Since the price point on these are so low, what's the feasibility of doing mass grid computing on these machines?
First of all "price point" is marketing speak. I think you mean "price".
For educational purposes it'll work. I've seen screenshots of early alpha boards booting plain jane Debian. I set up a grid of surplus P75s (basically free) about a decade ago with Debian and learned a whole heck of a lot about parallel processing and grid administration and how it all works in general. I intend to buy about 4 to 10 of these boards upon release to basically nostalgically relive my misbegotten youth. Also I like the idea of a Beowulf cluster that fits in a kids lunchbox.
For actual production, I think you'll find that much like my P75 experiment a decade ago, a surplus desktop in the $100 class will probably wildly outproduce a large cluster of these devices.
GPIO, SPI and I2C will all be there. No on-board analog IOs.
Fess up, AC, I'm pretty sure one of you ACs is Eben. Just register for a/. account like every other technowizard has done over the past fifteen years. That way I can "/. friend you" and your posts won't go into unviewed "AC hell".
So how do you flash them again? How do you "get linux on them"?
From the FAQ:
We will be selling SD cards with the distros preloaded.
Thankfully Debian is not BSD, and has plenty of GPL so you can't "sell" Debian without providing a free download link and free sources etc... On the other hand, to voluntarily support the project, I'd be willing to purchase a SD card at a reasonable donation-ish price... Or, if they would stick my name on a donors page, or sell me an honorary gold plated SD card on a thank you plaque, I'd willingly drop $100 or so.
I know a guy with an alpha preview test board. I've seen the screenshots. Pretty cool. Not vapor at all.
I haven't installed Debian off a cd / dvd / other removable media in many years, I netboot to install. The model B has an ethernet port, so assuming it PXE boots, or you can stick a PXE booter on a SD card, that would make a fine second option.
To answer the original weird question that seems to assume the device is a USB peripheral instead of being a USB host:
Is there power over USB? No. Raspberry Pi is a USB host, not a USB device, and you can’t draw power from the uplink port of a hub.
This would imply to connect to a desktop or whatever you need one of those weird active USB to USB "psuedonetwork" cables. Personally I'd just stick a bluetooth adapter in the USB hole on each machine and be done with it, but to each their own...
Any plans or exposing some pads or even pins for Digital IO, SPI, and or I2C? Maybe even a few A2Ds? Such a device would be very handy for embedded systems. Things like weather stations, robotics, data logging and so on. Could be very handy in any science class room.
IO ports are where its at... This would make an ideal embedded dev platform.
Write my control loop simulation in GNU Octave on the desktop, run the same control loop file for real on the little embedded box.
You physically copy the certificate request to the signing computer by floppy disk or flash drive or similar method and copy the certificate back the same way.
I think he's getting at, how do you upgrade Debian's openssl package if its not on the net... Well, there are ways around that, again involving flash drives and such...
I think this is not just casual LOL type watching, but scientifically carefully studying the reaction to a semi-credible threat, to figure out how to work around their reaction in a future (real?) event.
How has the collapse of diginotaurus or whatever affected other CAs response?
I feel like all the training I did to be able to code games** in a PC is going to be obsolete before I know it.
Training rapidly becomes useless. Education never becomes useless.
Memorize how to use a linked list library in Pascal = rapidly useless
Learn what a linked list is, why and when you'd use it = useful forever
Also much like human languages (supposedly) the first three languages are pretty tough, but once you learn a bit of ten or so, its pretty simple since all the concepts are the same. The hard part is knowing how to index thru an array without a picket fence mistake and figuring out how to troubleshoot it. The easy part is remembering or googling the syntax.
They're the two biggest because American consumers love the result of how those corporations are run, even if they claim to hate the behavior.
They're not the biggest because their employees love them. If that were the case I'd expect family owned companies to the worlds largest, not faceless multinationals.
The only successful "communists" I know of are communal Mennonites and Amish.
US Armed forces?
Not successful as a closed system, in fact they require a trillion or so per year to keep on bombin' on the input side, and on the output side the resulting death and destruction is probably more than a $1T but aside from that, its pretty much from each according to their ability and to each according to their need.
This was widely told to me when I was in the US Army in the early 90s, always without mentioning the cost, as one of those ironic things that here we were cold waring the commies or having beat the commies, yet we run the army internally on purely commie principles. Basically a little commie country but without any territory outside our bases.
He had a solution. Require that shares in companies be own solely by individuals that are working there. In all other ways the economy would be the same. But, because the workers would own their own company, there would be a vested interested in balancing the needs of the workers with the demands of the market. And because there would be multiple companies doing the same thing you'd still get market forces.
Ah but very weak market forces because price discovery is hard when there's only a tiny number of people in the market.
I worked at a place like that. "beneficial units of interest". Tax laws make dividend payments unappetizing, so the place was always swimming in cash, trying to find something to invest in. Leading to lots of really stupid investments. The lower the bank interest rate, the stupider and riskier the investment. If they were momentarily short of cash or about to buy something big, you'd have your bosses boss boss lecturing you in his office about how you really need to take your yearly bonus in stock not cash and up your percentage of salary toward stock, and he can't legally tell you what to do, but plenty of unspoken implications... Price discovery was a bean counter was pressured to give the highest possible unit price that would not result in the auditors swooping in to destroy him, not exactly as efficient as a free market. The company encouraged buying on margin, so you'd have a thousand people borrowing a thousand bucks at crazy interest rates from a hundred banks instead of the company as a whole taking a simple $1M loan at a low corporate interest rate. Makes money for the banks, not so much for the rest of us. Old timers who owned more than say, a quarter million of stock were semi-untouchable, as the company would have to raise a quarter mil cash to buy them out if they quit, retired, or were fired, and/or hire enough people or otherwise encourage the rest of us to buy a 1/4 mill of stock, which made employee / employer relations a bit strained at times, as you couldn't treat peons like dirt as standard at other corporations. Some old timer quits or dies and you can't upgrade your server due to a momentary cash crunch. The execs decided to pay themselves obscene amounts of company stock; they already had more shares than us peons, so they were untouchable. Any random public owner of local electric company stock can complain about the electric company's management's performance because they don't work there... Any owner of company stock has to be a yes man wrt to management performance or get downsized. So we had the best rated, highest paid, yet most incompetent senior management I've ever heard of at any company. If you quit, you make a lot of money. If you're downsized due to a downturn, the price has dropped, you don't make so much money, so its extremely risky. In the good times you don't leave because you'd lose possible gains on the stock. In the bad times, you'd be locking in losses, right when you don't need them. On the other hand, I always owned public stock in competitors, because if they go down, my job goes up, and if my job goes away, my competitor stock goes up, more or less.
Overall, it was a weird, yet none the less profitable, experience. It's possible, but I'm not sure how well it would scale, and given typical levels of corruption I think it would end up collapsing into a reimplementation of our current system. Probably interlocking boards of directors paying themselves obscene bonuses of multiple companies internal stocks, while the rank -n- file workers own about 1% of outstanding shares, in other words pretty much what we have now, but minus the middlemen at NYSE and NASDAQ.
Hardly a new phenomena. Compare the apparent financial condition of "leave it to beaver" or "brady bunch" or "married with children" with what an actual family in those employment scenarios would realistically have...
I remember "The Cosby Show" being savaged by the press because the shows lifestyle was supposedly too extravagant for the father's income as a successful doctor, although they never hassled the detectives on "Miami Vice" for the same mistake, probably was a racial thing.
Barring direct financial contribution to political candidates and forcing them to run on equal funds would help. Barring the movement between high public office and private business, especially government contractors, would help as well
A lotto system is acceptable for jury trials. It should work for legislative and executive branch as well. Provide a lifetime pension of 8 times the median income and mandatory felony prison time if their tax return ever shows any other source of income. Income multiple is high to prevent familial corruption; "vote this way and I'll give your kid a really nice job", "naw, I got enough pension income that he doesn't have to get a job".
Yes, but most of the time you had to pass credit checks to get one.
Not much has changed, in that regard. Also they cost "about" the same, weirdly enough. A couple thousand bucks for about two years of service. Some things never change.
If by "cell phone" you mean more generically any cordless phone where you don't own/maintain the base station, IMTS was pretty popular in the late '60s, if you were wealthy enough.
Errors can come from nearly anything; slightly imperfect knowledge of the characteristics of your device
Non-linear effect anywhere in the RX or TX chain, or intermod from surrounding objects is a big problem.
I've done quite a bit of RF design work, microwave ham radio stuff, etc. The big problem is historically low noise stuff which makes a great receiver tends to blow up when subjected to power, and high power gear tends to have horrific weak signal noise characteristics.
A great low noise fractional dB noise figure preamp is off the shelf and cheap, and it'll be vaporized by say 20 dBmW. A great 30 dBmW MMIC 1 GHz amp is off the shelf and cheap, I have used the watkins johnson devices (yes I know they have a new marketing name which I've temporarily forgotten), and its weak signal noise performance... is not good.
In contrast, this usage of full-duplex means that both directions are transmitting simultaneously on the same channel, without segmenting or multiplexing it.
I don't actually know how they solved the problem, though, and the article is light on details.
If I had to do it, I'd do traditional 70s era spread spectrum code division multiple access CDMA. Imagine a psuedorandom voltage generator feeding the RX VCO attached to the RX mixer. Then imagine a different psuedorandom voltage gen, or at least the same generating polynomial at a different offset, feeding the TX VCO attached to the TX mixer. Two completely separate RF paths, maybe up to the antenna. Synchronizing two separate psuedorandom voltage gens is merely twice as fun as just one, kinda, I guess.
The other way was to use an old fashioned yet highly effective RF circulator. They are large, and heavy, and frankly kinda hard to make. Think like a hockey puck of ferrite with a big ole magnet. RF only flows clockwise. This is old, old stuff. Larger and heavier than a "brick" cellphone from the 90s, although they worked perfectly fine at the base station.
There's another way to do it using PLLs and the two transmitters in quadrature, but that's getting bizarre (like, have I been drinking this morning already?) and synchronization is gonna be an absolute bear. The hard part isn't static stability, but dynamic as it switches in and out of sync, or multipath interferes with it.
In a world where bandwidth demands are increasing exponentially, a simple doubling of capacity ought to get us by for, oh, I don't, know, 6 months?
Demand might be increasing, but the supply is being choked by throttling and overage charges. Great, now I can reach my cap in 10 minutes instead of an hour. Whoo hoo.
Also there's a pretty hard limit to BW demands... What would I do with a tiny little hand held device with a two hour battery life that would use more than HDTV bandwidth? I reached this limit at home around 2003, and both my monthly bill and comparative service levels have gone from "exotic" to "average" in that interval. I could certainly afford to upgrade to 100 meg service, but I have no idea what I'd do with it that I'm not already doing...
1) FAX machine : Tell the phone person to please provision a POTS analog phone line to a jack right there, and tell me the external number. Fax machines are cheap and can be bought on the office expense account (the one used to buy paperclips). For bonus points tell the receptionist your new departmental fax number. Unbox fax, plug in, you're running. You know if it works or not because every far end tells your near end in some manner that is "OK" or not. Support is, if it breaks, buy another. It just works.
2) Scan and email : Fill out request form for IT dept for the hardware. They need to follow the capital expense forms and procedures to buy your $100 flatbed scanner, along with possible competitive bidding, assuming they even have the capital budget remaining for the year. Your bosses bosses boss may need to get permission from his boss to transfer $100 of his capital budget to IT, assuming he has the budget. Its quite trivial to spend thousands in labor on meetings and arguments about spending $100. It may or may not arrive in 3 months and may or may not meet your needs, but you're stuck with the hardware. Fill out a request form for IT to get the scanner software installed on your locked down PCs. Argue endlessly about who will support the system, and how much it will be supported. Eventually you get it working, and every time you send an email with a scan, you have to call or wait for an email response to prove their anti-virus didn't eat it. Its a nightmare.
At home I would never use a fax. But I understand why they're the path of least resistance at businesses.
...and would require integration with multiple OS's, require drivers (the quality of which we don't know), etc.
Making it transparent to the OS means you don't have any of those problems. It's a trade off, and for a first gen drive, probably the better way to go.
Don't forget massive vendor lockin, to be avoided like the plague that it is.
Hmm, only works under one OS, perhaps one kernel version of that OS, with one brand, perhaps one model of drive, perhaps they can find a way to lock in the motherboard too. Um, no thanks. Run away! Run away!
You have a 12TB file server? WTF are you putting on that thing, pirated content you'll never have enough time in your life to watch?
My wife has done semi-pro video editing (as in she's not being fairly compensated for her labor) and as her local tech support guy aka husband I can guarantee that its very difficult to get enough storage space without... semi-pro storage gear, like maybe a ebay surplus NAS or something. Its a huge PITA to use off the shelf solutions. Like if you try to do it all out of a tigerdirect catalog or buying on amazon. She ended up using multiple external HDs in cases, plugging the usb or fw into her mac, and storing raw HD interview footage on the external drives and only "good clips" on the main server for final editing.
For high def documentary type work, figure you'll end up with about 10 tb of raw video per hour of finished high def video. This is assuming storage space is not going to be your limiting factor. If your limiting factor is "filming" time on-site, maybe you can't take enough video to fill 10 TB. If your limiting factor is editing time to meet a deadline for duplication, then maybe you can't edit 10 TB worth of video so no sense taking it. The limiting factor is not always video storage space, although often it is.
Note that if your idea of high end video capture device is a dusty fingerprint smudged cellphone that records 320x240 at a couple frames/sec, then you probably don't need 10 tb. On the other hand, if your camera cost more than a mortgage payment, then you probably need more than 10 TB, if not online, at least available in external plug in cases.
I don't really know much about how she does it, I'm just the guy who hears "the server is full!" no matter how big the drives are. I think about 50 TB online would handle the multiple simultaneous project situation. Any idea how to put 50 TB in my basement, more or less contiguous, in a sensible scalable maintainable reliable manner? Also I only use Debian for infrastructure, if that matters.
is planning to take part in an unprecedented science project — a 20-square-mile model of a small U.S. city.
Note the military has quite a few of these, although smaller scale. Also full of bullet holes. Which might actually be a bonus if you're planning on a technology deployment in "urban" areas.
...for hobos and other mobile homeless. How will they be kept out of all those uninhabited buildings? They may look uninhabitable to people with a place to live, but to the homeless, they might look not too bad.
Don't need a floor for testing? Don't waste time and money building one.
The place could be infested just like a city park. But it should be less of a problem than an abandoned tract of mcmansions, and they seem to be doing fine.
Because burning down 35,000 real homes when your molten salt solar energy storage system fails, or having your water recycling system backflow bad water through the tap is a bigger problem when you fill them with families of 4?
Even "normal failure" needs to be covered up. If it takes 50 revisions to get your SDHW panels not to leak, the last thing you want is 50 families whining on facebook and twitter about how revisions 1 thru 49 of your new panel design leaked water all over their priceless scrapbooks.
Also payment negotiations with a zillion individuals would be a huge PITA.
That was my first thought. I'm still waiting for someone to build a city like Eureka with... well... slightly lower requirements for residence.;)
Huntsville Alabama? Last time I visited, in the early 90s, around 80% of the population was pure binary, either military uniform or doctorate degree. What do you get if you take a sleepy farm town in 1945 and drop 50 times its population of german V-2 rocket scientists, and friends, on top of it? Nobody could talk at the bar about what they did at work, but you knew whatever it was, it was cool.
Slashdot Mirror
Since the price point on these are so low, what's the feasibility of doing mass grid computing on these machines?
First of all "price point" is marketing speak. I think you mean "price".
For educational purposes it'll work. I've seen screenshots of early alpha boards booting plain jane Debian. I set up a grid of surplus P75s (basically free) about a decade ago with Debian and learned a whole heck of a lot about parallel processing and grid administration and how it all works in general. I intend to buy about 4 to 10 of these boards upon release to basically nostalgically relive my misbegotten youth. Also I like the idea of a Beowulf cluster that fits in a kids lunchbox.
For actual production, I think you'll find that much like my P75 experiment a decade ago, a surplus desktop in the $100 class will probably wildly outproduce a large cluster of these devices.
GPIO, SPI and I2C will all be there. No on-board analog IOs.
Fess up, AC, I'm pretty sure one of you ACs is Eben. Just register for a /. account like every other technowizard has done over the past fifteen years. That way I can "/. friend you" and your posts won't go into unviewed "AC hell".
So how do you flash them again? How do you "get linux on them"?
From the FAQ:
We will be selling SD cards with the distros preloaded.
Thankfully Debian is not BSD, and has plenty of GPL so you can't "sell" Debian without providing a free download link and free sources etc... On the other hand, to voluntarily support the project, I'd be willing to purchase a SD card at a reasonable donation-ish price... Or, if they would stick my name on a donors page, or sell me an honorary gold plated SD card on a thank you plaque, I'd willingly drop $100 or so.
I know a guy with an alpha preview test board. I've seen the screenshots. Pretty cool. Not vapor at all.
I haven't installed Debian off a cd / dvd / other removable media in many years, I netboot to install. The model B has an ethernet port, so assuming it PXE boots, or you can stick a PXE booter on a SD card, that would make a fine second option.
To answer the original weird question that seems to assume the device is a USB peripheral instead of being a USB host:
Is there power over USB? No. Raspberry Pi is a USB host, not a USB device, and you can’t draw power from the uplink port of a hub.
This would imply to connect to a desktop or whatever you need one of those weird active USB to USB "psuedonetwork" cables. Personally I'd just stick a bluetooth adapter in the USB hole on each machine and be done with it, but to each their own...
Any plans or exposing some pads or even pins for Digital IO, SPI, and or I2C? Maybe even a few A2Ds? Such a device would be very handy for embedded systems. Things like weather stations, robotics, data logging and so on. Could be very handy in any science class room.
IO ports are where its at... This would make an ideal embedded dev platform.
Write my control loop simulation in GNU Octave on the desktop, run the same control loop file for real on the little embedded box.
videos (which this won't support anyway).
Supposedly has native H.264 decoding or hw acceleration or whatever.
I have thought it over.
You physically copy the certificate request to the signing computer by floppy disk or flash drive or similar method and copy the certificate back the same way.
I think he's getting at, how do you upgrade Debian's openssl package if its not on the net... Well, there are ways around that, again involving flash drives and such...
Untrusted CAs aren't included in the web browser
Insert simpsons voice "ha ha". The whole point is that is just not so.
3. Watch as they scramble in panic
I think this is not just casual LOL type watching, but scientifically carefully studying the reaction to a semi-credible threat, to figure out how to work around their reaction in a future (real?) event.
How has the collapse of diginotaurus or whatever affected other CAs response?
I feel like all the training I did to be able to code games** in a PC is going to be obsolete before I know it.
Training rapidly becomes useless. Education never becomes useless.
Memorize how to use a linked list library in Pascal = rapidly useless
Learn what a linked list is, why and when you'd use it = useful forever
Also much like human languages (supposedly) the first three languages are pretty tough, but once you learn a bit of ten or so, its pretty simple since all the concepts are the same. The hard part is knowing how to index thru an array without a picket fence mistake and figuring out how to troubleshoot it. The easy part is remembering or googling the syntax.
They're the two biggest because American consumers love the result of how those corporations are run, even if they claim to hate the behavior.
They're not the biggest because their employees love them. If that were the case I'd expect family owned companies to the worlds largest, not faceless multinationals.
The only successful "communists" I know of are communal Mennonites and Amish.
US Armed forces?
Not successful as a closed system, in fact they require a trillion or so per year to keep on bombin' on the input side, and on the output side the resulting death and destruction is probably more than a $1T but aside from that, its pretty much from each according to their ability and to each according to their need.
This was widely told to me when I was in the US Army in the early 90s, always without mentioning the cost, as one of those ironic things that here we were cold waring the commies or having beat the commies, yet we run the army internally on purely commie principles. Basically a little commie country but without any territory outside our bases.
He had a solution. Require that shares in companies be own solely by individuals that are working there. In all other ways the economy would be the same. But, because the workers would own their own company, there would be a vested interested in balancing the needs of the workers with the demands of the market. And because there would be multiple companies doing the same thing you'd still get market forces.
Ah but very weak market forces because price discovery is hard when there's only a tiny number of people in the market.
I worked at a place like that. "beneficial units of interest". Tax laws make dividend payments unappetizing, so the place was always swimming in cash, trying to find something to invest in. Leading to lots of really stupid investments. The lower the bank interest rate, the stupider and riskier the investment. If they were momentarily short of cash or about to buy something big, you'd have your bosses boss boss lecturing you in his office about how you really need to take your yearly bonus in stock not cash and up your percentage of salary toward stock, and he can't legally tell you what to do, but plenty of unspoken implications... Price discovery was a bean counter was pressured to give the highest possible unit price that would not result in the auditors swooping in to destroy him, not exactly as efficient as a free market. The company encouraged buying on margin, so you'd have a thousand people borrowing a thousand bucks at crazy interest rates from a hundred banks instead of the company as a whole taking a simple $1M loan at a low corporate interest rate. Makes money for the banks, not so much for the rest of us. Old timers who owned more than say, a quarter million of stock were semi-untouchable, as the company would have to raise a quarter mil cash to buy them out if they quit, retired, or were fired, and/or hire enough people or otherwise encourage the rest of us to buy a 1/4 mill of stock, which made employee / employer relations a bit strained at times, as you couldn't treat peons like dirt as standard at other corporations. Some old timer quits or dies and you can't upgrade your server due to a momentary cash crunch. The execs decided to pay themselves obscene amounts of company stock; they already had more shares than us peons, so they were untouchable. Any random public owner of local electric company stock can complain about the electric company's management's performance because they don't work there... Any owner of company stock has to be a yes man wrt to management performance or get downsized. So we had the best rated, highest paid, yet most incompetent senior management I've ever heard of at any company. If you quit, you make a lot of money. If you're downsized due to a downturn, the price has dropped, you don't make so much money, so its extremely risky. In the good times you don't leave because you'd lose possible gains on the stock. In the bad times, you'd be locking in losses, right when you don't need them. On the other hand, I always owned public stock in competitors, because if they go down, my job goes up, and if my job goes away, my competitor stock goes up, more or less.
Overall, it was a weird, yet none the less profitable, experience. It's possible, but I'm not sure how well it would scale, and given typical levels of corruption I think it would end up collapsing into a reimplementation of our current system. Probably interlocking boards of directors paying themselves obscene bonuses of multiple companies internal stocks, while the rank -n- file workers own about 1% of outstanding shares, in other words pretty much what we have now, but minus the middlemen at NYSE and NASDAQ.
Just watch any of the "Real Housewives" shows
Hardly a new phenomena. Compare the apparent financial condition of "leave it to beaver" or "brady bunch" or "married with children" with what an actual family in those employment scenarios would realistically have...
I remember "The Cosby Show" being savaged by the press because the shows lifestyle was supposedly too extravagant for the father's income as a successful doctor, although they never hassled the detectives on "Miami Vice" for the same mistake, probably was a racial thing.
Barring direct financial contribution to political candidates and forcing them to run on equal funds would help. Barring the movement between high public office and private business, especially government contractors, would help as well
A lotto system is acceptable for jury trials. It should work for legislative and executive branch as well. Provide a lifetime pension of 8 times the median income and mandatory felony prison time if their tax return ever shows any other source of income. Income multiple is high to prevent familial corruption; "vote this way and I'll give your kid a really nice job", "naw, I got enough pension income that he doesn't have to get a job".
Yes, but most of the time you had to pass credit checks to get one.
Not much has changed, in that regard. Also they cost "about" the same, weirdly enough. A couple thousand bucks for about two years of service. Some things never change.
If by "cell phone" you mean more generically any cordless phone where you don't own/maintain the base station, IMTS was pretty popular in the late '60s, if you were wealthy enough.
http://en.wikipedia.org/wiki/Improved_Mobile_Telephone_Service
I'm told the original MTS from the '40s is still sorta around, like in the wilderness.
It sounds like what 4chan /tv/ does every Monday night when there's a new House episode on.
We used to do exactly what has been patented, in Second Life, almost one decade ago. Lets say, 2004-2005ish.
Errors can come from nearly anything; slightly imperfect knowledge of the characteristics of your device
Non-linear effect anywhere in the RX or TX chain, or intermod from surrounding objects is a big problem.
I've done quite a bit of RF design work, microwave ham radio stuff, etc. The big problem is historically low noise stuff which makes a great receiver tends to blow up when subjected to power, and high power gear tends to have horrific weak signal noise characteristics.
A great low noise fractional dB noise figure preamp is off the shelf and cheap, and it'll be vaporized by say 20 dBmW. ... is not good.
A great 30 dBmW MMIC 1 GHz amp is off the shelf and cheap, I have used the watkins johnson devices (yes I know they have a new marketing name which I've temporarily forgotten), and its weak signal noise performance
In contrast, this usage of full-duplex means that both directions are transmitting simultaneously on the same channel, without segmenting or multiplexing it.
I don't actually know how they solved the problem, though, and the article is light on details.
If I had to do it, I'd do traditional 70s era spread spectrum code division multiple access CDMA. Imagine a psuedorandom voltage generator feeding the RX VCO attached to the RX mixer. Then imagine a different psuedorandom voltage gen, or at least the same generating polynomial at a different offset, feeding the TX VCO attached to the TX mixer. Two completely separate RF paths, maybe up to the antenna. Synchronizing two separate psuedorandom voltage gens is merely twice as fun as just one, kinda, I guess.
The other way was to use an old fashioned yet highly effective RF circulator. They are large, and heavy, and frankly kinda hard to make. Think like a hockey puck of ferrite with a big ole magnet. RF only flows clockwise. This is old, old stuff. Larger and heavier than a "brick" cellphone from the 90s, although they worked perfectly fine at the base station.
There's another way to do it using PLLs and the two transmitters in quadrature, but that's getting bizarre (like, have I been drinking this morning already?) and synchronization is gonna be an absolute bear. The hard part isn't static stability, but dynamic as it switches in and out of sync, or multipath interferes with it.
In a world where bandwidth demands are increasing exponentially, a simple doubling of capacity ought to get us by for, oh, I don't, know, 6 months?
Demand might be increasing, but the supply is being choked by throttling and overage charges. Great, now I can reach my cap in 10 minutes instead of an hour. Whoo hoo.
Also there's a pretty hard limit to BW demands... What would I do with a tiny little hand held device with a two hour battery life that would use more than HDTV bandwidth? I reached this limit at home around 2003, and both my monthly bill and comparative service levels have gone from "exotic" to "average" in that interval. I could certainly afford to upgrade to 100 meg service, but I have no idea what I'd do with it that I'm not already doing...
So you wanna start send documents.
1) FAX machine : Tell the phone person to please provision a POTS analog phone line to a jack right there, and tell me the external number. Fax machines are cheap and can be bought on the office expense account (the one used to buy paperclips). For bonus points tell the receptionist your new departmental fax number. Unbox fax, plug in, you're running. You know if it works or not because every far end tells your near end in some manner that is "OK" or not. Support is, if it breaks, buy another. It just works.
2) Scan and email : Fill out request form for IT dept for the hardware. They need to follow the capital expense forms and procedures to buy your $100 flatbed scanner, along with possible competitive bidding, assuming they even have the capital budget remaining for the year. Your bosses bosses boss may need to get permission from his boss to transfer $100 of his capital budget to IT, assuming he has the budget. Its quite trivial to spend thousands in labor on meetings and arguments about spending $100. It may or may not arrive in 3 months and may or may not meet your needs, but you're stuck with the hardware. Fill out a request form for IT to get the scanner software installed on your locked down PCs. Argue endlessly about who will support the system, and how much it will be supported. Eventually you get it working, and every time you send an email with a scan, you have to call or wait for an email response to prove their anti-virus didn't eat it. Its a nightmare.
At home I would never use a fax. But I understand why they're the path of least resistance at businesses.
...and would require integration with multiple OS's, require drivers (the quality of which we don't know), etc.
Making it transparent to the OS means you don't have any of those problems. It's a trade off, and for a first gen drive, probably the better way to go.
Don't forget massive vendor lockin, to be avoided like the plague that it is.
Hmm, only works under one OS, perhaps one kernel version of that OS, with one brand, perhaps one model of drive, perhaps they can find a way to lock in the motherboard too. Um, no thanks. Run away! Run away!
You have a 12TB file server? WTF are you putting on that thing, pirated content you'll never have enough time in your life to watch?
My wife has done semi-pro video editing (as in she's not being fairly compensated for her labor) and as her local tech support guy aka husband I can guarantee that its very difficult to get enough storage space without... semi-pro storage gear, like maybe a ebay surplus NAS or something. Its a huge PITA to use off the shelf solutions. Like if you try to do it all out of a tigerdirect catalog or buying on amazon. She ended up using multiple external HDs in cases, plugging the usb or fw into her mac, and storing raw HD interview footage on the external drives and only "good clips" on the main server for final editing.
For high def documentary type work, figure you'll end up with about 10 tb of raw video per hour of finished high def video. This is assuming storage space is not going to be your limiting factor. If your limiting factor is "filming" time on-site, maybe you can't take enough video to fill 10 TB. If your limiting factor is editing time to meet a deadline for duplication, then maybe you can't edit 10 TB worth of video so no sense taking it. The limiting factor is not always video storage space, although often it is.
Note that if your idea of high end video capture device is a dusty fingerprint smudged cellphone that records 320x240 at a couple frames/sec, then you probably don't need 10 tb. On the other hand, if your camera cost more than a mortgage payment, then you probably need more than 10 TB, if not online, at least available in external plug in cases.
I don't really know much about how she does it, I'm just the guy who hears "the server is full!" no matter how big the drives are. I think about 50 TB online would handle the multiple simultaneous project situation. Any idea how to put 50 TB in my basement, more or less contiguous, in a sensible scalable maintainable reliable manner? Also I only use Debian for infrastructure, if that matters.
is planning to take part in an unprecedented science project — a 20-square-mile model of a small U.S. city.
Note the military has quite a few of these, although smaller scale. Also full of bullet holes. Which might actually be a bonus if you're planning on a technology deployment in "urban" areas.
...for hobos and other mobile homeless. How will they be kept out of all those uninhabited buildings? They may look uninhabitable to people with a place to live, but to the homeless, they might look not too bad.
Don't need a floor for testing? Don't waste time and money building one.
The place could be infested just like a city park. But it should be less of a problem than an abandoned tract of mcmansions, and they seem to be doing fine.
Because burning down 35,000 real homes when your molten salt solar energy storage system fails, or having your water recycling system backflow bad water through the tap is a bigger problem when you fill them with families of 4?
Even "normal failure" needs to be covered up. If it takes 50 revisions to get your SDHW panels not to leak, the last thing you want is 50 families whining on facebook and twitter about how revisions 1 thru 49 of your new panel design leaked water all over their priceless scrapbooks.
Also payment negotiations with a zillion individuals would be a huge PITA.
That was my first thought. I'm still waiting for someone to build a city like Eureka with... well... slightly lower requirements for residence. ;)
Huntsville Alabama? Last time I visited, in the early 90s, around 80% of the population was pure binary, either military uniform or doctorate degree. What do you get if you take a sleepy farm town in 1945 and drop 50 times its population of german V-2 rocket scientists, and friends, on top of it? Nobody could talk at the bar about what they did at work, but you knew whatever it was, it was cool.