month to have a steady stream of information. Sounds like it could be a million dollar annual income right there, and that's got to be very hard to walk away from for any company.
As for commercial abuse, if they know it's happening, they presumably are taking steps to quash it as well, without much luck. Probably like playing whack-a-mole.
Let's all hope Google comes to the rescue.
That's friggen BRILLIANT!
Add to the list of benefits: if the computers are deemed stolen, the local District Attorney's office should have evidence after the audit to prosecute someone, even if that's the former head of the school board.
Um, even in 480 BC (well before the 2nd century BC), professional armies were whupping the heck out of amateur armies through superior training and tactics.
The HP Sojourn (a rebadged Mistubushi Pedion) explored this corner of the design space in the very late 1990s, just about 10 years ago. I've owned a few of them. Fantastically thin at 0.72 inches (just 0.02 inches thicker than the claimed thickness of the new Intel device), with a then state-of-the-art 2 or 6 GB 9 mm disk drive and Pentium II 233 MMX processor with 64 MB of main memory. Very nice display, too. 2 PCMCIA slots, one USB 1.1 port, but no network interface. Excellent support under Linux including sleep and hibernation modes. To achieve such a slim form factor, the keyboard was chicklet-style (not unlike the HP calculators) which really didn't appeal to the power user. It was marketed to executives with the even-then astronomical pricetag of USD 6k. I bought my first one used at USD 1.5k (they were really bad at holding their value); currently you see them on eBay for under USD 200.
These are really, really thin. With full-sized keyboards and 12.1 inch displays. Slimmer than many padfolios. I never understood why that part of the design space wasn't more fully populated, as it's such an obvious (to me) win to have a really light, really thin, computer with a full-sized display and keyboard. Perhaps we're coming back to it.
"In short, at least give the penguin a fair viewing. If you still don't like it, that's ok: that's why I'm boss. I simply know better than you do."
With the more complete context, my reading is that he was being playfully facetious!
No, you're wrong. On the screen font you happen to be using, under the OS you happen to be using, on the browser you happen to be using, using the rendering engine you happen to be using, it might or might not support kerning. So, you might or might not see the effect.
To a great extent, the resolutions available for computer screens sufficient to even think about kerning have only recently become available to the mainstream. Kerning is a subtle, but important, effet that most screen fonts are designed to not require (because of the limited rendering resolution).
You can, however, easily tell the difference between a properly hinted font where the hints have been correctly used and one where things are just wrong by printing out some text on a decent printer (ie, nearly anything manufactured in the last 2 years). If you typeset, for example "VA" on a kerned font, and very very carefully compare where the "V" ends and the "A" begins, you'll see they overlap just a hair. As in 1/72nd of an inch. The difference between a properly typeset font and one that's lacking kerning is the difference between a beautifully drawn pen-and-ink illustration and something hacked together with Powerpoint.
From TFA: The salespeople assured her that the C304NR was "Linux ready."
If true, this is an implied warranty for fitness for a particular purpose (which is, in some states, a right that cannot be taken away from you even if the written warranty excludes it, and even if you sign a document saying you give up such rights), and the original place of purchase is liable for deceptive business practices if, indeed, HP's warranty states you cannot load another operating system on the machine.
Disk space is ridiculously inexpensive these days -- a good 500 GB hard drive will set you back under USD 140 when on sale. You didn't specify how large your collection is in terms of uncompressed space, but let's say 20,000 tracks represents about 1,300 albums on CD (15 tracks per CD). That's about 800 GB (600 MB per CD). Four 500 GB drives (two for data, two for backup ... you *were* going to have an on-line backup, right?) and you're set.
As other replies on this thread have observed, lossless compression recovers about 30%. Is your time and worry converting to a lossless format worthwhile? Tagging, yes, but compression for your reference collection? To me, the answer isn't clear, and as disk prices continue to plummet, the answer will sometime soon become, "don't bother compressing."
I read TFA. It states that a forensic analyist went through at least one of eight computers that the defendant is suspected of using. The one in question is from her home, if I recall. If the report is accurate (probably close enough) and complete (who knows), neither her ISP nor Google were involved in determining what the person using that computer was searching for.
While a fresh-faced junior in college, I took a tour of Europe. One day, I left my camera in what I thought was a secure place. It was stolen, and along with it went all of my exposed film. Today, while film is essentially dead, the spectre of catastrophic loss looms even larger because it's even more convenient than before to store all of your information in one place, like your laptop.
Therefore, my advice is, no matter what devices you plan to bring with you, take pains to have multiple copies of your data (photos, songs, whatever) and keep them physically separated to reduce the likelihood that both would get stolen at the same time. I use a laptop drive in an external USB enclosure as backup storage when traveling. Professional photographers have a mantra-like phrase, "protect the take," that means to guard the captured data.
IAAA. I am an academician. My entire career has been in academia. I spend time in my real office and get paid real money, with which I pay very real bills. Your world is somehow more real than mine?
I, too, like the extra hours of daylight in the evening, but this attitude that academia isn't relevant needs to go. Or, maybe, you'd rather live without antibiotics and nearly every other medical advance, telecommunications, computers, and most of the trappings of modern society? You use Linux, right? Where did that come from? You use email? The Web? Google? Academia isn't some parasite on society: it's where the seeds of the future grow.
Who do you thinks going to get the credit when they finally invent fission? The person who thought it up or those that actually made it?
t ml). Doing it in a controlled way to extract energy or explosive power, well the idea is something like 100 years old, and it was brought to fruition about 60 years ago. We remember the names of both sets of people (eg, Einstein, Bohr, Oppenhiemer, Fermi, Rutherford, Teller, etc.).
I'm not sure you've written what you meant to here.
First, fission, that is, the breaking of atomic nuclei into separate clusters of particles with a concomitant release of energy, has been around on Earth for billions of years, as it's a natural phenomenon (see, eg, http://www.ocrwm.doe.gov/factsheets/doeymp0010.sh
Second, you probably mean fusion. That, too, is really old, as it's a natural phenomenon. Powers the Sun, you see. Now, perhaps you meant a controlled fusion reaction in order to extract energy, or explosive power. The latter has already been done, and we remember both sets of names (see the list above). Perhaps, you really meant controlled fusion reactions to extract electrical power. That's still being worked on, but saying that it will be, "finally invented," is, well, an odd turn of phrase. It *has* been invented. The ideas (see that same list above) are all pretty old. The devil, it turns out, is in the details, and the details are really, really hard. So hard and so expensive that the current best hope, ITER, requires cooperation from seven countries (if you lump all of the EU into one country). We don't know who will be running the show when ITER, or some successor, finally creates a sustainable energy-positive burn, but I hope we remember their name as well as we remember the names above.
Clean burning fuel has nothing to do with C02, but everything to do with nitrogen and sulfer compounds, often call NOx and SOx ("x" because the number of oxygen atoms varies depending on the species). Those two classes of compounds are responsible for smog, acid rain, and, in part, the ozone layer depletion. Given the choice between burning, say, coal, which produces an excess of NOx and SOx, and methane which produce only traces of same when properly combusted, I'll take the methane, thank-you-very-much.
While this is a wonderful recognition of some fantastic work, the Slashdot editors should bear in mind that science and mathematics are not the same thing. To call solving the Poincarre Conjecture a breakthrough in Science (breakthrough of the year, no less!) is disrespectful to both scientists and mathematicians.
There have been some breakthroughs in Mathematics that were simultaneously notable in Science (solving the 4 Colors Problem, for example, the first time a computer was used to experimentally and exhaustively validate the results of a theorem), but these are rare. To the limits of my mathematical knowledge, this was not one of them, despite it being remarkable work.
[My candidate for the best microcomputer instruction set from the programmer's POV -- hands down, the MC6809]
Amen, brother! While I haven't been coding for quite as long as you (for me, it was 1976 when I started), I've used a hefty number of instruction sets and designed a handful myself. The 6809 was always my favorite. I still have a well-worn copy of the 6800 instruction set manual in my library; so clear, so beautiful. This was back when instruction set design was based purely on merit (what is the best way to compute?) rather than market forces (what is the best way to run MS applications?).
You must be working in a fairly non-commerical, non-medical arena...
IAAMR (I am a medical researcher), and let me tell you, the funding picture isn't pretty here, either. At a recent meeting of 25,000 of the world's neuroscientists in Atlanta, the NIH (National Institutes of Health) tried to present the current funding situation in a positive light, but no matter how hard they tried, "bleak" is the word that best describes it. I was in the audience along with a few thousand other researchers.
It's not that great proposals are getting funded and good ones no longer are -- it's that many if not most of the great ones are now not getting funded. The current success rate for new applications to the NIH is approximately 10%, and that's if you use the best interpretation of the NIH's own numbers. The less favorable interpretation has the funding rate at 7%. That is, seven percent of the submitted projects get monetary awards of some kind. My project, to make a Six Million Dollar Man / Jeordi LaForge style artificial vision device, has been among the unlucky 93% that did not get an award; you would think restoring sight to the blind would be fundable, huh?
There is, to first approximation, no research money available from the US Federal Government for new projects. Not even for sure-fire medical devices that have exceptionally good PR potential. I agree with the parent poster (who disagrees with the grandparent) in observing that this fellow's not securing funding from the US government is relatively meaningless because of the dire situation for non-weapons-research money available from the Feds.
I have been making the following argument for years (check the Slashdot archives for evidence of that assertion):
1. Diamond, it turns out, is an excellent semiconductor substrate. High band-gap, excellent thermal tolerance. It has the same crystal structure as silicon.
2. The semiconductor industry is able to spend billions of dollars in R&D. They have already done so to perfecting silicon crystals that are far, far more pure and far, far larger than any other single crystals. We're talking seven 9s purity single crystal ingots that are too big for one person to lift. I have a silicon boule head on my desk that is much larger than the Hope diamond, and much purer, that I bought for USD 25.
3. Once diamond has been developed as a semiconductor substrate, the jewelery market for expensive diamonds will no longer exist, except for the niche market of natural diamonds, because of the ready availability of diamonds that are far superior to what can be mined. (And if inclusions is what you want, colors, etc., that can be added back in. You could use beam epitaxy to draw designs, in color even.)
Within a few feet of the reader is probably a few carats' worth of what started out as ultra-pure single crystal silicon. One can currently buy blank silicon wafers on the surplus market for a few dollars. If that were diamond instead, diamond jewlery wouldn't have much value at all. De Beers might be powerful, but I'd put my money on Intel, AMD, and their bretheren.
if you work at Microsoft you'll have a life outside of work. If you work at Google, then work will be your life.
I visited the Google campus two weekends ago. On a Saturday. I counted only three working employees (in the Pirate group) other than the contractors who were setting up something in the main auditorium: the whole place was cavernously empty. The corporate culture is that life outside of Google is first, working at Google second. When it's time for work, everyone's there. When it's time to go home, people enjoy the rest of their life. And this makes for some very loyal employees.
I don't know about the Microsoft corporate culture, but the one at Google is definitely not what your professor described in the least.
You also forgot to add that exact film processing conditions (concentrations, age of the reagents, temperatures, humidity, etc.) can affect the image tremendously in often unreproducible ways. Digital imaging reduces many of these uncontrolled variables.
Electronic voting machines smell like a scam to me.
Mechanical voting machines work just fine and are comparatively simple, easy to verify, and last a long, long time. The major complaint against them is that they require maintenance, and that the parts are no longer available. For the millions of dollars of development and equipment costs to program and manufacture electronic voting machines that will surely last only a few years before being declared obsolete, a new set of dies could be designed and struck to make any spare parts necessary for another 50 years of mechanical voting using the same machines we have now (that get used, what, once per year at most for a few hours?). If we were in fact concerned with getting accurate, reliable counts, it seems to me we should be more concerned with proper maintenance and custodial care of our mechanical voting machines.
Higher acuity vision (like the proverbial baseball players with 20/10 or 20/6) have much better resolution in their foveas, but I don't know what the exact numbers would be. Also note that it is definitely possible to have your optometrist correct your eyes to much better than 20/20 -- for most people -- but that you probably would find it uncomfortable.
There are so many different variables involved in determining acuity that it's very very hard to nail one specific value down.
Acuity depends on the task, on the luminance contrast level, on the chrominance contast level, on the illumination level, whether you're using one eye or two, how much time you're allowed to view the test, how well your lenses focus on your retinae, and, probably, how much coffee you've had, how much you slept the night before, and what the phase of the moon is.
OK, kidding about the last one.
If you measure acuity with gratings, you get a very different number than if you measure it with split lines (vernier acuity), for example. I've heard people talking about different acuities for vertical vs. horizontal vs. non-cardinal gratings. It's pretty sloppy, thus I've no doubt you'd be able to find many different reports with better or worse findings than 3 minutes of arc. But, for 20/20 vision, the equivalent resolution would be about 3 minutes of arc.
Could this be considered "full human" resolution?
IAAVN (I Am A Visual Neuroscientist). The answer to your question is, "no." The article pointed to claiming 15 million pixels specifically states the pixels are variable resolution. The photosensors in the central part of human (and primate) vision are packed at a much, much higher resolution than those at the periphery. The standard resolution in central vision for people with 20/20 vision is about 3 minutes of arc; at 3 degrees away from the fovea, this drops to 1/2 that figure; and at only 20 degrees eccentric (about two fist widths held at arm length), it's at 1/10. (If you've never heard that vision is variable resolution, try this trick: open a book or newspaper and stare at a single word in the middle of a paragraph; then, without moving your eyes, see how far to the left, right, up and down, you can read. You will find that the limits are astonishingly narrow. Evenly sampled high resolution vision is a powerful illusion based on the extreme resolution we have in the central part of vision, the ability to move our eyes, and some incredible circuitry in our brains.)
More importantly, saying you have N by M pixels alone doesn't give visual resolution, it gives object resolution: it is not possible to resolve individual pixels in an 8x10 photo printed at VGA resolution held 10 meters away, despite the relatively low resolution of the image. It is necessary to know not only the resolution of the image but the viewing distance as well to be able to say if the combination approaches the limits of human vision.
MIT, Stanford and UC Berkley have been doing this for years. Years. Those are the three I personally know about, and where those heavy-hitters go, others are sure to follow, so it's likely there are many more. (A little-known, short-lived institution called Ars Digita University was the first to do it; MIT was the first major player.) So why not go and study their systems? You and your employer are not treading virgin territory, so, before designing your system, it would be prudent to understand the paths other smart people and institutions have already taken when facing the same challenges.