I hate replying to posts that sound like trolls. Who cares if it originally demoed on a Cray or an SGI? Game development frequently takes a few years and plans can change. Once upon a time, Halo was supposed to be a PlayStation game too.
But that was when it was a Myth/scifi hybrid. It was much more of a RTS game, not any sort of first-person game.
Bungie posted a history of Halo a few months ago...
Let's not forget that Connectix wrote Virtual Game Station (VGS), a PlayStation emulator for Mac and PC. So they certainly have experience writing game console emulators.
So yes, they certainly have plenty of in-house experience if all of those Connectix folks are still around.
They're posting the videos of each level on Mondays, Wednesdays and Fridays over the next few weeks--the contest just finished. The first level has already been posted and is an impressive piece of work.
The BitTorrent for the Pillar of Autumn video is here.
No, I've upgraded and can confirm it does not fix any exploits that were not fixed by the recent security update patch (i.e., only the help/runscript exploit).
I'd agree that a careful overhaul is needed to properly fix these exploits. But the clock is ticking on the exploit problem!
XGrid is an extremely interesting project, but it's not designed to take on a dedicated, custom-designed cluster like VT's Big Mac.
Some calculations can be split into pieces that don't require much "talk" with other pieces. For example, Apple's Mandelbrot demo--you don't need to know what's running on other processors.
OTOH, many problems require quite a bit of cross-talk with other processors. For example, most of the quantum chemistry calculations I run require calculating big integrals. These are run across multi-proc boxes or clusters, but the speedup depends a *lot* on the latency of the network. So XGrid won't really help here--most of the ad-hoc networks serviced by XGrid would have something like 100MBs Ethernet, which is slow.
I'm willing to put up $$ to use supercomputing centers like VT's Big Mac because they're *designed* to handle hard-core parallel number-crunching. Right now, I'm running jobs on a 24-proc POWER3 cluster with 4GB RAM per processor. (Yes, the extra RAM really helps too since I don't hit the hard drive much.)
I think XGrid will see a lot of use for academic or corporate environments to allow adhoc clustering. As an example, I can run some calcs on an XGrid "cluster" at night on all of the desktop Macs in a lab or across an office. These won't be anywhere near as fast as a well-designed cluster. But it will give me access to "untapped" CPU cycles.
Granted, they're a series of novels, but the Red Mars, Green Mars, Blue Mars books by Kim Stanley Robinson explores a lot of these issues.
I won't give away the plot, since a lot of posters here seem not to have read the novels. But suffice to say I think he's written an excellent summary of many issues and I think it's a fairly reasonable scenario politics/sociology-wise.
I'm not sure I agree with "interface details need to be smaller..." Ever use a large-screen monitor with an older user? My parents, for example, have a high resolution (to edit full-page newsletters, etc.) but you can see my Mom pause when she targets a close box.
And a professor here was checking out my AlBook with 10.3 and said he thought the writing was too small for the menus, etc.
Personally, I think it's pretty decent. You think they're too big. My parents think they're too small.
I don't think you can win, except that you can change the screen resolution. Interface elements are a much smaller % at high resolution.
OK, you ask some good questions. I'm going to replace your "retrosynthetic chemistry programs" with "a really good synthetic chemist." Right now, my friends in synthetic chemistry research aren't worried for their jobs because there are a few programs that attempt retrosynthetic analysis. It's really, really hard stuff. (Ask any orgo student.)
(a) I think what you're asking is something like this... If I draw an arbitrary molecule on paper that looks like it's chemically plausible (i.e. no bonds to helium atoms), how likely is it that *some* synthesis could make that molecule?
I don't know the answer. I don't know if anyone does right now. Chemists don't usually work that way. We pick specific targets, not arbitrary ones.
But I'd guess the answer is "maybe 80%" or more just can't be made under standard conditions. Chemistry is good at making molecules, but a lot of things I see are actually made under inert atmosphere, no water, high pressure, etc. Even ignoring that, there are plenty of examples in the chemical literature of a target molecule that no one seems to be able to make, period, even though they seem plausible. (So let's say "50%" can't be made period.)
Octanitro-cubane is a great example. It was finally made last year or so, after something like 10 years of research, several lab explosions and a few grad student injuries. (I think someone lost fingers.)
And in my research, I've done QM calculations on molecules that are certainly plausible. But as it turns out, adding group X to that part of this known molecule makes this bond here pretty weak and so the whole thing isomerizes into a more stable form.
Take cases where you have two chemically similar molecules called "tautomers." It's hard to know which one is more stable and they often interconvert. Even good chemists do some work and realize that the isomer they want will always convert to the other form.
Bonds rotate. Bonds vibrate. A lot of chemistry happens inside a molecule itself--a hydrogen moves or a double bond changes orientation. Yuck!
Anyway, that's a few things that can happen chemically that can stop you from making an arbitrary molecule. That's why good chemists have great job security.:-) Oh, I didn't mention that it's a broad enough field that if I drew molecule X, I might not have the experience in these types of molecules to have the slightest clue how to do it. Synthetic chemists focus on one very small area in grad school so they can at least be good on a few types of reactions.
(b) Ah... Certainly you're correct that a lot of reactions just have crummy yield and require a lot of purification. So if we had this magical nanoassembler, new products could be made. And we'd probably learn more mechanisms--our current techniques don't reveal much on this length scale and mechanistic chemistry can have some "hand waving" explanations for some reactions.
But there are plenty of cases I've had myself, when a supposedly perfect reaction should work and simply does not produce the product you expect (if it produces anything besides tar). Talk to a synthetic chemist and you'll find that they sometimes run the same reaction a few times with different results.
Put simply, some bonds just don't seem to be possible. And we don't know why.
You back off of some claims I've heard from Drexler's camp when you say "wide variety of products." See, we probably can make a "wide variety" of molecules--we already do that in lab! But a "wide variety" is a far cry from being able to design an arbitrary diamondoid nanogear.
I honestly think that after a few hundred years of synthetic chemistry, the fraction of things that we can make, relative to the molecules I could draw on paper is still quite small.
(Full disclosure, I would describe my research as nanoscience and I think the field has lots of promise.)
The problem is that the facility you mention (and the particular equipment you mention) is not a nano-assembler in Drexler's sense. It is not *itself* nano-scale.
See, if you use an AFM or something macroscale to make things atom-by-atom, you're largely limited to making things one at a time.
If you have a nanoscale assembler, you could have 10^23 of them and then you can make something in a reasonable amount of time (e.g. microsecons) and make a reasonable number too (e.g. millions).
We can make macroscopic objects out of complex components already.
But the problem is this--Drexler's theory is that we can make an arbitrary object. That's not necessarily true from biochemistry. There are a great diversity of molecules made by nature. But synthetic chemistry has been able to make molecules never made by nature.
Does that mean we can use biochemical techniques to assemble macroscopic assemblies? No.
The trick that life uses is called "self assembly." We haven't the least clue how proteins form 3D shapes from their constituents. It's a great unsolved problem in biology and chemistry. The first one to solve it wins at least ONE Nobel prize.
From current research, we know that we cannot self-assemble every molecule we can imagine. Some will self-assemble and some different types of assemblies are possible. But we're still a *long* way from being able to assemble an abitrary combination--which Drexler requires.
And if you resort to what life can do, we're quite limited. Has life ever made a skyscraper?
The notion that mechanosynthesis will not work seems to contradict current chemical methods where chemical reactions occur by random interactions between atoms/molecules. If these aren't random mechanical interactions then I must misunderstand chemistry.
Oh sorry, I forgot to rebut your point about "random mechanical interactions." If you take mechanism classes, you'll find out that 95% or more of chemical reactions are quite intricate. When you're working on 10^23 molecules, you can win--some of those will line up correctly and go through. Some won't.
Drexler works on the scale of *ONE* assembler at a time. What does it do if it doesn't have the orbitals in the "tip" aligned correctly? How does it make sure that pathway goes through in a given time?
I'm sorry, that's irrelevant. That's like saying someone told a false statement because he's a Republican. (Take your pick on insult you'd like to throw.)
I could be a bum on the street and still tell you the correct science. You might not believe me, but it's still correct.
As for Mr. Drexler, I've read Nanosystems. Mr. Drexler doesn't know chemistry. If he did, he could tell me all the cool new reactions we need to create the stuff he proposes. Or the chemistry/physics needed to do a nanoassembler.
I've done plenty of computational chemistry research--it's about 90% of my Ph.D. And you know what? I can happily draw whatever molecule I want on the screen and predict the properties. Can I make it?
NO, not necessarily!
There's a reason a lot of people hate orgo class in college. Chemistry is tough--there are a lot of exceptions and the best synthetic chemists have years upon years of experience in lab bumping their heads against walls trying to make things.
Drexler needs to try some synthetic chemistry. Maybe then he'll rethink his nanoassembler idea.
Nobody knows. The assumption is that this is possible--after all, we're not always made of the same atoms. If I ingest some isotope-labeled food and I see that some of the nitrogen ends up in my brain, then a few weeks later it's gone.
Our bodies recycle proteins, carbohydrates and fats on a continual basis. We have to be more than just the sum of our quantum states. But how can we be sure what's important and what's not?
Do you want to run that experiment?
Do you remember that old IBM ad where someone looks like they're shoplifting--stuffing things in his coat, then he walks out and the security guard gives him his receipt?
We already have RFID tags to prevent shoplifting CDs and stuff. Now people want to print RFID circuits onto everything and have the supermarket line just scan everything by radio wave instantly.
The chemistry for printed electronics exists and circuit designs have been made. What remains (as the parent poster indicated) is the cost. Walmart bailed on RFIDs because they're still too expensive. Everyone wants them under $0.01.
Actually, it's relatively easy to build in some sensor tech into your RFID circuits. Many people want printed RFID tags using organic conductive molecules and there are already shipping "electronic nose" devices for detecting all sorts of airborne molecules with the same compounds.
Add in a mass sensor to realize your paint cans are half-empty, one to determine that there's pigment in the "white" paint, etc.
In *theory* at least you can do all this with the same circuits. It's chemically possible, the electrical engineering and the physics has been done, etc. Now if someone actually makes a cheap enough RFID that can return this sort of information anytime soon is someone else's thing.
Dammit Jim, I'm a scientist, not an economist! -Geoff
Sounds like a great idea, doesn't it? Then you realize how expensive it is to make your RFID circuits keep all those bits around. Printed electronics will probably bring RFID cheaply enough to be useful, but then you have the cost and area constraints on the number of bits.
A recent research paper showcased a 40-bit adder via printed electronics. And you want how many bits for your tags? Remember each of these RFIDs should cost $0.01 to print.
A lot of people would like to use "printed electronics" for RFID tags. (Organic conductive materials are often quite soluble and can be printed in inkjet printers or probably even offset presses.)
The catch is that it's still much harder to make a bit with printed electronics than with silicon, plus adding bits requires adding more addressing lines, control circuits, etc. So while those of us in the chemistry lab are making single transistors, the electrical engineers are showcasing things like 40-bit adders from printed electronics. And that's a pretty big circuit area-wise.
If you want RFID tags to take off, you want them for like 1-cent each, from what I've heard in seminars recently. Plus there's that area restriction--you don't want it to be the size of your entire cereal box.
So my guess is the 96-bits vs. 128 comes from the current printed electronics research.
That would be my reading of that announcement too. Of course it's the initial proposal, so it'll get all sorts of changes along the way.
No, it's not necessarily a bad thing. There's a whole lot of work that's completely hidden from view that would be opened up to academic research. I can think of several chemistry programs I'd love to get in source form.
But it would be quite interesting to see how they decide to make the cutoff. TCP/IP was government-funded research. Does that mean anything that uses it must be released? (This is why IANAL.)
-Geoff
Wonderful news! What's next?
on
Open Source Science
·
· Score: 3, Insightful
This would be great news and will certainly level the research playing field. The major research universities and companies can currently afford the massive prices required to subscribe to a wide variety of journals. One of the first responces, IMHO will be the increase in research quality outside the current "research farms."
But I'm not sure I agree there are "excessive profits" at journals, especially since some of them have recently spent big $$ to digitize and archive old articles--in many cases dating back over a hundred years. But since many of us are almost exclusively using online access to journals, distribution charges will decrease dramatically.
So the big question isn't whether this should happen--it should. Science ideally should be a meritocracy of ideas, not dependent on how much your school is willing to spend on journals. But the big question is who pays in this new model. Someone has to review and edit articles. Someone has to pay for the bandwidth of the journal. So do we go back to the "you pay if you want to publish" method (bad idea--only the rich can publish) or will public funds go to public distribution (i.e. the public library model).
Slashdot Mirror
Sorry... sounded more like a "Macs don't get games until years go by" sorta thing.
:-)
I too thought that first Myth/Halo thing seemed really cool. I remember thinking about 64-player battles and that sort of thing.
Fun.
-Geoff
(waiting for Halo 2 goodness... hope I have the free time...
I hate replying to posts that sound like trolls. Who cares if it originally demoed on a Cray or an SGI? Game development frequently takes a few years and plans can change. Once upon a time, Halo was supposed to be a PlayStation game too.
e xboxhistory040904&p=42
i on=History&subsection=Main&page=6
But that was when it was a Myth/scifi hybrid. It was much more of a RTS game, not any sort of first-person game.
Bungie posted a history of Halo a few months ago...
http://www.bungie.net/News/TopStory.aspx?story=pr
And of course they're pretty upfront about their history too.
http://www.bungie.net/Inside/CustomPage.aspx?sect
Just wanted to get some tru7h out there...
Cheers,
-Geoff
Let's not forget that Connectix wrote Virtual Game Station (VGS), a PlayStation emulator for Mac and PC. So they certainly have experience writing game console emulators.
So yes, they certainly have plenty of in-house experience if all of those Connectix folks are still around.
-Geoff
They're posting the videos of each level on Mondays, Wednesdays and Fridays over the next few weeks--the contest just finished. The first level has already been posted and is an impressive piece of work.
The BitTorrent for the Pillar of Autumn video is here.
-Geoff
Let's see...
With a Bluetooth enabled Mac, there's:
Salling Clicker:
http://homepage.mac.com/jonassalling/Shareware/Cl
Romeo:
http://www.irowan.com/romeo/
And of course there are the standard media remotes, e.g. Keyspan Media Remote:
http://www.keyspan.com/products/usb/remote/
Heck, pretty much any PowerPoint remote control could probably be used for remote iTunes control, within reason. But the Bluetooth remotes are sweet.
-Geoff
No, I've upgraded and can confirm it does not fix any exploits that were not fixed by the recent security update patch (i.e., only the help/runscript exploit).
I'd agree that a careful overhaul is needed to properly fix these exploits. But the clock is ticking on the exploit problem!
-Geoff
XGrid is an extremely interesting project, but it's not designed to take on a dedicated, custom-designed cluster like VT's Big Mac.
Some calculations can be split into pieces that don't require much "talk" with other pieces. For example, Apple's Mandelbrot demo--you don't need to know what's running on other processors.
OTOH, many problems require quite a bit of cross-talk with other processors. For example, most of the quantum chemistry calculations I run require calculating big integrals. These are run across multi-proc boxes or clusters, but the speedup depends a *lot* on the latency of the network. So XGrid won't really help here--most of the ad-hoc networks serviced by XGrid would have something like 100MBs Ethernet, which is slow.
I'm willing to put up $$ to use supercomputing centers like VT's Big Mac because they're *designed* to handle hard-core parallel number-crunching. Right now, I'm running jobs on a 24-proc POWER3 cluster with 4GB RAM per processor. (Yes, the extra RAM really helps too since I don't hit the hard drive much.)
I think XGrid will see a lot of use for academic or corporate environments to allow adhoc clustering. As an example, I can run some calcs on an XGrid "cluster" at night on all of the desktop Macs in a lab or across an office. These won't be anywhere near as fast as a well-designed cluster. But it will give me access to "untapped" CPU cycles.
Granted, they're a series of novels, but the Red Mars, Green Mars, Blue Mars books by Kim Stanley Robinson explores a lot of these issues.
I won't give away the plot, since a lot of posters here seem not to have read the novels. But suffice to say I think he's written an excellent summary of many issues and I think it's a fairly reasonable scenario politics/sociology-wise.
-Geoff
I'm not sure I agree with "interface details need to be smaller..." Ever use a large-screen monitor with an older user? My parents, for example, have a high resolution (to edit full-page newsletters, etc.) but you can see my Mom pause when she targets a close box.
And a professor here was checking out my AlBook with 10.3 and said he thought the writing was too small for the menus, etc.
Personally, I think it's pretty decent. You think they're too big. My parents think they're too small.
I don't think you can win, except that you can change the screen resolution. Interface elements are a much smaller % at high resolution.
Fair enough. But what else are you going to use?
Chemistry, broadly defined, has always been about making molecules.
I won't argue that Drexler hasn't been influential or that he hasn't raised interesting problems. But I stand by what I said.
How do you create an arbitrary molecule?
OK, you ask some good questions. I'm going to replace your "retrosynthetic chemistry programs" with "a really good synthetic chemist." Right now, my friends in synthetic chemistry research aren't worried for their jobs because there are a few programs that attempt retrosynthetic analysis. It's really, really hard stuff. (Ask any orgo student.)
:-) Oh, I didn't mention that it's a broad enough field that if I drew molecule X, I might not have the experience in these types of molecules to have the slightest clue how to do it. Synthetic chemists focus on one very small area in grad school so they can at least be good on a few types of reactions.
(a) I think what you're asking is something like this... If I draw an arbitrary molecule on paper that looks like it's chemically plausible (i.e. no bonds to helium atoms), how likely is it that *some* synthesis could make that molecule?
I don't know the answer. I don't know if anyone does right now. Chemists don't usually work that way. We pick specific targets, not arbitrary ones.
But I'd guess the answer is "maybe 80%" or more just can't be made under standard conditions. Chemistry is good at making molecules, but a lot of things I see are actually made under inert atmosphere, no water, high pressure, etc. Even ignoring that, there are plenty of examples in the chemical literature of a target molecule that no one seems to be able to make, period, even though they seem plausible. (So let's say "50%" can't be made period.)
Octanitro-cubane is a great example. It was finally made last year or so, after something like 10 years of research, several lab explosions and a few grad student injuries. (I think someone lost fingers.)
And in my research, I've done QM calculations on molecules that are certainly plausible. But as it turns out, adding group X to that part of this known molecule makes this bond here pretty weak and so the whole thing isomerizes into a more stable form.
Take cases where you have two chemically similar molecules called "tautomers." It's hard to know which one is more stable and they often interconvert. Even good chemists do some work and realize that the isomer they want will always convert to the other form.
Bonds rotate. Bonds vibrate. A lot of chemistry happens inside a molecule itself--a hydrogen moves or a double bond changes orientation. Yuck!
Anyway, that's a few things that can happen chemically that can stop you from making an arbitrary molecule. That's why good chemists have great job security.
(b) Ah... Certainly you're correct that a lot of reactions just have crummy yield and require a lot of purification. So if we had this magical nanoassembler, new products could be made. And we'd probably learn more mechanisms--our current techniques don't reveal much on this length scale and mechanistic chemistry can have some "hand waving" explanations for some reactions.
But there are plenty of cases I've had myself, when a supposedly perfect reaction should work and simply does not produce the product you expect (if it produces anything besides tar). Talk to a synthetic chemist and you'll find that they sometimes run the same reaction a few times with different results.
Put simply, some bonds just don't seem to be possible. And we don't know why.
You back off of some claims I've heard from Drexler's camp when you say "wide variety of products." See, we probably can make a "wide variety" of molecules--we already do that in lab! But a "wide variety" is a far cry from being able to design an arbitrary diamondoid nanogear.
I honestly think that after a few hundred years of synthetic chemistry, the fraction of things that we can make, relative to the molecules I could draw on paper is still quite small.
(Full disclosure, I would describe my research as nanoscience and I think the field has lots of promise.)
Oh sure, we can make things atom by atom.
The problem is that the facility you mention (and the particular equipment you mention) is not a nano-assembler in Drexler's sense. It is not *itself* nano-scale.
See, if you use an AFM or something macroscale to make things atom-by-atom, you're largely limited to making things one at a time.
If you have a nanoscale assembler, you could have 10^23 of them and then you can make something in a reasonable amount of time (e.g. microsecons) and make a reasonable number too (e.g. millions).
*That* is the key problem.
We can make macroscopic objects out of complex components already.
But the problem is this--Drexler's theory is that we can make an arbitrary object. That's not necessarily true from biochemistry. There are a great diversity of molecules made by nature. But synthetic chemistry has been able to make molecules never made by nature.
Does that mean we can use biochemical techniques to assemble macroscopic assemblies? No.
The trick that life uses is called "self assembly." We haven't the least clue how proteins form 3D shapes from their constituents. It's a great unsolved problem in biology and chemistry. The first one to solve it wins at least ONE Nobel prize.
From current research, we know that we cannot self-assemble every molecule we can imagine. Some will self-assemble and some different types of assemblies are possible. But we're still a *long* way from being able to assemble an abitrary combination--which Drexler requires.
And if you resort to what life can do, we're quite limited. Has life ever made a skyscraper?
The notion that mechanosynthesis will not work seems to contradict current chemical methods where chemical reactions occur by random interactions between atoms/molecules. If these aren't random mechanical interactions then I must misunderstand chemistry.
Oh sorry, I forgot to rebut your point about "random mechanical interactions." If you take mechanism classes, you'll find out that 95% or more of chemical reactions are quite intricate. When you're working on 10^23 molecules, you can win--some of those will line up correctly and go through. Some won't.
Drexler works on the scale of *ONE* assembler at a time. What does it do if it doesn't have the orbitals in the "tip" aligned correctly? How does it make sure that pathway goes through in a given time?
I'm sorry, that's irrelevant. That's like saying someone told a false statement because he's a Republican. (Take your pick on insult you'd like to throw.)
I could be a bum on the street and still tell you the correct science. You might not believe me, but it's still correct.
As for Mr. Drexler, I've read Nanosystems. Mr. Drexler doesn't know chemistry. If he did, he could tell me all the cool new reactions we need to create the stuff he proposes. Or the chemistry/physics needed to do a nanoassembler.
I've done plenty of computational chemistry research--it's about 90% of my Ph.D. And you know what? I can happily draw whatever molecule I want on the screen and predict the properties. Can I make it?
NO, not necessarily!
There's a reason a lot of people hate orgo class in college. Chemistry is tough--there are a lot of exceptions and the best synthetic chemists have years upon years of experience in lab bumping their heads against walls trying to make things.
Drexler needs to try some synthetic chemistry. Maybe then he'll rethink his nanoassembler idea.
Nobody knows. The assumption is that this is possible--after all, we're not always made of the same atoms. If I ingest some isotope-labeled food and I see that some of the nitrogen ends up in my brain, then a few weeks later it's gone. Our bodies recycle proteins, carbohydrates and fats on a continual basis. We have to be more than just the sum of our quantum states. But how can we be sure what's important and what's not? Do you want to run that experiment?
Besides Ontrack, there's also Drive Savers, which has an excellent reputation. We've used them here and had excellent service.
Of course it's pricey. Much better to have a very, very solid backup system.
-Geoff
She's in the "Help" commercial trailer on the website.
I don't think it's a great turnoff for me--assuming there's some explanation for it in the plot (e.g. Smith finds Oracle and she has a "rebirth.")
-Geoff
No, no scanning. RFID tags.
Do you remember that old IBM ad where someone looks like they're shoplifting--stuffing things in his coat, then he walks out and the security guard gives him his receipt?
We already have RFID tags to prevent shoplifting CDs and stuff. Now people want to print RFID circuits onto everything and have the supermarket line just scan everything by radio wave instantly.
The chemistry for printed electronics exists and circuit designs have been made. What remains (as the parent poster indicated) is the cost. Walmart bailed on RFIDs because they're still too expensive. Everyone wants them under $0.01.
-Geoff
Actually, it's relatively easy to build in some sensor tech into your RFID circuits. Many people want printed RFID tags using organic conductive molecules and there are already shipping "electronic nose" devices for detecting all sorts of airborne molecules with the same compounds.
Add in a mass sensor to realize your paint cans are half-empty, one to determine that there's pigment in the "white" paint, etc.
In *theory* at least you can do all this with the same circuits. It's chemically possible, the electrical engineering and the physics has been done, etc. Now if someone actually makes a cheap enough RFID that can return this sort of information anytime soon is someone else's thing.
Dammit Jim, I'm a scientist, not an economist!
-Geoff
Sounds like a great idea, doesn't it? Then you realize how expensive it is to make your RFID circuits keep all those bits around. Printed electronics will probably bring RFID cheaply enough to be useful, but then you have the cost and area constraints on the number of bits.
A recent research paper showcased a 40-bit adder via printed electronics. And you want how many bits for your tags? Remember each of these RFIDs should cost $0.01 to print.
-Geoff
A lot of people would like to use "printed electronics" for RFID tags. (Organic conductive materials are often quite soluble and can be printed in inkjet printers or probably even offset presses.)
The catch is that it's still much harder to make a bit with printed electronics than with silicon, plus adding bits requires adding more addressing lines, control circuits, etc. So while those of us in the chemistry lab are making single transistors, the electrical engineers are showcasing things like 40-bit adders from printed electronics. And that's a pretty big circuit area-wise.
If you want RFID tags to take off, you want them for like 1-cent each, from what I've heard in seminars recently. Plus there's that area restriction--you don't want it to be the size of your entire cereal box.
So my guess is the 96-bits vs. 128 comes from the current printed electronics research.
-Geoff
That would be my reading of that announcement too. Of course it's the initial proposal, so it'll get all sorts of changes along the way.
No, it's not necessarily a bad thing. There's a whole lot of work that's completely hidden from view that would be opened up to academic research. I can think of several chemistry programs I'd love to get in source form.
But it would be quite interesting to see how they decide to make the cutoff. TCP/IP was government-funded research. Does that mean anything that uses it must be released? (This is why IANAL.)
-Geoff
But I'm not sure I agree there are "excessive profits" at journals, especially since some of them have recently spent big $$ to digitize and archive old articles--in many cases dating back over a hundred years. But since many of us are almost exclusively using online access to journals, distribution charges will decrease dramatically.
So the big question isn't whether this should happen--it should. Science ideally should be a meritocracy of ideas, not dependent on how much your school is willing to spend on journals. But the big question is who pays in this new model. Someone has to review and edit articles. Someone has to pay for the bandwidth of the journal. So do we go back to the "you pay if you want to publish" method (bad idea--only the rich can publish) or will public funds go to public distribution (i.e. the public library model).
Too bad public libraries are often underfunded.
-Geoff
There was even just a review of it not so long ago: Nanotechnology Review. It's definitely aimed at the "popular science" realm. Cheers, -Geoff