US Army Spent $2.7 Billion On Crashing Computer 196
An anonymous reader writes "According to two former US Army intelligence officers, the multi-billion-dollar DCGS-A military computer system that was designed to help the US Army in Iraq and Afghanistan simply doesn't work. DCGS-A is meant to accrue intelligence, surveillance, and reconnaissance, and provide real-time battlefield analysis and the current location of high-value targets — but instead, it has hindered the war effort rather than helped. Major General Michael Flynn, the top intelligence officer in Afghanistan, says that DCGS-A's faults have even resulted in a loss of lives (PDF)."
high-tech armies are vulnerable (Score:5, Interesting)
T.E. Lawrence and the Mind of an Insurgent
"Lawrence distilled six fundamental principles of insurgency that even today have remarkable relevance.
First, a successful guerrilla movement must have an unassailable base - a base secure not only from direct physical assault, but from attack in other forms as well, including psychological attack.
Second, the guerrilla must have a technologically sophisticated enemy. The greater this sophistication, the greater this alien force would rely on forms of communications and logistics that must necessarily present vulnerabilities to the irregular.
Third, the enemy must be sufficiently weak in numbers so as to be unable to occupy the disputed territory in depth with a system of interlocking fortified posts.
Fourth, the guerrilla must have at least the passive support of the populace, if not its full involvement. By Lawrence's calculation, 'Rebellions can be made by 2 percent active in striking force and 98 percent passively sympathetic.'
Fifth, the irregular force must have the fundamental qualities of speed, endurance, presence and logistical independence.
Sixth, the irregular must be sufficiently advanced in weaponry to strike at the enemy's logistics and signals vulnerabilities."
http://www.findarticles.com/p/articles/mi_qa3723/is_200507/ai_n14685818
--------
In the words of Scotty, Star Trek III: "The more you overtake the plumbing the easier it is to stop up the drain."
The complexity of modern armies is their Achilles heel.
Typical... (Score:5, Interesting)
Meantime, the Republicans want to cut *every* social service, but won't cut a single dollar of "defense" spending, which is how the US Army spends more per year ($20 billion) providing Air Conditioning in Afghanistan, than NASA's entire budget.
We cannot sustain fighting three or more Wars (I've lost count), without new taxes. And since nobody wants more taxes, the wars must end. What happened to Rumsfeld promising that we'd get Iraq's Oil, and it would pay for the war???
Cripes we're in a bad situation.
Re:What's so special about this computer system? (Score:4, Interesting)
Marketing gimmick (Score:5, Interesting)
Re:Pentagon Irresponsibility (Score:5, Interesting)
The Pentagon does not write its own budget. Our military is civilian led, which means the place to point fingers is at the Senate Defense Appropriations committee: http://en.wikipedia.org/wiki/United_States_Senate_Appropriations_Subcommittee_on_Defense [wikipedia.org]
Note the list of Republicans: all of them are 00's-style big spenders, and perfect complements to their democratic counterparts. There is not a single voice on that committee for fiscal conservatism or budgetary restraint.
I agree that we need to slash and gut the military budget. We can run a better, cheaper army, but first we have to gut the appropriations committee (and the Senate Armed Forces committee). For my part, I have supported primary challengers to ever Republican on that list (to little effect). I urge democrats to do the same.
Now everyone's a freakin' expert! (Score:4, Interesting)
It's really easy to produce a system that meets the easy 80% of requirements. It's A LOT harder to complete the job. The 'lives lost' statement is a consequence of 'missed operational opportunities', where the computer is only an enabler. It still takes a human to decide to act on information (in a timely fashion.) I've met very few people who are both trained intel analysts and experienced/competent programmers or system engineers and therefore competent to pass judgement on the implementation of a large complex distributed (and hopefully fault-tolerant) system that must deal with incomplete/inconsistent information and communications problems. (But I've met a lot of military/government people writing requirements who are happy to specify things that are theoretically impossible...)
This reads like someone trying to do 'procurement via public relations,' something that was particularly blatant during the USAF Tanker recompete.
And of course the Slashdot postings are full of posturing based on political persuasion and no knowledge of the actual system or its requirements or implementation.
I'm not defending DCGS-A, I'm just pointing out observations from a career spent doing these kinds of systems in both military and non-military government contexts. I do not have any knowledge of DGCS-A requirements or implementation nor do I speak for anyone besides myself. If caught or captured, my secretary will disavow any knowledge of my actions.
Re:Typical... (Score:5, Interesting)
That was horse-shit fantasy from day one ... did you really believe that Iraq was going to pay you for the troubles of overthrowing their government, and that they'd be beholden to you and sell you cheap oil for decades?
That was one of those purely bullshit things the previous administration was prone to saying (like "Mission Accomplished" [wikipedia.org]) that was so far detached from reality as to be offensive. Oh, sure, they'll give you billions of dollars in oil to offset your costs, and they might throw in a pony as well.
I find it hard to believe that anybody actually believed that the upshot of overthrowing Iraq would be cheap oil -- unless, of course, the whole invasion really was a pretext to try to grab the oil. Mostly, it's just another example of how Bush et al had their heads up their collective asses.
Re:Typical... (Score:4, Interesting)
Democrats don't want to cut defense spending either.
Yes, because the ones that do get called traitors by the same blowhards who whine about how the government isn't fiscally responsible enough.
Re:Marketing gimmick (Score:4, Interesting)
The JUONS (PDF) linked in the article is likely pro-Palantir or pro-Something without coming out and saying it, too. They are written with "requirements" that usually only one system can fulfill. It's not necessarily malicious, though. The writer is sure they know which system they need to satisfy their own requirements.
Re:What's so special about this computer system? (Score:5, Interesting)
Re:Sad for the lives lost (Score:4, Interesting)
It's terrible that people actually died as a result of shoddy programming but I am not surprised.
People were going to die either way; this is the military we're talking about. Seems better that an invader should die than someone defending his home, doesn't it?
Windows Server OS LoL!! (Score:2, Interesting)
Good Heavens...
https://l3com.taleo.net/careersection/l3_ext_us/jobdetail.ftl?job=208541&src=JB-10095 [taleo.net]
Windows OS server. Now, given the experience I have running WIndows, there is no way in _HELL_ I would use it in life or death situations.
I mean the largest application domain for windows is playing GAMES, not business and certainly not for combat operations.
These people must be complete idiots.
-Hack
Re:high-tech armies are vulnerable (Score:2, Interesting)
Well, a quick overview:
First, a successful guerrilla movement must have an unassailable base - a base secure not only from direct physical assault, but from attack in other forms as well, including psychological attack.
None of our enemies currently have this. Pakistan is getting a little restless, so the situation could change in the future, but this has not been an issue so far.
Second, the guerrilla must have a technologically sophisticated enemy. The greater this sophistication, the greater this alien force would rely on forms of communications and logistics that must necessarily present vulnerabilities to the irregular.
This is just stupid. Greater sophistication decreases vulnerability.
Third, the enemy must be sufficiently weak in numbers so as to be unable to occupy the disputed territory in depth with a system of interlocking fortified posts.
Yes, this is usually the main problem.
Fourth, the guerrilla must have at least the passive support of the populace, if not its full involvement.
Again, not really a problem in the most recent conflicts. Iraq had too many factions for any one particular group to have "passive support of the populace", and in Afghanistan the Taliban has very little support, though they have been successful at terrifying significant fractions of the populace into not opposing them.
Fifth, the irregular force must have the fundamental qualities of speed, endurance, presence and logistical independence.
This doesn't really mean much. You can rephrase it as "the irregular force must be an irregular force".
Sixth, the irregular must be sufficiently advanced in weaponry to strike at the enemy's logistics and signals vulnerabilities."
This hasn't been an issue for western nations in a long, long time. Even in Vietnam, logistics and comms weren't significantly impacted by enemy action. These days the "irregulars" we face depend solely on shifting political opinion rather than achieving military goals like disrupting communications abilities or supply chains.
Conclusion: T.E. Lawrence lived in a completely different era. In his time, "communications" still meant messengers on horseback, and maybe telegraph lines running through the desert. While a couple of his observations are still valid today, the majority are just a quaint reflection on the attitudes and tactics of ancient armies. They have as much bearing on combat today as the musings of a caveman would have had on the armies of Lawrence's time.
I used to work for them... (Score:5, Interesting)
I used to work for one of the suppliers (the one most "at fault" according to the article, with the shitty code and shitty UI we provided).
Here are some things to consider:
The company's business model was to procure IDIQ contracts...they succeeded for several years by purposefully providing broken bits and pieces, in order to assure that more fixes would be purchased later. It finally caught up to them because you can only pile so much crap on existing crap before the whole thing breaks.
Palantir is great software, but people in the Army don't like it. They think it's pretty with no functionality. They are wrong. It's awesome. There are two problems with Palantir, in that you have to store your data on THEIR servers, and the owner of the company is not a US Citizen. They have some inroads, like the links suggest, but they'll never be able to get the most sensitive contracts because of the US Persons requirement.
DCGS-A sucks because it is closed-source garbage that runs only on Microsoft components, and relies heavily on SQL-server. Plus all the people I used to work with are overpaid self-taught jackasses who got the job because they could code in visual basic and they had a clearance.
In all, I'm glad to see the Army and military in general understand and accept that they are suckers and slaves to politicians and "the free market" mentality of PACs and lobbyists. Too bad this garbage (and even bigger garbage FCS/BCTM that finally got axed last month) wasted so much money in the meantime.
Screw the free market. Time to put all this money into government R&D and churn out some decent software for the investment. The NSA alone has enough talented programmers to make this happen.