Hacker News Reader

Comment by illumin8 | Hacker News Reader

The funny thing is that if the DoD contracted out to build these, they would end up costing over $1 million each. There are a lot of unique opportunities for startups that are willing to use off the shelf technology and package it in unique ways. Provided, of course, you don't have any ethical qualms about selling technology that might be used for good or evil.

View on HN

You are off by at least an order of magnitude in your price estimates, milspec rugged robots are expensive but on the order of $30,000 - $100,000 each not $1 million. Now those are complex systems with articulated arms, complex comms, and full control station style remote. If the military spec'd out a simple system, like the one in this article, it would cost much less. An existing contractor like iRobot would be perfectly capable of producing it at a low cost. They make the existing pack bots used by the military as well as cheaper consumer grade robots like the Roomba.

I totally agree. Here's the Packbot from iRobot who make Roomba:

http://www.irobot.com/gi/ground/510_PackBot

They're slow. The full military version is $120K but simple versions are half that.

Another analogy is the Firstlook:

http://www.irobot.com/gi/ground/110_FirstLook/

(There's a nice video).

Also commercially available:

http://www.qinetiq-na.com/products-talon.htm http://www.qinetiq-na.com/products-dragon-runner.htm

All but the FirstLook, which is new, are in routine use.

The article is sensationalist by not saying that.

This is true (it would probably be more like $100k but it would still seem absurd) but the DOD would control the entire supply chain and intimately know the specs and capabilities of every part.

What if the military came to depend on these and showed up at the toy store only to find the asian manufacturer discontinued that model and replaced it with a weaker one with a remote that interfered with the wireless cameras you were using?

Some of whats bought for that absurd seeming price is expertise and most importantly repeatability.

Actually, if you don't sell it for at least 1M$, you won't get any contract, it doesn't look serious enough. Terrestrial drones were being done during WWII, Russians were remote-controlling full-sized tanks in the 40s, these techs should be obvious by now.

I have been working for a defense project about instrumenting some armored vehicles with video cameras and monitors. The project was more than 10 years old when I arrived in it. I couldn't help but thinking "bring the prototype to any hackerspace and pay 20k of salary for two weeks of dev and you'll be done."

Let me be the first to say thank god they don't.

We should all count our blessings that hackers tend to be peaceful people by nature. Imagine if we were trying to kill each other.

Well who do you think writes software that controls missiles?

I don't doubt there are some smart people who are trying to figure out how to kill each other. I'm just glad that's not most of us.

Edit: Although I guess I can extend that to human societies in general, not just hackers.

Edit 2: Man, no idea who manages to disagree with that.

I can't remember the exact quote, or even the source, but it was along the lines of "For engineers, the biggest [moral] issue with weapons development is that it is fun".

You can build some really neat things when you have a military R&D budget behind you. The downside is of course that it actively or passively hurt people.

And then there's the whole calculus-of-death/unintended consequence thing, where in theory, a more accurate missile is better for everyone because it reduces collateral damage. But because it kills less innocents, there are fewer qualms about using it more often. This applies strongly to the current UAV trend - You can politically spin (or just bury in accounting) "Ooops, We lost a $xMM drone", versus actual pilots dying.

Yup. I am sure they pitch it as you can build something that saves lives. Little Bob can ride his bicycle safely because of your work. As opposed to you can build something that can kill the enemy in mass in seconds.

There's the old joke: a decent software engineer would never write a destroyLisbon() procedure - he would write a destroyCity() procedure and pass the city as an argument.

> "For engineers, the biggest [moral] issue with weapons development is that it is fun".

This reminded me of a very similar quote from Edward Luttwak about the disconnect between students of war and the people who fight wars: http://www.youtube.com/watch?v=7vz1BpRf3Lc&feature=playe...

The entire interview is fascinating and worth the time. It's a bit dated (from 2007), but still has a number of really interesting tidbits throughout.

Yes, lets leave that to the physicists please.

It seems like the truck used in this example costs around 300 usd. If I were a soldier, I think I would not wait for the government to commission a million dollar truck. I would get 3 of these for now.

That toy truck cost 300 fucking dollars??

On the low end it seems.

That's a hobbyist RC car, meant for adults with money, not kids. I have a friend who has a similar model, and he spends time buying new parts and upgrading, like any other hobbyist. It has a gas motor and I suspect this one does as well-it goes as fast as their Humvee on the limiter. A cheap toy store model wouldn't last a week in those conditions and wouldn't hit the speed to make it effective.

At that point you'll have one. Now try to make it it work through extreme weather - hot, cold, pouring rain, etc. Now make it reliable to shock so you aren't fixing thing every 50 miles or so. Ok, write the reference manuals so that 18yr old mechanics how to install, maintain, replace them. You've got some sharp mechanics there, but you have a few dumb ones too. The manual will involve breaking things down to a painful level. Setup logistics for contracts, replace/repair. Oh now setup a production line to make hundreds or thousands. And, by the way, that production line may or may get shut down after the first order. Or you may get another order for exactly the same thing two years later after the last one rolled off the line.

It shouldn't take 10 years, but it isn't a simple trip into a hackerspace either.

Yeah, that's gonna take more than a trip to the hackerspace— maybe a Kickstarter.

Some of the things you're listing are must-haves (secure channel), but most are not. If increasing reliability and versatility makes the drone last 10 times longer and costs 10 times more -- it's not worth it in the long run: it's goal to trigger the first IED it can find, anyway. Even if it's not, it still would be cheaper and easier to buy 10 drones or make them modular enough so anybody with some X-Box experience can snap a new one together in the field. Manuals? -- nah, see the X-Box thing above. Besides, who reads RC cars' manuals? Replace/repair? -- a dumpster full of modular parts is lightyears ahead of any repair technician in the field. Production lines? -- there already exist production lines for RC cars, aren't they?

To put it in the familiar context: what you're describing (and what DoD is used to) is a mainfraime. These guys just made a PC.

I'm no military historian, but there are terrible failure stories for products pushed out before they're ready to the frontlines too. Beauacracy is a waste, but some of those steps seen as wasteful in the DoD procurement process are "scar tissue" from where things went wrong either in the battlefield or, as often, protection against incapable, or worse, fraudulent contractors.

You're right, my post wasn't a list of requirements, just an illustration of some of the things one might end up having to care about when packaging up a military product used in dangerous situations. There's a balance to be made, it should be made on a case by case basis.

some of those steps seen as wasteful in the DoD procurement process are "scar tissue" from where things went wrong either in the battlefield

This is precisely part of the mechanism that causes every military to be well prepared for the previous war but unprepared for the current one.

I propose greater emphasis actual feedback mechanisms based on experience in the field. These can be made to work on timescales of months or weeks. When bureaucracy comes into things, iteration timescales can stretch to years and change can become generational.

The path you mention exists, but not every program is fast tracked - nor should they all be. If you go too fast and mess up, soldiers can die. If you go too slow and mess up, same result.

I get frustrated too at too-slow, too-big companies running projects with billion dollar price tags resulting in no workable product, but I'm not as sure that the solutions are simple. I am sure that without constant oversight from within and without it could be much worse.

http://en.wikipedia.org/wiki/OODA_loop

> If increasing reliability and versatility makes the drone last 10 times longer and costs 10 times more -- it's not worth it in the long run: it's goal to trigger the first IED it can find, anyway.

Google's approach to building server farms was likewise to buy hardware that was ultra-cheap because it had failed to meet manufacturing specs. Google expected a certain percentage of hard disks, etc., to fail, and simply swapped them out when they did. That resulted in a significantly lower TCO. (So says Steven Levy in in his recent book In the Plex.)

While this is a legit perspective, there is a meaningful difference in that Google can throw out their failed hardware without any real risk of having it return to haunt them, but broken toy trucks means either handing materiel to the enemy or having extra useless weight to carry around.

You might be able to tweak specifics to account for this, but you also might not.

The solution is to have force redundancy and to make each individual element completely disposable (and detonatable).

(and detonatable)

No way terrorist types could misuse that one.

No, I mean if the gear can't be recovered you tell it to self-destruct.

Having lessened reliability is ok until the 2 that fail are the 2 you had in your truck as you're driving down a hostile road. Or if they won't work in the rain, that's ok until you have to go on patrol in the rain. Or if they won't work in harsh desert conditions... oh, wait.

It's similar to saying that having expendable bullets and grenades is OK until you use the last one while still on patrol. Solution: preparing for it by taking more. Knowing that they're designed to fail/be repaired easily changes the approach. The argument I'm trying to make is that it would be cheaper/lighter/more_versatile to get many expandable droids than to get one robust and universal.

Also, highly-adapted droid != highly-adaptable droid. There is no need for an amphibious droid in Afghanistan, just like one doesn't use the same apparel in all climate zones. As long as interface/principles are the same (like a PC), various versions of it can be used in (almost) any environment without extensive re-training.

While I see where that makes sense, the phrase "The perfect is the enemy of the good" comes to mind. The soldier in the field would probably rather have something like this instead of nothing at all or worse, wait (and die) while something "perfect" is developed.

On the bright side, at least they weren't forbidden from using there own solution.

Actually I was very surprised to discover that security was not taken into consideration in the main protocol used by drones operators. There has been stories about Iraq US drones sending or receiving some data unencrypted.

I am not allowed to comment about the specific project I have worked on, but I can tell you that a trip to the hackerspace would have resulted in a better product in that respect too.

I wish I were free to comment on this one too.

Agree with the first part. However, I think that as long as you were fairly firm that you were only interested in building technology that saved lives (no RC toys with guns on them), I would not feel guilty in the slightest. After all, this truck can't really be used to hurt anyone, just save lives.

It absolutely can be. The XM-7 Spider is basically the same thing, except over secured remote control, and it holds munitions instead of video cameras. It's pretty difficult to build technology that only saves lives.

"To every man is given the key to the gates of heaven; the same key opens the gates of hell." - Bhuddist Proverb, via Feynman on whether science can be good or bad

It could be used to scout targets.

Just tie some explosives to it, like the RC-XD (explosive device) in Modern Warfare 2.

You beat me to the punch, but it was actually Call of Duty Black Ops. :)

Any device designed to operate in a war zone is an implicit support of the activity. Even life-saving devices. If you made bullet-proof jackets and sold them to the US military, you are supporting the military. You may even be impacting the outcome of an engagement, or influencing the endgame.

If the truck did not exist, 6 US soldiers would likely be dead. The insurgents would probably prefer this. Because the truck existed, the soldiers may now go on to fulfill their missions, which may result in insurgents being killed.

Basically, war is an ugly thing. Don't expect to get involved without getting a little dirty.

P.S. In the interest of transparency, I am an American and a strong supporter of the US military. I have family currently serving, and I'm glad that there was a device like the truck that saved lives.

I was going to reply, but dkokelly below summed up my thoughts exactly. By saving the lives of our soldiers, there is a very real chance that they will go on to take the lives of some enemy soldiers in the future. So, doing any business with the DoD you have to think about the possibility that your products will either directly or indirectly cause the loss of human life. Whether you agree with the mission or support the troops or not, some people don't like selling products that cause the indirect loss of human life.

I think it's because this solution is pretty specific to the region. The toy truck works on sandy roads in the middle east. If the roads were muddy or large gravel the toy would get stuck. The DoD would probably have requirements for this to work in multiple terrains.

Having different solutions for different terrains would become a logistics problem. The DoD already has enough problems getting equipment like AA batteries to the correct locations in sufficent quantity.

Well, one reason is feature creep. More and more spec and features will just get tacked on (doesn't matter if it's suppose to be disposable).

1 guy w/ a rigged RC truck bears no liability. A defense contractor shipping these devices will have a team of lawyers chasing them around if it fails to find a bomb that kills someone.

Great the legal environment we have created for ourselves prevents effective tools to be available to our troops.

I guess individual Americans are going to have to step up and fill the gap since contractors may not be willing, at low cost, to assume the liability.

I'd happily pay $500 plus the shipping to Afghanistan if it helps soldiers do their jobs faster and safer. I'm sure that I am not the only one.

They'd sell for a high price not necessarily due to price gouging, but because it costs a lot to sell to the military. The overhead of quality control, support, and all of the bureaucracy involved in dealing with a military customer would dwarf the actual per-unit cost.

That sounds like an incredible waste with the overhead. I absolutely don't understand it. I think people in those power positions really need to stop treating the whole structure like a black box and realize it shouldn't be that damn hard to send some RC cars to Iraq to save lives.

The DOD does take quite a bit of overhead to work with. They also do buy some stuff at pretty competitive prices. As I remember the tent contracts were actually pretty reasonable compared to what commercial 30-man tents cost (actually less than quite a few).

// Dad used to work for place that made tents and bags for DOD

The DoD is the ultimate Waterfall Method practitioner.

Isn't that true of any technology?

Not the PSP controllers they use to control various things. The DoD gets those cheaper than we do.

Reply on news.ycombinator.com