Slashdot Mirror

There's a DARPA BAA on the street right now for "Machine Common Sense" that is hoping to address this by asking AI researchers to design AI to learn "common sense" the same way human babies do. One of the examples in the text of the BAA is "I saw the Grand Canyon flying to New York." A context-aware AI or one with "common sense" would understand that this sentence really meant "...WHILE flying to New York" rather than inferring that the Grand Canyon was flying.

"Common sense" in DARPA's context is not really what I would call the widespread understanding of what that phrase means, but is more oriented around understanding basic physics and behaviors and recognizing when something doesn't make sense. They're also taking...um....baby steps with this BAA, just trying to get some basic behavior around recognizing un-physical scenarios and that sort of thing. It's pretty cool though.

Read about the BAA here. . Download the ~1.9MB PDF for the full text of the BAA.

You have missed an even better story. Due to their stellar job at keeping our person information secure, Equifax was awarded a no bid contract for personal identification for the IRS. Oh man, now we are really f*%ked.

Seventeen contracts are linked to in the Intercept article. Docker's is one of them.

Docker: Open platform to build, ship, and run distributed applications. Contract is here: https://www.fbo.gov/spg/TREAS/BPD/DP/SS-CFPB-16-032/listing.html

"The Bureau of the Fiscal Service (Fiscal Service), on behalf of the Consumer Financial Protection Bureau (CFPB), intends to contract with Docker Inc., 144 Townsend Street, San Francisco, CA 94107, on a sole source basis, under authority of FAR 6.302-1, for one (1) Node License including Docker Engine, Docker Trusted Registry, Docker Universal Control Plane & Business Day Support 6AM - 6PM PT, Monday - Friday, one (1) year subscription. Docker Inc. has been identified as the only source capable of providing the required content in an arrangement that suits the needs of the CFPB."

Yep, the Intercept just called out Docker for being the sole source of Docker.

You can read about the specifics here: https://www.fbo.gov/index?s=op....

The call is for a human-deployable system after 4 years. It should read from a million neurons and be able to write to 100,000 neurons, 1000 neurons in full duplex read-write, with 60 dB channel isolation, all in a tiny package that doesn't significantly overheat the brain tissue its up against.

Who thinks that's possible?

Here's the Bureau of Prisons solicitation for such a system.

Which is here:

https://www.fbo.gov/index?s=op...

Used to be that folks had the courtesy to do that, even though of course we understand that you want to drive traffic to your blog.

Re: other comments on shotgunning the things, (sounds like fun, but might do more harm than good if they fall in the wrong place), what they actually want is to disable the active control of the buggers:

"The Air Force Global Strike Command is requesting three (3) systems to counter unmanned aircraft systems (UAS), also commonly known as personal drones. There are three main areas in counter UAS (cUAS): detect, identify and defeat. This system should address the defeat portion. This portion needs to disrupt the control link between a commercial UAS and the pilot causing the UAS to fall into its preprogrammed "lost link" protocol. The system should provide the additional ability to disrupt the UAS's ability to receive and use satellite navigation signals (GPS and GLONASS) for navigation purposes."

Urm, maybe the latter might cause some hilarity if you're using it around an (air)port...

Also, loved this part:

"The system must have the below attributes:
      Low complexity: no software, no firmware"

So they want this hard-wired from transistors, or better-still valves?
HAM radio boys everywhere, fire up your breadboards!

A happy and peaceful Christmas to all...

Must be able to disrupt communications on 2.4 and 5.8 GHz ISM bands, Stop autonomous waypoint flights

Why are these two clauses ganged in one line item? Why does the second clause start with a capital. This indicates they were two separate line items and were edited to be a a single. Why? Nothing wrong with it other than being oddly written.

Disrupt ISM. Check.

Stop autonomous waypoint flights. ?. To do that you have to disrupt GPS. Why not just say that? Well, they do in the very next line item.

Disrupt satellite navigation on GPS L1 and GLONAS L1

I don't think a lot of thought went into this RFI.

At the moment, they are doing exactly what you say they should do: request for information. If after this they decide that they need something completely home-grown, then we can criticize. Right now they are just seeing what is available.

It's a bit late for these businesses, but one of the pro's of Free and Open Source software is that you always have the right to get the source code and pay somebody else to support your operating system version when the official supplier pulls their support. That's something that Microoft makes very clear is illegal for Windows users to do.

https://www.fbo.gov/index?s=op...

Well that does throw something into this doesn't it?

Looking at the solictation for bids it's clear that someone at the SSA decided to architect the system before the contractor was selected. This was back in 2009, two years before the contractor was selected.

I can imagine how the RFP cycle went - try to guess what design they had in mind, put together a proposal to do that whether it was a good architecture or not.

The proposed new Disability Case Processing System will be implemented using Service Oriented Architecture (SOA) principles and methods, and will incorporate services, which will be available for use Enterprise wide...The new system is also expected to integrate seamlessly with existing software applications in use by the State DDS's and other agency components as well as access historical data where and when necessary. Further, it must also comply with the SSA'S enterprise architecture and database standards and strategy.

Then spend the next three years trying to do what the proposal says.

To the people who modded this off topic, It seems that the IRS *REALLY DOES* use NetAPP storage controllers.

Here's a publicly disclosed bidding process document on the matter.

https://www.fbo.gov/index?s=op...

It's dated at last year. I can't imagine that they would just scrap that hardware in under a year's time.

Because the fly on the wall is still too noticeable. We need the flea on the wall.

If anybody is going to fulfill DARPA's request for hybrid electronic/biological cyborg surveillance insects, they'll need some pretty tiny electronic gear....

For example, the Navy uses 2190 mineral oil to cool the 23699 synthetic oil in the LM2500 gas turbines that move most of the fleet. The 2190 is also used to lube the main reduction gear that steps down the RPM of that LM2500 by a ratio of ratio of 21.3746 to 1 (ISTR it was 27:1 on the old Ticonderoga-class, but this is a different drive train).
The 2190 mineral oil system has a heat exchanger, trading all that lovely hotness with seawater.
The rationale for using 2190 to cool the high-performance 23699 is that, in case of a heat exchanger failure, a bit of mineral oil in the synthetic (for which the engineers test repeatedly throughout the day) is a lot less damaging than getting seawater in there.

"Acoustic zoom" beam-forming microphones have been available since 2010. Their main commercial product is an "acoustic camera" with 128 microphones on an 0.4 meter disk. They have other surveillance products, but they are "not approved for unlicensed users". This is already in use at FCI Otisville, a US prison. "This technology allows an operator to listen to various locations within the range of the system without any movement of the equipment. ... (T)his ability means the operator does not have to move about in order to "point" the equipment at his target and thereby draw attention to him potentially compromising the investigation."

With these systems, if you have enough recording bandwidth, you can record all the microphones and do the beam-forming later. So it's possible to pick the target at playback time. Squarehead is partnering with Galleon Embedded Computing, which makes 8 terabyte recorders full of flash devices capable of recording at gigabit Ethernet rates, so that's presumably what they're doing.

There are several other vendors now. This isn't really that hard to do.

So this technology is already out there, listening to crowds and pulling out single conversations.

Congrats! You are one of the few, the proud, the readers of the FA.

Here's an article

and the government program

https://www.fbo.gov/index?s=opportunity&mode=form&tab=core&id=974186e6ff5bfc7bfed500f5d51af352&_cview=0

Look at the Star Trek cosplay, not the firearms we're stocking up on!

The Internal Revenue Service (IRS) intends to purchase sixty Remington Model 870 Police RAMAC #24587 12 gauge pump-action shotguns for the Criminal Investigation Division. The Remington parkerized shotguns, with fourteen inch barrel, modified choke, Wilson Combat Ghost Ring rear sight and XS4 Contour Bead front sight, Knoxx Reduced Recoil Adjustable Stock, and Speedfeed ribbed black forend, are designated as the only shotguns authorized for IRS duty based on compatibility with IRS existing shotgun inventory, certified armorer and combat training and protocol, maintenance, and parts.

(not that it's much compared to the DHS)

>It would make the US troops seem more human

>That is a good point, but it's the only thing better about using animals.

I disagree. I believe in most situations it's best to make US soldiers seem, to the extent possible, like technodemons summoned from the cauldrons of American science wizards.

The less known and the more presumed about a US soldier's abilities the easier it will be to fight. Give them night vision and guns that can shoot around corners. Give them air conditioned self supporting strength enhancing armor. Give them networks that let every soldier know where every other soldier in his squad is. Give them flying death robots and laser guns. Emulate every desired superpower you can from ESP to precognizance to golems and X-ray vision. Make every US soldier a nightmare for his enemies, inflating his capabilities beyond any enemy simulation and in the end we'll save money by not having to actually use these abilities.

This is actually a DARPA help wanted ad. And from description of the project sounds like a good job opportunity for some slashdoters.

here is the ad:
http://www.darpa.mil/NewsEvents/Releases/2012/12/14.aspx

and here is the proposers' day conference:
https://www.fbo.gov/index?s=opportunity&mode=form&id=e21984e31d49c3780966a53983daa4f6&tab=core&tabmode=list&=

You mean that Social Security Administration ammunition story last week?

No.

That's a drop in the bucket.

Take a look for yourself at what DHS has/is contracted/taking bids for. Search results shown covers 12-'08 to 09-'12.

https://www.fbo.gov/index?s=agency&mode=form&tab=notices&id=44bca752df5e7239418a48165240f419

Of course, you can use the search function to narrow the search to just .40 caliber HP or widen it to include terms like "tactical", etc to see even more stuff that's being bought for non-military domestic use in unprecedented quantities for not being in a major world war, especially regarding the DHS & TSA, and other domestic governmental departments.

You can figuratively stick your fingers in your ears if you want if it makes you feel better, but it's apparent that, even leaving out speculative etc contract proposals and requests for quotes, at least it's apparent the government expects something big domestically is going to happen sometime relatively soon.

None of the possibilities can be good for US citizens.

Strat

That's UP TO 750 million on TOP of the 450 million already ordered.

Nope, don't know where the 750M comes from, since it says right in the contract that the max is 70M rounds per year, and it's a current year +4(IE 5 year) contract. 70M*5=350M. Page 25 of their horrible scanned contract. "The maximum limit for the resultant award(s) is 70 million rounds per year. If more than one (1) contract is awarded, the maximum limit will be split between all awards".

As for the 450M, their 'source' says that that is ANOTHER 'up to' contract. "The order comes under an indefinite-delivery/indefinite-quantity contract from the U.S. Department of Homeland Security, U.S. Immigration and Customs Enforcement for HST bullets."

Here's what I'm seeing: The DHS is ordering 'up to' 350M rounds of training ammunition; which would be a 'incredibly unlikely doomsday scenario' level. They MIGHT buy 70M one year and have a good amount of stock for subsequent years if they get a windfall of money. It's what happens in other departments.

The 'up to 450M' rounds order is for OPERATIONAL ammunition, and again, is a 5 year order.

This is not for training rounds.

Page 1-2: "This contract is for commercial leaded training ammunition (CLTA) in accordance with this document"
I already addressed how many rounds you can go through to train a single person, such that the quantities requested are actually pretty reasonable. Unless you think that 500 rounds a year, split between rifle and pistol, is unreasonably high? You can go through 60 rounds just sighting in a rifle - 10 groups of 3 for iron sights, 10 groups for scope - Sight picture training makes it practice as well, which is why you don't just have like 3 sets of 3. You have to assume that the firer doesn't fire anytime other than training/qualification, so you have to work the basics a lot.

As you said, "most". All it takes is one person who wants bad things to happen.

The military have bigger concerns while performing their duties, they shouldn't have to worry about someone intentionally or accidentally leaking information which may lead to their demise.

I've known families of deployed sailors. For the most part, they aren't told when their family is coming home until the last minute. They usually can't even say where they are deployed. It really only takes one message being intercepted.

Say you're a sailor on a ship, and you send a note home to mom (email, physical letter, IM, etc) saying "We're heading over to the Persian Gulf. We'll be there on July 1st. They already told us there's something big happening, so I won't be able to write for a few weeks.."

Then mom posts it on Facebook. " :( Little BitZ wrote. Most of you know he's on the USS SomeCarrier. They're doing something big in the Persian gulf starting July 1st."

Oh, and did I mention that mom isn't that great with Facebook, and everything she writes is public? Nah, that'd never happen. And no intelligence agency would monitor public posts for information.

Robots are starting to work in unstructured situations. I was there at the moment when this was recognized - the 2005 DARPA Grand Challenge at the California Motor Speedway in Fontana, CA. That's when everything changed.

The 2004 Grand Challenge, remember, was a pathetic joke. No vehicle got further than 7 miles, and that was CMU's. The CMU approach at the time wasn't even really autonomous. Entrants got the route on a CD an hour or so before the start. CMU had imagery of the whole area and tried to plan obstacle avoidance manually just before the start, using a huge team of people in a semitrailer full of workstations. Didn't work; the DoD people in charge had moved some obstacles during the night. And that was the best result. One vehicle came out of the gate, turned hard, and ran back into the starting gate. One flipped over. The big Oskosh entry demolished a SUV parked as an obstacle to be avoided. The whole thing was embarrassing.

DARPA was very displeased with the performance by the universities that had long been receiving DARPA funding for robotics. It was quietly made clear to some major CS departments that their performance had to improve or funding would be cut off. That's why entire CS departments were suddenly devoted to the DARPA Grand Challenge in 2005.

In 2005, things were completely different. Everybody who got that far had already been through an elimination, and every vehicle at the 2005 challenge was better than any of the 2004 entries. There was considerable press coverage, and at first, the press treated it as a joke. But suddenly there were over 20 vehicles running around autonomously, and they weren't crashing into stuff. When multiple vehicles finished the course, it was viewed as a triumph.

Finally, the state of the art had reached the point that money and determination would get problems solved. That wasn't true in the 1980s. NASA threw over $100 million at the Flight Telerobotic Servicer project, and got nothing that worked.

Now check out the DARPA Humanoid Challenge. (There's much dreck about this on blogs and in the popular press. Read the DARPA announcement instead.) They have an approach that's likely to work, and demand simulated demos (in their simulator) in 9 months, with demos on real hardware in 18 months. I personally think they'll get something able to do most of the mobility tasks and some of the manipulation tasks in that time. Useful humanoid robots will be a lot closer in two years.

Price will still be a problem. But not an unsolveable one. These things could be brought down to the price of an SUV, if not lower, through production economies alone. The parts count is probably lower than that for an SUV.

Your 'logic' assumes all humans drive exactly the same (we don't), and that all AI cars were a distinct departure requiring a vastly different protocol for working amongst (they don't).

As far as I can tell, from the perspective of a fellow motorist, the only two differences between this car and any other vehicle on the road are the drivers' hands not being on the wheel (which I see every day anyway) and the additional sensors (which might be a little distracting but no moreso than a normal car with anything on a roof-rack.) Are there other factors I'm missing? Is there something a Google'mobile does that the average sedan cannot? Do the brake lights and turn signals operate differently? Does it read and react to street signs and stop lights differently than a human?

End result: really, we'll just have to wait and see. But while some people express concern, distrust and general FUD, I approach this concept from a perspective of optimism. Even if it doesn't work, at least someone is trying something crazy and out there. A.I. cars were a DARPA pipedream less than a decade back. Welcome to the future, bitches. If all DARPA Challenges hit viable solutions this quickly, we'd have freaking androids before my kids graduate High School

"Fricking SNOW" research is getting hot for the USA. They want a new ring of sensors and unmanned systems way up north.
http://www.networkworld.com/community/blog/darpa-wants-ultimate-technology-sensor-network-monitor-vast-arctic
Assured Arctic Awareness
https://www.fbo.gov/index?s=opportunity&mode=form&id=6d24d0650fc1d4fa7c62a83bd41dff20&tab=core&_cview=0

Well, we may actually see some progress. Supposedly Sandia had a demo day in November where they invited "Potential licensees and commercialization partners"

The fbo synopsis states,
"This is a synopsis of proposed contract action FA4452-12-R-0013... The Government plans to issue a Request for Proposal (RFP) for a minimum of 63 and a maximum of 18,000, iPad 2, Brand Name or Equal devices. This equipment will be utilized as Electronic Flight Bags (EFB) for flightcrew members and trainers..." at the fbo.gov solicitation page, for what it's worth.

I think you're a little brainwashed by the MSM and the POTUS rhetoric. Yes, there is far too much collaboration among the CEO / Washington Administrator club (same guys, moving back and forth), but don't forget that only the government has the monopoly on force. And, yep, there are armed bureaucrats all over the place.

I'm not really worried about corporate power until they have government backing.

Once Senator Palpatine's bill is passed they'll be able to revoke your citizenship and throw you in Gitmo (or a domestic camp) as well.

So you can now be detained at anytime for any reason - so long as the "terrorist" label is applied.

Meanwhile, the feds have been taking bids from companies who can setup and run large scale "relocation" camps in 72 hours or less.

The bid is right here: FedBizOps.gov

More info here.

PDF with full details available at:
https://www.fbo.gov/spg/ODA/DARPA/CMO/DARPA-BAA-12-11/listing.html

Press release at:
http://www.darpa.mil/NewsEvents/Releases/2011/12/05.aspx

Um... the linked blog post links to an official document that was first posted in March of this year.

In the middle of the greatest deficit it's good to see our government spending money on games.

Serious games are actually useful and they can save not only money but lives. One area of serious gaming are training simulators. Think beyond flight simulators. They are serious games that teach soldiers how to interact with members of a very different culture. There are serious games that present fire fighters with different types of chemical spills to see how they handle it and react to unfolding events. This particular game also has a very serious and seemingly worthwhile goal:

"The goal of the Sirius Program is to create experimental Serious Games to train participants and measure their proficiency in recognizing and mitigating the cognitive biases that commonly affect all types of intelligence analysis. The research objective is to experimentally manipulate variables in Serious Games and to determine whether and how such variables might enable player-participant recognition and persistent mitigation of cognitive biases. The Program will provide a basis for experimental repeatability and independent validation of effects, and identify critical elements of design for effective analytic training in Serious Games. The cognitive biases of interest that will be examined include: (1) Confirmation Bias, (2) Fundamental Attribution Error, (3) Bias Blind Spot, (4) Anchoring Bias, (5) Representativeness Bias, and (6) Projection Bias."
https://www.fbo.gov/index?s=opportunity&mode=form&tab=core&id=1793ab48906acabaf923c76486c29c0f&_cview=0

FedBizOpps: https://www.fbo.gov/

This bid request (fbo.gov) was sent out last year -- notice phrases such as "existing inventory," and ask yourself what's going on here?

I did a quick google search and came up with this

https://www.fbo.gov/index?s=opportunity&mode=form&tab=core&id=bd97a8f9b4c3bf8393ff69b58026ffb1&_cview=0
"The NexGenCI project seeks to significantly upgrade existing x-ray backscatter, whole-body-imaging, bulk detection portals already in use by the TSA"

/tinfoil hat...

Around about 1990, I attended a talk by Robert L. Forward at Hughes Aircraft, where he reviewed the results of a study he had completed for the USAF regarding anti-matter propulsion, primarily regarding further development of anti-matter production and storage methods. A reasonably good introduction can be found here. It was fascinating, in large part because virtually every concrete step towards reliable production of "fuel" ended with the comment "this would make a great graduate project."

I haven't looked at the details for the DARPA RFI (PDF), but I'm guessing that by this point, a lot of the grad work has been done (thesis refs, anyone?) and that for this first 1 year contract, they're looking for input on the economics of scaling up the grad work. For those seriously interested, I'd bet the 100 Year Starship Public Symposium this September in Orlando would be an excellent next step. Unfortunately, it's too late to submit papers, or respond to the RFI.

I see where to apply, a link in one of the articles:
    https://www.fbo.gov/?s=opportunity&mode=form&id=406db188e0e1935a806c143a5603eb48&tab=core&_cview=0

If slashdot allowed longer tittle I woudl have called it: "Social Semantic Desktop for Sensemaking on Threats AND OPPORTUNITIES"

We'll see if they like some variation on:
    http://groups.google.com/group/openmanufacturing/msg/2846ca1b6bee64e1
"Summary: This note is essentially about how civilians could benefit by have access to the sorts of "sensemaking" tools the intelligence community (as well as corporations) aspire to have, in order to design more joyful, secure, and healthy civilian communities (including through creating a more sustainable and resilient open manufacturing infrastructure for such communities). It outlines why the intelligence community should consider funding the creation of such FOSS "dual use" intelligence applications as a way to reduce global tensions through increased local prosperity, health, and with intrinsic mutual security."

The VA doesn't seem to want to use this system; in fact, they have an open "Request for Proposal" for the creation of an open source Electronic Health Records system.

Perhaps some people here may be interested in helping. The VA is huge; whatever they standardize on, especially if it is free, is likely to be implemented in many places.

The actual request (with more information linked) is available here

I want to be optimistic. When I chose engineering as a career, my goal was to aid humanity in colonizing space, because I could see that we've run out of terrestrial expansion room.

But TFA is Michael Cooney's Layer 8 blog. Cooney mines the Federal Business Ops website for RFIs and RFPs and then writes entire articles based on conjecture and conclusions reached by means of Boots of Springing and Striding. I've worked on programs that have received Cooney's attention and was amazed at how wrong he was on so many points, and how he presented his erroneous assumptions as facts. It's hard to take anything I read on Layer 8 credibly.

For instance, Cooney regularly glosses over the transient nature of the RFIs he cites. Keep in mind that an RFI is merely a "Request for Information." It's an unfunded solicitation of ideas and white papers, used to identify whether there's anybody credible out there who has an idea plausible enough and attractive enough to warrant going back to the DARPA Director and, eventually, Congress with a budget request for a real RFP and phase I study program. Many RFIs result in either nothing, or an RFP for an unfunded IDIQ or a shoestring SBIR type contract. They're fishing expeditions. And sometimes they're done for internal projects just to get new ideas for free, or for programs hardwired for an existing contractor just as a sort of threat. (But on the other side of the coin, DARPA is usually not tricksy like that... but there's still no guarantee of any money available.)

Still, I'm very glad that DARPA is soliciting ideas, at least... there's a phrase in the R&D world: "DARPA Hard." DARPA doesn't consider ideas that are just matters of engineering -- making existing tech lighter/faster/cheaper. They want to push the state of the art and hope to sponsor real, fundamental science that opens up new possibilities. Starships are indeed DARPA Hard.

If so, I don't think they're planning on using it much mid-mission. From the PDF on the drones (pg. 6):

Autonomy
    Independently deployable unmanned system
    Sparse shore-based remote supervisory command and control model
    Capable of autonomous risk-based mission continuation through command and control interruptions
    Adaptive constraint set based on strategic context, mission phase, internal state, and external conditions
    Compliance with the Convention on the International Regulations for Preventing Collisions at Sea (COLREGS) and maritime law
    High fidelity surface navigation and situational awareness sensors
    High fidelity internal state awareness and dynamic performance assessment

It seems like they're planning on launching these things and then having them operate completely autonomously - most likely it'll only communicate with its land-based handlers at the very beginning/end of its mission.

A interesting RFI for Pre-Packaged Commercial Meals mentioned the New Madrid Fault System
https://www.fbo.gov/index?s=opportunity&mode=form&id=eaea338540a0aea155a48a650a077352&tab=core&_cview=0 (Jan 20, 2011)
"...FEMA request for Information is to identify sources of supply for meals in support of disaster relief efforts based on a catastrophic disaster event within the New Madrid Fault System for a survivor population of 7M...."
36 months of remaining shelf life upon delivery.
http://www.reuters.com/article/2008/11/20/us-earthquake-study-idUSTRE4AJ9EV20081120 from 2008
would cause "the highest economic losses due to a natural disaster in the United States."
New Madrid Seismic Zone Catastrophic Earthquake Response Planning Project (Statistics for Eight-State Region on page 28)
https://www.ideals.illinois.edu/bitstream/handle/2142/14810/Volume%20II_Part%201.pdf

source: https://www.fbo.gov/index?s=opportunity&mode=form&id=06a877fddd2dedaf6a52520345f64eda&tab=core&_cview=0

from the fedbizops:

  "Promotion of new technologies to support declassification. Striking the critical balance between openness and secrecy is difficult but a necessary part of our democratic form of government. Striking this balance becomes more difficult as the volume and complexity of the information increases. Improving the capability of departments and agencies to identify still-sensitive information and to make declassified information available to the public are integral parts of the classification system."

Right. If you actually read the announcement, it's not that they want yet more boondoggle supercomputing centers. What they want is more crunch power in small boxes. Read the actual announcement (PDF). See page 17. What they want is 1 petaflop (peak) in one rack, including cooling gear. The rack gets to draw up to 57 kilowatts (!).

From TFA, written by Michael Cooney and propagated by the summary:

Dubbed extreme scale computing, such machines are needed DARPA says to "meet the relentlessly increasing demands for greater performance, higher energy efficiency, ease of programmability, system dependability and security."

It looks like these "extreme scale computing" systems are needed before things like "ease of programmability" can be acheived. I call bullshit.

The actual notice from DARPA is named Omnipresent High Performance Computing (OHPC). From the first paragraph of that page:

... To meet the relentlessly increasing demands for greater performance, higher energy efficiency, ease of programmability, system dependability, and security, revolutionary new research, development, and design will be essential to enable new generations of advanced DoD computing system capabilities and new classes of computer applications. Current evolutionary approaches to progress in computer designs are inadequate. ...

That makes a lot more sense.

Now, will someone please go and smack Michael Cooney up the back of head for writing like that?

I trust Google.
Ten years from now when the founders leave, I may not.
Its innovative to give Google 5years, no more, no less to do it.
They will do a fantastic job, but it should go out to bid, shouldn't it?


Not all feds are good, and not all feds are bad. I didn't always believe it, but I now do. [but thats a POLL I imagine...]

Heres something the feds won't do as well as we are.

The workshop will: (a) Introduce the research community (industry, academia, and Government) to the TX program vision and objectives;

So who wants to start a pool on which agencies/industry power hitters make the biggest contributions? Lockheed Martin has a great military aircraft record but Boeing seems to work magic in the advanced controls systems. Personally I would put my money on Northrop Gruman or some university coming up with the most significant design contributions. Both of those sources have quite the tenacity for half-crazed cutting edge ideas that the government loves to gobble up.

Did we read the proposal linked from the article? It specifies PlayStation 3 model CECHP01 which does, indeed, run Linux. I wondered, though, how successful they are going to be at finding 2,200 units. Distributors are running out of new/old stock of this model, as many compute cluster builders are trying to get them before they're all gone.

Here is the proposal for those who didn't actually bother to RFA:
https://www.fbo.gov/index?s=opportunity&mode=form&id=bac60f8808fa1e221597573901a7cd6b&tab=core&_cview=1&cck=1&au=&ck=

From reading the actual BAA, it sounds like this is not an effort to replace IP networks but to supplement them with additional protocols. In fact, the requirements explicitly state that MNP must carry legacy IPv4 and IPv6 traffic.

Now the DoD is going to pay Microsoft to develop a new networking protocol for them (http://defensenews-updates.blogspot.com/2009/10/lockheed-martin-awarded-31m-research.html).

Wonder what licensing fees MS will charge for the "Military Networking Protocol" (https://www.fbo.gov/download/afa/afa738a71c6cbdf8024d54ec5e141a1a/2008_10_28_mnp_baa_final.pdf) or whether they'll even allow FOSS apps to license MNP?