Buy several large 4 liter cans of shop grade mineral oil (It doesnt have the "perfume" normally found in baby oil, which is also mineral oil), and a small aquarium. Put some shiny black light aquarium rocks in the bottom.
Put the systemboard, including the heatsink and fan, directly into the aquarium. (I would suggest something in a tiny form factor, like a mini ITX, since you can use a beefy wall wart to power it, rather than a full blown PSU.)
Route cables up and out the top through a repurposed aquarium filter tube. Where the charcoal filter would be in the tank filter, use wireties to control the cabling. You can put a cable switch box in there to serve as the port end connect point, so your normal AV and HID cables arent hanging out of the "water".
Populate the tank with those mechanical bobbing fish, and some bubbler toys.
Fill the tank with the mineral oil, drop in some aquarium bubbler stones for dramatic effect, and place a blacklight in the aquarium tank light of the cover.
VOILA. Something that wont sound like a jet engine, has heavy thermal mass to dissipate the heat of the system, has a continual supply of lubricant for the ball-bearing CPU fan, and makes an interesting accent to the room all in one go.
The fish arent real, and the mineral oil doesnt evaporate, so you never have to clean the tank, assuming you keep ambient dust out.
Really? Every place I have looked, only 2gb sticks were available. 8gb sticks would be awesome.
(No. There is no real ambiguity in saying 1gb == 1024mb, 1mb == 1024kb, 1kb == 1024b. The ambiguity was injected by shyster disc manufacturers, wanting to claim 1mb as 1000kb, instead of the correct 1024kb, because they wanted to sell a lower capacity device as larger than it really was. 1gb is 1gb. I refuse to adopt a whole different suffix just because of marketing drones trying to reinvent the term.)
Say,you want to leave the us. Perhaps a business trip, perhaps a vacation. 99% of the rest of the world uses GSM. That means if you want to keep you phone, and go to another coutry, you will HAVE to use ATT.
You would be foolish to think att wouldn't abuse that natural monopoly, to fleece people it thinks have lots of cash. I would expect att to totally stop giving out unlock codes for their phones shortly after a successful att-TMo merger. Afterall, why would you, a consumer, need to unlock the sim lock if att was the only game in town anyway?
Allowing this merger is deadly to competiton, and unthinkable for international business.
They have reasonable prepaid plans. I can get unlimited text, voice, and data (throttled, but meh) at 50$ a mo. I can get unthrottled data at 70.
The android phones they offer can make use of my home wifi to make and recieve calls, even if the cellular coverage is spotty. I live in the boonies, and this is a major perk. It allows me to keep a big city number where the phone company would charge me long distance otherwise.
They actually give a shit about their customers, or at least appear to more than ATT does.
They are the only other US carrier that is GSM besides the bloated whore that is ATT. The last thing I want to see is ATT shove another cellular carrier up its chancre riddled snatch.
That said, ATT does NOT need T-Mo's spectrum. What they need to do is deploy the spectrum they have more sensibly. Rather than trying to shove 10 thousand subscribers on a single tower, then bitching when they all use the maximum allowed bandwidth-- they need to deploy 10 reduced power output towers that each service 1000 subscribers. They can go ahead and deploy the high power towers in rural areas to maintain their "We have the best coverage!" nonsense (because it is a lie, but meh), but for urban areas such persistent signal is deleterious due to reflections off buildings causing multipath issues, in addition to the obvious one of trying to satisfy the data demands placed on such a network.
So, rather than buying T-Mo, patching the problem in a manner that would require most ATT customers to buy new phones (that have the T-Mo/UK frequency antennas), and then using the GSM monopoly to play king of the mountain-- they need to use the money they would have spent on buying T-Mo, decommission the high power transponders on the urban area towers they have, replace them with lower power ones, and then build more total towers in the poorly serviced urban areas.
Oh, but that is that whole "Invest in infrastructure" thing that they dont want to do.
Fuck ATT. Fuck them with an iron spike on a jackhammer.
All the packager needs to do is pack them in a clear polybag type wallet that is 4 x 4 pockets in size, and include 8 such wallets in a rectangular cardboard box. My quick and dirty geometry skills shows that I could reasonably stuff 128 of them in approximtely the same size box used by a 10 pack cdr box. The biggest cost would be slipping the cards into the pockets of the cheapo polybags.
Current packaging is expensive, because they use blisterpacks that also include a rugged clear plastic case. You don't want that in a bulk pack.
Licensing costs would be the hard one to deal with. Corporate drones don't like to renegotiate licenses.
1) in this case I don't mean all the suppression genes. Duh, of course you need them if you live longer than a gnat...:-) I mean the ones explicitly tied to fibroblast behavior. A few of these genes cause scar formation instead of proper tissue regeneration when they are active. We don't want neoplasms of any sort, we just want to reduce scarring from injury.
2) you misread. The idea is to CAUSE apoptosis when the parasite determines that its host cell is cancerous, or otherwise unsalvageable, not to prevent it. The idea is to kill off anomalous tissue, not keep it around to cause pathologies.:-)
3) again, I am not a geneticist. My original idea was to use an exotic synthetic dna analog, such as xdna, or some of the structural dna like molecules being used for nanotechnology applications. that would require quite a few custom enzymes and proteins though. (Not a trivial matter) if we use xdna, we could use the extra base pairs as parity information though, and the wider structure makes it a little more durable.
4) I remember watching a canned lecture on cellular analysis which featured a set of special inhibitory compounds that could lock a dividing cell in metaphase mitosis for extended periods, at least under a microscope. This would be the ideal time to crawl an enzyme down both sets of replicated chromosomes (host cell is in mid mitosis, just stuck, so all the chromosomes would be neatly unpacked aligned on centromere locations, and bound to chromatin), evaluate the telomere length, and compare against a custom variant of mrna produced by the parasite from the reference copy. An idea of where to start for this custom enzyme might be dna transcriptase. It should generate short chain sequences (say, up to centromere positions?) Of rna from the replicating host genome, direct those to the parasite, where both copies, and another enzyme that crawls the backup dna interact. This enzyme should have 4 active sites. One for each of the two input streams from the host, one attached to the backup genome, and the 4th to generate repair message rna chains. It basically checks to make sure that the host has properly replicated the nuclear genome, and that it also matches the stored one. If it doesn't, it creates the needed messages to cut and repair the host dna, to create a trna snippet from the backup, and then release these into the host cell's cytoplasm.
One of the things I think would be useful is the dna ligation and repair enzymes found in some species of yeast. Excellent thing to add to the parasite's capsid, so that it can create and release these when repairs are needed.
Holding off completion of mitosis is dangerous to the cell though, so these processes would need to complete pretty quickly. This is a pretty complicated process (but then again, so is ordinary transcription), and not garanteed to ensure viable copies, but errors would still have a chance to get caught on the next mitosis cycle.
The replication of the backup genome would occur after the parasite has determined that the copied host genome is correct. It (the parasite) divides, uses the already pulled in rna templates from the host that it used to check with, and uses them to verify the new copies of the backup genome prior to dividing itself. It then permits the host to finish dividing, and a new copy of the parasite goes into each daughter cell.
One way we might get the messages from the host cytoplasm into the parasite is though a custom receptor site on the parasite's cellular membrane, and an engineered protein tag stuck to the message. This would keep ordinary mrna and trna from getting inside. Once inside the parasite, we can do all kinds of things, assuming the parasite genome can handle lots of stuff in it....:-)
By checking at host mitosis time, we would have 3 copies of the host genome to compare against, and should be able to spot transcription erros with a "best 2 out of 3" approach.
As you said though, the devil is in the details. I wouldn't e
While you can't put usb sticks in a wallet, you CAN put sdcards in one. Specifically one made for trading cards.
I would love to see sdcard media get sold in bulk packs like cdrs are. There is a slight problem with capacities not rounding evenly with optical formats... (640-700mb cdr : 1gb sdcard. 4.5gb dvd : 8gb sdcard. 9gb dvd : 12/16gb sdcard) but the form factor is much smaller, you can store waaay more data in a similar sized wallet, and they are less easily destroyed by frequent handling.
Yes. I KNOW they are more expensive. I also remember when cdrs cost over a dollar a pop. These devices don't have to be blazing fast to replace optical media, and while I know it won't be a popular subject with the demographic here, it WOULD work quite well with software firms, because sdcards have to be able to support special hardware drm features to be spec compliant. (This means that your spiffy boxed 3d game you bought off the shelf can chug slowly on install, use your fast sata drive at runtime, and use the sdcard as a dongle to verify game purchase, all in the same package. I am surprised that no software house has tried it yet.
The cards themselves don't need to be fast really, so cheap organic semiconductors, like those used in flexible displays that can exceed amorphous silicon speeds could be used to make the bulk pack cheapo ones.
Like any product, as long as it remains a niche, specialty product it will be expensive, but when it becomes a widespread multi use product, economies of scale drive down the price. I can easily see flash going that way, especially for slow but cheap sdcards.
I have wondered to myself a few times if it would be possible to reprogram an intracellular parasite to become a new "immortality" organelle.
Take for instance, the work with toxoplasma gondii. This is already an intracellular parasite, which has been fully sequenced and even fully reprogrammed in the lab.
we suspect that much of 'old age' is the genetic breakdown of chromosomes from cellular mitosis, which causes a limit to the number of times a healthy cell culture can divide, and further impact the functional health of tissues made from such aged cellular populations.
Incorporating a failsafe backup of the chromosomes of the host, detecting cancer factors, and selectively disabling some the tumor suppression genes in the host that restrict tissue regeneration would radically increase the lifespan of the host.
The idea I had in mind was for the endoparasite to contain a normal bacterial genome capsid, for the organism's own cellular activites, and for the cancer detection and apoptosis trigger of the host--- but also to contain a fully synthetic non-replicating copy of the host's genome. (Perhaps it could be phosphorilated or in some other manner rendered bioologically inactive in the parasite.)
The idea is that as the telomeres of the host's genome break down, it triggers the biological equivalent of running fsck on the host genome, then rebuilds the host telomeres- essentially restarting the cell division clock, and rejuvenating the host tissue.
The problem I haven't come up with a suitable answer for, is how to cope if the organisms end up in the WRONG host.
We don't want aunt mae turning into uncle ben on the genetic level after they shag, for instance.
The organisms need a way to update the template, withou updating to a BROKEN template in the host.
I am not a genetic engineer, so I haven't thout too deeply on the matter, but I could deffinately see something like this turning somebody essentially immortal.
Sounds like what I do. I suggest design changes, and point out critical design flaws all the time. (I have also been given tasks of giving design alternatives in the past.)
Prior to taking this job, I used to custom fab farm equipment and did computer technical work. It is only now that I realize that I was always an engineer.
I am certain that there are no shortages in the total numbers of such people. The problem is that our society looks down on manufacturing and fabrication as being somehow "dirty", and uncouth. It instead seems to epitomize "ideas" rather than solutions.
Software people recognize this immediately when they "work" for people that give them a nebulous set of ideas without any thought put into how it could (or should) be implemented, and treat the finished work as if it was all their hard work, and downplay the real intellects that turned their pipedream into a reality.
Engineers often get the same shortchange, and get replaced just as easily because the pipedreamers look for whoever they can exploit the cheapest.
To further add insult to injury, hobby engineers often get handed an intractible situation where industry and government work to create a perfect storm that requires absurd credentials to do just about anything, which steals even the joy from hobby engineering and fabrication.
(That is to say, granny McParanoid down the street sees him bulding something, thinks it might be a bomb or something because she has no idea about such things, and the next thing he knows, he is being handed fines for doing his hobby while his creation gets exploded by the bomb squad. The usual "do you have a permit for this?" Where the "permit" is only granted to large firms, for crazy money, and where "this" is the area of engineering practice his hobby takes him- be it chemical, electrical, biological, et al. I am very much reminded of the polymer chemist on the east coast who had all his lab notes and research samples destroyed by the local police for daring to do safe polymer research in his house instead of a high priced and zoned research lab.)
If the president wants to turn around the brain drain trend, then he needs to make it safe to be an independent engineer, and put a stop to the anti intellectual and fearmongering madness that current politicians are thriving on.
It might contribute to an education or information delivery crisis. (That crisis being that the reduced commincation infrastructure is creating local pockets of populations that are so poorly educated as to be an outstanding burden on the rest of the country. Ignoring the infrastucture problem would only worsen the crisis.)
Not saying that is the case- more likely just media hyperbole- but possible in theory. I don't know much about the uk's telecom system to know.
What about artificially increasing gravity momentarily?
Reduce the circumference, and rotational velocity increases proportionally. If enertial mass is uneffected, the heavier elements would spin out to the sides, leave the artificial gravity well, and fly off?
You mean like all the other low power broadcasting devices all over that would give false positives, on top of the wasted logistics of aiming rf sniffers up every transient bird's bum?
The idea is that it looks like a bird, and the enemy base crew rightly ignores it. It is a social engineering hack to gain unauthorized intelligence access.
Much like people in call centers don't to background checks on everyone that calls, (and thus fall victim to such attacks), the ground crews of restricted areas don't check out every single thing that enters their airspace. The more realistic the cammoflauge, the less rf antennas get aimed at it.
It could be screaming like nobody's business like a spark gap transmitter, and still be overlooked as the cause. (At least until one gets captured, and the ruse discovered.)
Agreed. The otp is transmitted between the base station and the uav at preflight over a physical data connection prior to activation of the radio com link. Eve would have to live in the data cable to get to the otp when it would matter.
The fpga is indended for extensibility after production. (Say, field replacable modular optical devices, or other special purpose snap ins. The idea is to be able to totally change the way the system behaves without requiring a screwdriver and soldering iron.) It can also be made to nuke itself in the event of a crash, and cheap ones are just that.. cheap.
I agree, a z80 would be more than enough processing oomph, but you might need special functionality "right now" on the field prior to launch.
The problem here is that you assume a single, repeated com channel, and not a military discriminator.
This means that "eve" needs to already know the frequency to be monitoring prior to obtaining the otp to decode the recorded stream. When doing ANYTHING cryptographic, you don't want to have a consistent variable, or the system becomes security through obscurity. (Sony ps3, for instance. You simply don't repeat the same keypair and hold the salt if you want a secure transaction.)
The discriminator should be hopping channels to avoid allowing eve to record the transmission. The encryption should make the raw transmission nearly indistinguishable from random noise for those people without the keys.
The otp can be used to dynamically change aes keys reasonably safely, if needed, and to synchronize discriminators. Each otp is specific to a drone/station pair, is reloaded with real random data on each mission launch, and would be useless if captured after the fact.
More fun can be had id the system is designed to kill all power to volatile memory on the event of a crash landing, killing the otp in the drone's memory in the process. Like I said, no persistent onboard storage other than flight control software in nand.
Aes needs a keypair, iirc. The otp satisfies one of these keys, is specific for that mission and that data stream. (This opposed to reusing keys.)
For weight, and intelligence reasons the device should not have any more onboard storage than a few mb of ram for the flight computer and for the camera to store the image data prior to encryption and broadcast, and some flight control software in nand. This won't need to do complicated vector math like a stealth fighter, so a sophisticated flight computer would be unnecessary.
The design implies short range base stations which should be in direct transmission range. This means the feed should be live to avoid exactly the same problem you just described. (They capture the drone and playback what it was lookinng at. No storage, no local playback. Only a otp, for that one drone. The most they could glean is the encrypt type and the discriminator freq used for data. True random otps would make it computationally unsensible to use captured pads for forensic intelligence purposes.)
I watched the video, which has additional views of the interior.
The part mark plate on the component marked "fmc" and the few metal components of the fuselage of the airframe look suspiciously like lockheed martin's work.
(Disclaimer: I work in aerospace. This looks like their engineering in the metal bulkhead design. If not them, a subsidiary. Do not know the model. The part mark placcard stinks of LM. BOEING uses inkjet partmarking, as did raytheon aero before hawker beech bought them.)
This does not look like a top secret device. No, really, it doesn't.
It looks like a low cost "expendable" craft intended to fly over restricted air spaces.
I say that because the wing and airframe profile appear to have been modeled on the "gliding" look and behavior of a large goose. It would be exremely wasteful of military ordinance to shoot down everything that looks like a goose 100ft in the air that flies over a restricted area.
If I were to design such a craft, it would 1) be very slow and as near to silent as possible. 2) contain absolutely no stealth technologies that might give enemy engineers clues about our radar abilities, and 3) implement an FPGA based one time pad data encryption system to transmit recon data to the nearby recon team.
Ths way if the craft crashes, gets shot down, or captured the expense of replacement is 1) very low due to nearly 100% plastic construction and cheap electronics. 2) data forensically uninteresting from either an engineering pov or from a data espianage point of view. 3) cannot be used to break mission critical data encryption technologies, due to 1:1 one time pad pairings, with quite possibly cheap commercial encryption methods. (256AES, etc.) By the time it is recovered and studied, that pad is black listed as belonging to an mia drone.
This thing has "field recon" practically painted all over it. Lightweight plastic airframe, electronic only propulsion, small battery... all add up to being a disposable device with very short range, low airspeed, and short active runtimes.
Whoever deployed this device was close by. (Unlike a predator which uses petrolium fuel and has a rigid metal airframe that can handle a reasonably fast cruise speed and can perform long mission flighttime, this device has none of those features, and as such cannot realistically be launched from miles away like a predator can.) This looks like it could well be a "backpack" type kit, that folds up for storage and portability. (That's how I would commision such a device anyway.)
All that said, this kind of setup would lend itself well to commercial mass production, since nearly the entire airframe could be injection molded on the cheap. For similar reasons the design would lend itself well to hobby enthusiasts with access to fab labs. Having access to aviation grade CAD equipment, I would *LOVE* to get some detailed photos of every inch of the airframe (with a mm scale metric ruler in the shots) and of the internal cavities.
I really would like to make some community models of this vehicle.
For apple, I can clearly see that being true, but for Microsoft I question that for historical reasons...
Microsoft at the time was not nearly so powerful as it is now. (WinSock hails back to the 3.1, pre internet days. I clearly remember using non-microsoft TCP stacks on win3x, with stand alone PPP solutions for dialup access.) Microsoft was literally scrambling to implement WinSock. (part of the reason why the earliest versions had so many problems. For an idea of how "new" to networking microsoft was at the time, take a more than cursory examination of their primary network protocol of that era--- NetBEUI. To avoid being "Last" to the internet table, and with a seriously broken solution, microsoft would have eaten GPL.)
BSD sockets was attractive because it was, in order of attractiveness:
1) Already mature and use tested. 2) Free 3) Could become part of a microsoft proprietary technology seamlessly.
Replacing with a GPL license, it would still have the first two.
So what I think they would have done is implemented the GPL type code, then started the afore mentioned "Wheel reinvention" as a seperate development project, and when it was mature, phase out the GPL based system.
(Compare to what they did with IBM and OS/2, and their creation of windows NT from it.)
It would have done so with great fanfare that it had built a new technology that has all the core functionality of the former, but now "Better and faster" because it would be 100% microsoft proprietary. This would have fit in perfectly with the "Embrace, Extend, Extinguish" practice of that era. The GPLed interfaces would then be depreciated, and in 2 releases, would be totally dropped.
In that time however, any additions or improvements to the GPLed implementation would have back-fed the community, which is exactly what the GPL is intended to do.
The problem with bsd licensed code is that it quickly becomes fragmented proprietary code.
Like for instance, the bsd sockets implementation.
Microsoft made heavy use of this code to make the earliest version of their winsock api. A modification that is closed.
As far as I know, osx uses a bsd flavored sockets implementation as well. It is quite possibly the most widely used tcp\ip reference stack implementation anywhere.
The issue is that osx sockets, winsock, bsd sockets, et al are all fragmented, and with the exception of the parent bsd implementation, all closed and proprietary.
Had it been licensed under gpl, all the child implementations of the parent would be open, and advancements or improvements could cross proliferate.
That is the real strength of the gpl. The improvements you make to the code to make it useful to you could very well be improvements that others can use to make the code work for them. Instead of fragmenting the code, it helps to unify the code, and helps it to evolve with much less "reinventing the wheel."
The bsd license has its place, but it is no substitute for the gpl.
Anyway, we made the actual list of "bad words" by working off several other lists provided to us from other schools and organizations that have been doing this themselves for years. We combined their lists and edited it down to what you see.
So, what you are basically saying is "Everyone else is doing it!", right? I don't really want to sound like a jackass here, but I think this is a good opportunity to take mom's advice about jumping off bridges to heart.
Personally, I have never understood the whole "That word is inappropriate." mantra. Who defines what is inappropriate, and under what criteria? Likewise for "Inappropriate behavior."
Are there specifically written down guidelines, or is it simply left up to individual faculty and school administrators to decide? If the former, can students and parents be given certified copies to enforce compliance, and if the latter, what is to stop several teachers from instituting mutually exclusive definitions of "Inappropriate"?
Simply blanketbombing "Naughty words" (who defines them as naughty? Is there any arbitration process involved, or is it purely authoritarian?) because certain people "Dont like them" is madness. Guess what, I dont like asparagus. That doesnt give me the right to forbid people eating it in my presence. similar should be true for "naughty" words.
Highly eliptic entry vector perpendicular to the ecliptic, with the close bend of the shot a min safe distance for the probe to not get roasted?
The sun's gravity dwarfs everything else in the system already, so a "wide" shot arc shouldn't pose too much problem.
For shits and giggles you could get more thrust by deploying a mylar screen on the escape portion of the shot to gain accelleration. The same screen would act to retard the probe after it enters the target star's heliopause.
Slashdot Mirror
Simple solution:
Buy several large 4 liter cans of shop grade mineral oil (It doesnt have the "perfume" normally found in baby oil, which is also mineral oil), and a small aquarium. Put some shiny black light aquarium rocks in the bottom.
Put the systemboard, including the heatsink and fan, directly into the aquarium. (I would suggest something in a tiny form factor, like a mini ITX, since you can use a beefy wall wart to power it, rather than a full blown PSU.)
Route cables up and out the top through a repurposed aquarium filter tube. Where the charcoal filter would be in the tank filter, use wireties to control the cabling. You can put a cable switch box in there to serve as the port end connect point, so your normal AV and HID cables arent hanging out of the "water".
Populate the tank with those mechanical bobbing fish, and some bubbler toys.
Fill the tank with the mineral oil, drop in some aquarium bubbler stones for dramatic effect, and place a blacklight in the aquarium tank light of the cover.
VOILA. Something that wont sound like a jet engine, has heavy thermal mass to dissipate the heat of the system, has a continual supply of lubricant for the ball-bearing CPU fan, and makes an interesting accent to the room all in one go.
The fish arent real, and the mineral oil doesnt evaporate, so you never have to clean the tank, assuming you keep ambient dust out.
Really? Every place I have looked, only 2gb sticks were available. 8gb sticks would be awesome.
(No. There is no real ambiguity in saying 1gb == 1024mb, 1mb == 1024kb, 1kb == 1024b. The ambiguity was injected by shyster disc manufacturers, wanting to claim 1mb as 1000kb, instead of the correct 1024kb, because they wanted to sell a lower capacity device as larger than it really was. 1gb is 1gb. I refuse to adopt a whole different suffix just because of marketing drones trying to reinvent the term.)
Right now, I can only get 2x 2gb sticks inside most laptops.
Given the inherent doubling of chip density this offers, when can I expect to be able to purchase 4gb SODIMM packages?
Say,you want to leave the us. Perhaps a business trip, perhaps a vacation. 99% of the rest of the world uses GSM. That means if you want to keep you phone, and go to another coutry, you will HAVE to use ATT.
You would be foolish to think att wouldn't abuse that natural monopoly, to fleece people it thinks have lots of cash. I would expect att to totally stop giving out unlock codes for their phones shortly after a successful att-TMo merger. Afterall, why would you, a consumer, need to unlock the sim lock if att was the only game in town anyway?
Allowing this merger is deadly to competiton, and unthinkable for international business.
The reasons I use T-Mobile:
They have reasonable prepaid plans. I can get unlimited text, voice, and data (throttled, but meh) at 50$ a mo. I can get unthrottled data at 70.
The android phones they offer can make use of my home wifi to make and recieve calls, even if the cellular coverage is spotty. I live in the boonies, and this is a major perk. It allows me to keep a big city number where the phone company would charge me long distance otherwise.
They actually give a shit about their customers, or at least appear to more than ATT does.
They are the only other US carrier that is GSM besides the bloated whore that is ATT. The last thing I want to see is ATT shove another cellular carrier up its chancre riddled snatch.
That said, ATT does NOT need T-Mo's spectrum. What they need to do is deploy the spectrum they have more sensibly. Rather than trying to shove 10 thousand subscribers on a single tower, then bitching when they all use the maximum allowed bandwidth-- they need to deploy 10 reduced power output towers that each service 1000 subscribers. They can go ahead and deploy the high power towers in rural areas to maintain their "We have the best coverage!" nonsense (because it is a lie, but meh), but for urban areas such persistent signal is deleterious due to reflections off buildings causing multipath issues, in addition to the obvious one of trying to satisfy the data demands placed on such a network.
So, rather than buying T-Mo, patching the problem in a manner that would require most ATT customers to buy new phones (that have the T-Mo/UK frequency antennas), and then using the GSM monopoly to play king of the mountain-- they need to use the money they would have spent on buying T-Mo, decommission the high power transponders on the urban area towers they have, replace them with lower power ones, and then build more total towers in the poorly serviced urban areas.
Oh, but that is that whole "Invest in infrastructure" thing that they dont want to do.
Fuck ATT. Fuck them with an iron spike on a jackhammer.
All the packager needs to do is pack them in a clear polybag type wallet that is 4 x 4 pockets in size, and include 8 such wallets in a rectangular cardboard box. My quick and dirty geometry skills shows that I could reasonably stuff 128 of them in approximtely the same size box used by a 10 pack cdr box. The biggest cost would be slipping the cards into the pockets of the cheapo polybags.
Current packaging is expensive, because they use blisterpacks that also include a rugged clear plastic case. You don't want that in a bulk pack.
Licensing costs would be the hard one to deal with. Corporate drones don't like to renegotiate licenses.
1) in this case I don't mean all the suppression genes. Duh, of course you need them if you live longer than a gnat... :-) I mean the ones explicitly tied to fibroblast behavior. A few of these genes cause scar formation instead of proper tissue regeneration when they are active. We don't want neoplasms of any sort, we just want to reduce scarring from injury.
2) you misread. The idea is to CAUSE apoptosis when the parasite determines that its host cell is cancerous, or otherwise unsalvageable, not to prevent it. The idea is to kill off anomalous tissue, not keep it around to cause pathologies. :-)
3) again, I am not a geneticist. My original idea was to use an exotic synthetic dna analog, such as xdna, or some of the structural dna like molecules being used for nanotechnology applications. that would require quite a few custom enzymes and proteins though. (Not a trivial matter) if we use xdna, we could use the extra base pairs as parity information though, and the wider structure makes it a little more durable.
4) I remember watching a canned lecture on cellular analysis which featured a set of special inhibitory compounds that could lock a dividing cell in metaphase mitosis for extended periods, at least under a microscope. This would be the ideal time to crawl an enzyme down both sets of replicated chromosomes (host cell is in mid mitosis, just stuck, so all the chromosomes would be neatly unpacked aligned on centromere locations, and bound to chromatin), evaluate the telomere length, and compare against a custom variant of mrna produced by the parasite from the reference copy. An idea of where to start for this custom enzyme might be dna transcriptase. It should generate short chain sequences (say, up to centromere positions?) Of rna from the replicating host genome, direct those to the parasite, where both copies, and another enzyme that crawls the backup dna interact. This enzyme should have 4 active sites. One for each of the two input streams from the host, one attached to the backup genome, and the 4th to generate repair message rna chains. It basically checks to make sure that the host has properly replicated the nuclear genome, and that it also matches the stored one. If it doesn't, it creates the needed messages to cut and repair the host dna, to create a trna snippet from the backup, and then release these into the host cell's cytoplasm.
One of the things I think would be useful is the dna ligation and repair enzymes found in some species of yeast. Excellent thing to add to the parasite's capsid, so that it can create and release these when repairs are needed.
Holding off completion of mitosis is dangerous to the cell though, so these processes would need to complete pretty quickly. This is a pretty complicated process (but then again, so is ordinary transcription), and not garanteed to ensure viable copies, but errors would still have a chance to get caught on the next mitosis cycle.
The replication of the backup genome would occur after the parasite has determined that the copied host genome is correct. It (the parasite) divides, uses the already pulled in rna templates from the host that it used to check with, and uses them to verify the new copies of the backup genome prior to dividing itself. It then permits the host to finish dividing, and a new copy of the parasite goes into each daughter cell.
One way we might get the messages from the host cytoplasm into the parasite is though a custom receptor site on the parasite's cellular membrane, and an engineered protein tag stuck to the message. This would keep ordinary mrna and trna from getting inside. Once inside the parasite, we can do all kinds of things, assuming the parasite genome can handle lots of stuff in it....:-)
By checking at host mitosis time, we would have 3 copies of the host genome to compare against, and should be able to spot transcription erros with a "best 2 out of 3" approach.
As you said though, the devil is in the details. I wouldn't e
While you can't put usb sticks in a wallet, you CAN put sdcards in one. Specifically one made for trading cards.
I would love to see sdcard media get sold in bulk packs like cdrs are. There is a slight problem with capacities not rounding evenly with optical formats... (640-700mb cdr : 1gb sdcard. 4.5gb dvd : 8gb sdcard. 9gb dvd : 12/16gb sdcard) but the form factor is much smaller, you can store waaay more data in a similar sized wallet, and they are less easily destroyed by frequent handling.
Yes. I KNOW they are more expensive. I also remember when cdrs cost over a dollar a pop. These devices don't have to be blazing fast to replace optical media, and while I know it won't be a popular subject with the demographic here, it WOULD work quite well with software firms, because sdcards have to be able to support special hardware drm features to be spec compliant. (This means that your spiffy boxed 3d game you bought off the shelf can chug slowly on install, use your fast sata drive at runtime, and use the sdcard as a dongle to verify game purchase, all in the same package. I am surprised that no software house has tried it yet.
The cards themselves don't need to be fast really, so cheap organic semiconductors, like those used in flexible displays that can exceed amorphous silicon speeds could be used to make the bulk pack cheapo ones.
Like any product, as long as it remains a niche, specialty product it will be expensive, but when it becomes a widespread multi use product, economies of scale drive down the price. I can easily see flash going that way, especially for slow but cheap sdcards.
Technically speaking, dna already has a raid1 configuration, this would be hybrid raid.
(Double helix has 2 sides that mirror each other, so already raid 1.)
I have wondered to myself a few times if it would be possible to reprogram an intracellular parasite to become a new "immortality" organelle.
Take for instance, the work with toxoplasma gondii. This is already an intracellular parasite, which has been fully sequenced and even fully reprogrammed in the lab.
we suspect that much of 'old age' is the genetic breakdown of chromosomes from cellular mitosis, which causes a limit to the number of times a healthy cell culture can divide, and further impact the functional health of tissues made from such aged cellular populations.
Incorporating a failsafe backup of the chromosomes of the host, detecting cancer factors, and selectively disabling some the tumor suppression genes in the host that restrict tissue regeneration would radically increase the lifespan of the host.
The idea I had in mind was for the endoparasite to contain a normal bacterial genome capsid, for the organism's own cellular activites, and for the cancer detection and apoptosis trigger of the host--- but also to contain a fully synthetic non-replicating copy of the host's genome. (Perhaps it could be phosphorilated or in some other manner rendered bioologically inactive in the parasite.)
The idea is that as the telomeres of the host's genome break down, it triggers the biological equivalent of running fsck on the host genome, then rebuilds the host telomeres- essentially restarting the cell division clock, and rejuvenating the host tissue.
The problem I haven't come up with a suitable answer for, is how to cope if the organisms end up in the WRONG host.
We don't want aunt mae turning into uncle ben on the genetic level after they shag, for instance.
The organisms need a way to update the template, withou updating to a BROKEN template in the host.
I am not a genetic engineer, so I haven't thout too deeply on the matter, but I could deffinately see something like this turning somebody essentially immortal.
Sounds like what I do. I suggest design changes, and point out critical design flaws all the time. (I have also been given tasks of giving design alternatives in the past.)
Prior to taking this job, I used to custom fab farm equipment and did computer technical work. It is only now that I realize that I was always an engineer.
I am certain that there are no shortages in the total numbers of such people. The problem is that our society looks down on manufacturing and fabrication as being somehow "dirty", and uncouth. It instead seems to epitomize "ideas" rather than solutions.
Software people recognize this immediately when they "work" for people that give them a nebulous set of ideas without any thought put into how it could (or should) be implemented, and treat the finished work as if it was all their hard work, and downplay the real intellects that turned their pipedream into a reality.
Engineers often get the same shortchange, and get replaced just as easily because the pipedreamers look for whoever they can exploit the cheapest.
To further add insult to injury, hobby engineers often get handed an intractible situation where industry and government work to create a perfect storm that requires absurd credentials to do just about anything, which steals even the joy from hobby engineering and fabrication.
(That is to say, granny McParanoid down the street sees him bulding something, thinks it might be a bomb or something because she has no idea about such things, and the next thing he knows, he is being handed fines for doing his hobby while his creation gets exploded by the bomb squad. The usual "do you have a permit for this?" Where the "permit" is only granted to large firms, for crazy money, and where "this" is the area of engineering practice his hobby takes him- be it chemical, electrical, biological, et al. I am very much reminded of the polymer chemist on the east coast who had all his lab notes and research samples destroyed by the local police for daring to do safe polymer research in his house instead of a high priced and zoned research lab.)
If the president wants to turn around the brain drain trend, then he needs to make it safe to be an independent engineer, and put a stop to the anti intellectual and fearmongering madness that current politicians are thriving on.
It might contribute to an education or information delivery crisis. (That crisis being that the reduced commincation infrastructure is creating local pockets of populations that are so poorly educated as to be an outstanding burden on the rest of the country. Ignoring the infrastucture problem would only worsen the crisis.)
Not saying that is the case- more likely just media hyperbole- but possible in theory. I don't know much about the uk's telecom system to know.
What about artificially increasing gravity momentarily?
Reduce the circumference, and rotational velocity increases proportionally. If enertial mass is uneffected, the heavier elements would spin out to the sides, leave the artificial gravity well, and fly off?
You mean like all the other low power broadcasting devices all over that would give false positives, on top of the wasted logistics of aiming rf sniffers up every transient bird's bum?
The idea is that it looks like a bird, and the enemy base crew rightly ignores it. It is a social engineering hack to gain unauthorized intelligence access.
Much like people in call centers don't to background checks on everyone that calls, (and thus fall victim to such attacks), the ground crews of restricted areas don't check out every single thing that enters their airspace. The more realistic the cammoflauge, the less rf antennas get aimed at it.
It could be screaming like nobody's business like a spark gap transmitter, and still be overlooked as the cause. (At least until one gets captured, and the ruse discovered.)
Agreed. The otp is transmitted between the base station and the uav at preflight over a physical data connection prior to activation of the radio com link. Eve would have to live in the data cable to get to the otp when it would matter.
The fpga is indended for extensibility after production. (Say, field replacable modular optical devices, or other special purpose snap ins. The idea is to be able to totally change the way the system behaves without requiring a screwdriver and soldering iron.) It can also be made to nuke itself in the event of a crash, and cheap ones are just that.. cheap.
I agree, a z80 would be more than enough processing oomph, but you might need special functionality "right now" on the field prior to launch.
The problem here is that you assume a single, repeated com channel, and not a military discriminator.
This means that "eve" needs to already know the frequency to be monitoring prior to obtaining the otp to decode the recorded stream. When doing ANYTHING cryptographic, you don't want to have a consistent variable, or the system becomes security through obscurity. (Sony ps3, for instance. You simply don't repeat the same keypair and hold the salt if you want a secure transaction.)
The discriminator should be hopping channels to avoid allowing eve to record the transmission. The encryption should make the raw transmission nearly indistinguishable from random noise for those people without the keys.
The otp can be used to dynamically change aes keys reasonably safely, if needed, and to synchronize discriminators. Each otp is specific to a drone/station pair, is reloaded with real random data on each mission launch, and would be useless if captured after the fact.
More fun can be had id the system is designed to kill all power to volatile memory on the event of a crash landing, killing the otp in the drone's memory in the process. Like I said, no persistent onboard storage other than flight control software in nand.
Aes needs a keypair, iirc. The otp satisfies one of these keys, is specific for that mission and that data stream. (This opposed to reusing keys.)
For weight, and intelligence reasons the device should not have any more onboard storage than a few mb of ram for the flight computer and for the camera to store the image data prior to encryption and broadcast, and some flight control software in nand. This won't need to do complicated vector math like a stealth fighter, so a sophisticated flight computer would be unnecessary.
The design implies short range base stations which should be in direct transmission range. This means the feed should be live to avoid exactly the same problem you just described. (They capture the drone and playback what it was lookinng at. No storage, no local playback. Only a otp, for that one drone. The most they could glean is the encrypt type and the discriminator freq used for data. True random otps would make it computationally unsensible to use captured pads for forensic intelligence purposes.)
No, its a small, off-duty chzecloslovakian traffic warden.
I watched the video, which has additional views of the interior.
The part mark plate on the component marked "fmc" and the few metal components of the fuselage of the airframe look suspiciously like lockheed martin's work.
(Disclaimer: I work in aerospace. This looks like their engineering in the metal bulkhead design. If not them, a subsidiary. Do not know the model. The part mark placcard stinks of LM. BOEING uses inkjet partmarking, as did raytheon aero before hawker beech bought them.)
This does not look like a top secret device. No, really, it doesn't.
It looks like a low cost "expendable" craft intended to fly over restricted air spaces.
I say that because the wing and airframe profile appear to have been modeled on the "gliding" look and behavior of a large goose. It would be exremely wasteful of military ordinance to shoot down everything that looks like a goose 100ft in the air that flies over a restricted area.
If I were to design such a craft, it would 1) be very slow and as near to silent as possible. 2) contain absolutely no stealth technologies that might give enemy engineers clues about our radar abilities, and 3) implement an FPGA based one time pad data encryption system to transmit recon data to the nearby recon team.
Ths way if the craft crashes, gets shot down, or captured the expense of replacement is 1) very low due to nearly 100% plastic construction and cheap electronics. 2) data forensically uninteresting from either an engineering pov or from a data espianage point of view. 3) cannot be used to break mission critical data encryption technologies, due to 1:1 one time pad pairings, with quite possibly cheap commercial encryption methods. (256AES, etc.) By the time it is recovered and studied, that pad is black listed as belonging to an mia drone.
This thing has "field recon" practically painted all over it. Lightweight plastic airframe, electronic only propulsion, small battery... all add up to being a disposable device with very short range, low airspeed, and short active runtimes.
Whoever deployed this device was close by. (Unlike a predator which uses petrolium fuel and has a rigid metal airframe that can handle a reasonably fast cruise speed and can perform long mission flighttime, this device has none of those features, and as such cannot realistically be launched from miles away like a predator can.) This looks like it could well be a "backpack" type kit, that folds up for storage and portability. (That's how I would commision such a device anyway.)
All that said, this kind of setup would lend itself well to commercial mass production, since nearly the entire airframe could be injection molded on the cheap. For similar reasons the design would lend itself well to hobby enthusiasts with access to fab labs. Having access to aviation grade CAD equipment, I would *LOVE* to get some detailed photos of every inch of the airframe (with a mm scale metric ruler in the shots) and of the internal cavities.
I really would like to make some community models of this vehicle.
No no silly! They will invest in the iRack!
For apple, I can clearly see that being true, but for Microsoft I question that for historical reasons...
Microsoft at the time was not nearly so powerful as it is now. (WinSock hails back to the 3.1, pre internet days. I clearly remember using non-microsoft TCP stacks on win3x, with stand alone PPP solutions for dialup access.) Microsoft was literally scrambling to implement WinSock. (part of the reason why the earliest versions had so many problems. For an idea of how "new" to networking microsoft was at the time, take a more than cursory examination of their primary network protocol of that era--- NetBEUI. To avoid being "Last" to the internet table, and with a seriously broken solution, microsoft would have eaten GPL.)
BSD sockets was attractive because it was, in order of attractiveness:
1) Already mature and use tested.
2) Free
3) Could become part of a microsoft proprietary technology seamlessly.
Replacing with a GPL license, it would still have the first two.
So what I think they would have done is implemented the GPL type code, then started the afore mentioned "Wheel reinvention" as a seperate development project, and when it was mature, phase out the GPL based system.
(Compare to what they did with IBM and OS/2, and their creation of windows NT from it.)
It would have done so with great fanfare that it had built a new technology that has all the core functionality of the former, but now "Better and faster" because it would be 100% microsoft proprietary. This would have fit in perfectly with the "Embrace, Extend, Extinguish" practice of that era. The GPLed interfaces would then be depreciated, and in 2 releases, would be totally dropped.
In that time however, any additions or improvements to the GPLed implementation would have back-fed the community, which is exactly what the GPL is intended to do.
The problem with bsd licensed code is that it quickly becomes fragmented proprietary code.
Like for instance, the bsd sockets implementation.
Microsoft made heavy use of this code to make the earliest version of their winsock api. A modification that is closed.
As far as I know, osx uses a bsd flavored sockets implementation as well. It is quite possibly the most widely used tcp\ip reference stack implementation anywhere.
The issue is that osx sockets, winsock, bsd sockets, et al are all fragmented, and with the exception of the parent bsd implementation, all closed and proprietary.
Had it been licensed under gpl, all the child implementations of the parent would be open, and advancements or improvements could cross proliferate.
That is the real strength of the gpl. The improvements you make to the code to make it useful to you could very well be improvements that others can use to make the code work for them. Instead of fragmenting the code, it helps to unify the code, and helps it to evolve with much less "reinventing the wheel."
The bsd license has its place, but it is no substitute for the gpl.
Anyway, we made the actual list of "bad words" by working off several other lists provided to us from other schools and organizations that have been doing this themselves for years. We combined their lists and edited it down to what you see.
So, what you are basically saying is "Everyone else is doing it!", right? I don't really want to sound like a jackass here, but I think this is a good opportunity to take mom's advice about jumping off bridges to heart.
Personally, I have never understood the whole "That word is inappropriate." mantra. Who defines what is inappropriate, and under what criteria? Likewise for "Inappropriate behavior."
Are there specifically written down guidelines, or is it simply left up to individual faculty and school administrators to decide? If the former, can students and parents be given certified copies to enforce compliance, and if the latter, what is to stop several teachers from instituting mutually exclusive definitions of "Inappropriate"?
Simply blanketbombing "Naughty words" (who defines them as naughty? Is there any arbitration process involved, or is it purely authoritarian?) because certain people "Dont like them" is madness. Guess what, I dont like asparagus. That doesnt give me the right to forbid people eating it in my presence. similar should be true for "naughty" words.
Use the sun?
Highly eliptic entry vector perpendicular to the ecliptic, with the close bend of the shot a min safe distance for the probe to not get roasted?
The sun's gravity dwarfs everything else in the system already, so a "wide" shot arc shouldn't pose too much problem.
For shits and giggles you could get more thrust by deploying a mylar screen on the escape portion of the shot to gain accelleration. The same screen would act to retard the probe after it enters the target star's heliopause.