OT: DARPA reinvents the wheel.

I don't know if they use contractors or what, but I know they make ridiculous toys instead of weapons. A mechanical mule that carries equipment and makes a noise the enemy can hear a mile away. A pistol "extender" that shoots around corners but can't possibly do it accurately, like shooting with the wrist turned 90 degrees. Not to mention that if you carry that you can't carry a rifle, and what moron is going to choose that over a rifle?

Compared to what Bell Labs, PARC, and IBM labs did, DARPA is nothing. Like the difference between what big pharma does and what the government would do if they ever take over that business.

Speaking as a 21-year veteran of IBM Watson Research, I beg to differ. Among funding agencies, DARPA gets the best bang-for-the-buck of any Federal agency I know of.

I've had the privilege of working on three or four DARPA programs, and I've found their folks to be uniformly of very high technical quality and also easy to work with, provided you don't try to snow them.

Since I'm not a US citizen, my involvement hasn't included the super hush-hush stuff, but as a US taxpayer I wish other agencies were as good as DARPA overall.

Cheers

Phil Hobbs

But they do make some absurd toys that no soldier would want.

My tax dollars at work. Here's the DARPA TV video: The reinvented wheel starts at 2:20. Lot of other strange and interesting devices are shown.

Of course they do. Their purpose is "advanced research" - they come up with new ideas, new possibilities, new technologies. Some will be directly useful, some will be impractical in itself but lead later to useful stuff, some will be of interest in completely different fields, and some will turn out to be dead-end ideas - expensive toys.

This thing must have hydraulics in the wheels, so changes of the whole assembly is not likely to be so easy.

Rick C.

That's not how it's done. They advertise for proposals in a multitude of ve ry broad areas and the various charlatans come a runnin. Then the bureaucra ts at DARPA superficially vet the responses from which point more detailed discussions and firmed up proposals emerge. DARPA usually hands off that pa rt to other "resources" who don't work directly for DARPA. The track record with Future Combat Systems (FCS) is not so good. Seems they allow the futu ristic stuff to self-bloat to around the $100B mark of future obligations b efore the plug is summarily pulled on it and everything gets relegated to t he useless curiosity museum.

It would be nice if there were a executive summary about how the mechanicals work they probably aren't telling though.

At least at a superficial level without knowing details whether the system is electronic, hydraulic, or some combination the immediate thought is something like "Jesus they put hydraulics in the tires of something that could be getting shot at?!" but again as DARPA engineers I would imagine they would do their best to account for the conditions combat vehicles are often exposed to

Agreed. However, DARPA has an interesting method of remaining up to date and innovative: ...DARPA doesn't like to keep its employees around for long. It has about 220 employees in six offices, most of whom hold their jobs for four or five years. The agency has a 25% annual turnover rate, according to the report. The typical turnover rate in most industries is about 15%.

I guess(tm) DARPA doesn't like serial inventors, who usually produce a continuous series of amazing inventions throughout their career.

DARPA employees have their "expiration date" printed prominently on their ID badges, giving them a reminder that time is limited.

Sigh, CNN at its best. All Federal CAC (Common Access Card) ID's have an expiration date printed on them. It doesn't mean that the employee or job expires, just the card:

There are exceptions, usually political.

The usual exceptions to the rules are by those who make the rules. I guess that doesn't really apply to the past DARPA directors: One served for 8 years. All the others served for 5 or fewer years.

Incidentally, there seems to be some creativity in counting DARPA employees. DARPA comprises approximately 220 government employees in six technical offices, including nearly 100 program managers, who together oversee about 250 research and development programs.

Perhaps program managers are not considered employees? The top of the Linkedin page says "See all 467 employees". My calculators says 220 + 100 = 320 employees. Google search says 240 employees. Crunchbase says 251-500 employees. CNN says 220 employees.

DARPA program managers IME are usually senior academics who take a leave of absence from their schools for four or five years, just enough time to run one or two programs. You want to talk with the ones who've just got there, because they haven't committed to anything much yet.

It might be different in the big-bucks military hardware offices--I've only worked on MTO programs for them myself.

Cheers

Phil Hobbs

I think by government employee they mean permanent civil service. The others are some kind of non-civil service temporary appointment types, with no limitations on pay as would be the case if hired under the GS system, which most of them don't want anyway.

In training, no soldier wants an entrenching tool. In combat, they treasure the things... If it looks stupid, but it works, then it's not stupid.

It's not as though the impracticality is never predictable.

Sometimes it is, sometimes it is not.

I am sure there is a proportion of DARPA budget and work that goes to things that have a very low likelihood of ever being useful, and that some of that is due to corruption, incompetence, or simply that there is budget money that needs to be spent or budgets will be reduced for the next year. But I would expect that for the most part, projects are done with the expectation that the expected cost-benefit ratios make sense.