[thorie]

I like the point he made about not giving exact instructions on taking a box from point A to B.

Looks like it's better to hire smart, capable people and not micromanage them - than it is to hire cheap people and tell them exactly what to do.

[jholman]

As hugh3 mentioned, it's not (necessarily) about hiring smart people, nor about people who arrive well-trained. The majority of entrants into the military system have no training (i.e. are 18y.o. kids), and I suspect that the mean intelligence is a tiny fraction south of that of the general population (because of the incentive structure).

From the FA, the military system is not "just trust everyone to get a job done, no matter what the job is".

From the FA, the system is: first you train people hard, then once they've passed testing you trust them to make autonomous decisions within the areas that they're trained to be competent on, and then you assess performance and as necessary retrain. AND REPEAT. (And, I suspect, on an ongoing basis you add training for new capabilities.)

I think it's a _serious_ mistake to simplify that very capable feedback cycle down to "hire clever people and get out of the way". Giving people feedback (and training as necessary) will beat leaving them hanging, hands down, every day.

(note: IANA soldier)

As for whether or not this is ideal for a tech company, it seems to me that one requisite of the train-trust-feedback cycle is that you have to be able to train and give feedback. That's incompatible when you hire people smarter or more skilled than you. And if you're trying to innovate on the bleeding edge, that complicates things hugely. So I humbly (IANA founder either) suggest that perhaps a good strategy for a tech company is to pursue this technique as far as possible, but no further.

[barry-cotter]

I suspect that the mean intelligence is a tiny fraction south of that of the general population (because of the incentive structure).

The US military is not legally allowed to take people with IQs below 80. They have dropped or relaxed requirements about tattoos, criminal records, drugs and other stuff rather than drop the minimum score required on the AFQT.

http://www.udel.edu/educ/gottfredson/reprints/1997whygmatter...

Lest IQ 80 seem an unreasonably high (i.e., exclusionary) threshold in hiring, it should be noted that the military is prohibited by law (except under a declaration of war) from enlisting recruits below that level (the 10th percentile). That law was enacted because of the extraordinarily high training costs and high rates of failure among such men during the mobilization of forces in World War II (Laurence & Ramsberger, 1991; Sticht et al., 1987; U.S. Department of the Army, 1965). Minimum enlistment standards since World War II have generally been higher than the 10th percentile, and closer to what they are today for the different ser- vices: the 16th AFQT percentile (Army, about IQ 85), 21st (Marine Corps and Air Force, IQ 88), and 27th (Navy, IQ 91).

[hugh3]

Many of the enlisted men in the military aren't all that smart, and most of 'em are pretty darn cheap.

If there's a takeaway lesson it's that even not-so-smart, cheap-to-hire people are actually surprisingly capable of doing stuff, if you create a management environment where they have to do it and know they aren't gonna get any mollycoddling from the higher-ups. Here is the task. Go do it. Come back to me when it's done.

[rdl]

The argument of elite selection vs. elite training is interesting. I think USMC, particularly in WW2 and in general during the draft period, had general-population induction, but through training and organization produced an elite force. Other elite units, like the Navy SEALs, seem to focus on elite recruitment and selection.

I'm not sure which works better for a tech company. Clearly only the USMC process scales once you need to hire a lot of people.

[lucasjung]

The SEALs do both: they are very picky about who gets into BUD/S, BUD/S weeds out all but the best, and then they continuously train to stay the best.

I actually believe that the USMC's formula for success doesn't scale without limit. They are way smaller than the Army, and I'm sure that's a big part of how they are able to maintain the kind of cohesive culture needed to sustain excellence.

Also, even during the draft era, the Marines had another tool that you didn't consider: attrition. I don't mean the combat kind, I mean washing out the people who can't cut it. It's amazing how thoroughly they can transform a young person, but even so there are some people that just aren't good enough, and the Marine Corps has always been the most willing of all services to get rid of dead weight. So it's not selection vs. training, it's a three-way combination of selection, training, and attrition.

[nhangen]

"Many of the enlisted men in the military aren't all that smart, and most of 'em are pretty darn cheap."

That's one of the most asinine statements I've ever read here.

[VladRussian]

you're forgetting that it is military. "Deliver the box to point B" is an order, and non-execution of an order isn't an option. Nothing brings out and motivates creativity in people as effectively as real necessity does.

http://www.nationaljournal.com/the-homemade-weapons-of-libya...

Edit: the precision is a "must-have" of a military order (and is the responsibility of the CO, dual to the responsibility of underlings to carry out the order) and must not be mistaken for micromanagment while absence of micromanagment must not be mistaken for vagueness, what is frequently happens in the non-military, especially corporate, environment where such duality of responsibilities is frequently not recognized/enforced.

[spitfire]

That's known as a mission type order. Or Mission Orders in marine. MCDP-1 and Maneuver warfare by William Lund are good books on the subject.