Army Unit to Intel Center: DCGS Doesn’t Work

Army Unit to Intel Center: DCGS Doesn’t Work

Before recently deploying to Afghanistan, a U.S. Army brigade warned the service’s own intelligence center that the battlefield intelligence system doesn’t work.

The 4th Infantry Division’s 4th Brigade Combat Team more than a year ago submitted a report criticizing the Army’s version of the so-called Distributed Common Ground System, or DCGS (pronounced “dee-sigs”) to the U.S. Army Intelligence Center of Excellence.

“DCGS-A does not provide the functionality needed by deployed intelligence Soldiers,” according to the document, a copy of which was obtained by Military​.com.

The report was written in late January 2013, after the brigade returned from a previous deployment to Afghanistan during which 10 of its soldiers were killed, including the command sergeant major.

Less than four months later, Lt. Gen. Mary Legere, the Army’s deputy chief of staff for intelligence, defended the intelligence system during a demonstration of the technology at Fort Belvoir, Va.

“It provides the underlying intelligence for every decision that our commanders and soldiers make in the field and it saves lives,” she said at the time.

The report, prepared by Gabriel Martinez and Jeremy Godfrey, who were identified as collectors of lessons learned for the intelligence center, contained many of the same criticisms that other units in Afghanistan would later echo, namely that the technology is complicated to use and unreliable.

For example, the system’s so-called multi-function work station “did not assist unit analysts with intelligence fusion” and was “prone to crashing,” its tactical entity database “was unstable and labor intensive,” and because of the bandwidth requirement, “several battalion level organizations were unable to use the system effectively,” the document states.

It concluded, “At this time, DCGS-A is not an enterprise system capable of replacing the Army’s multiple intelligence ground processing systems.”

The report is the latest internal assessment that’s likely to fuel debate on Capitol Hill. Rep. Duncan Hunter, R-Calif., has argued with Army Chief of Staff Gen. Raymond Odierno over why the Army remains committed to the acquisition program at the exclusion of potential commercial alternatives.

Foreign Policy last week reported that the Pentagon’s acquisition arm is withholding a report that concluded the Army was wrong to dismiss software made by Palo Alto, Calif.-based Palantir Technologies Inc. The software, it reportedly states, can meet many of the service’s intelligence needs.

The Army brigade returned to Afghanistan earlier this month with the existing intelligence system.

Soldiers with the 4th Infantry Division and its subordinate units in Afghanistan, as well as those in Kuwait and at Fort Carson, Colo., are using “every tool” available for intelligence collection, including the Distributed Common Ground System — Army, which is their primary resource, according to Col. Mike Mammay, who’s serving with the division in Kandahar in Regional Command — South.

“Intelligence enabled by DCGS-A facilitates situational understanding, reduces uncertainty, mitigates operational risk and supports decision-making — it is a critical tool used by our Military Intelligence Soldiers,” he said in an e-mail to Military​.com.

(Story was updated to include comments from Army colonel in the last two paragraphs.)

Associate Editor Brendan McGarry can be reached at

Tags: , , , , ,

Join the Conversation

No, no, no! You just have to adopt the mode of thought treasured by those who spend our money.

The fact that it doesn’t actually work for the end-user means that it is transformational and that the potential impact is immense. You simply have to spend more money on the system, re-purpose and expand it, and spend money finding out why the end-user is so wrong in their perception of unusability.

Admitting that the system is unworkable could lead to very unfortunate career and financial consequences at multiple levels and is therefore intolerable.

So it is simple. More money and more insistence that people be trained on and forced to use the system. It’s D.C. genius!

So it was tested under battlefield conditions and found wanting, eh? Scrap it and fire those responsible for wasting our time and money.

I use it now, in the field, and it works flawlessly. Why is this artice being released now when the source information is more than a year old? This is beating a dead horse. And no one finds it suspicious that Rep. Duncan Hunter is personally invested in Palantir? This whole thing is just corruption and lies. DCGS works. I’m sitting in front of it right now, using it, and it’s fine.

Putting software head to head is a art, not a science. Things like synthetic benchmarks only get you so far (e.g, runtime, memory consumption, etc). For data mining/machine learning there’s ROC and measuring TP/FP/FN/TN using blinded data to test ability to put the pieces together. Then there’s integration with disparate pieces of hardware in the field, and vertical integration of workflow components (e.g, do I have to export to format B and put into sub-program C with some regex parsing before loading into D?)

Detailed benchmarks are unlikely to be become a matter of public record. It is probable they use different algorithms to do their data mining and that DCGS has some things Palantir doesn’t (which is true of almost every duopoly system). At some point one system will win out over the other.

So, shouldn’t there be an independent test (menaing no one in DoD) on which reality is correct? How about a defense version of Underwriters Laboratory or Consumers reports?

Lt. Gen. Mary Legere: “It provides the underlying intelligence for every decision that our commanders and soldiers make in the field and it saves lives,”

That’s about the DUMMEST thing I’ve ever heard come from the mouth of a general officer… NO COMPUTER PROGRAM can “provide…intelligence for every decision that our commander and soldiers make in the field” without a HUMAN providing data input, analyzing that data, and then applying human intuition to synthesize it into meaningful intelligence.

Computer programs ARE A TOOL and are only helpful if they provide a USEFUL repository, network and workspace to facilitate the previously mentioned HUMAN functions effectively and efficiently. I’ve not yet encountered any US Military proprietary software that does not make such functions MORE difficult and time-consuming.

Sadly that is impractical as it will only lengthen the amount of time to put platforms in the field.

The quickest, best way to evaluate DoD software it to give it to the Soldiers who are supposed to use it and see if they can employ it effectively without 40+ hours certification training and a battery of civilian contracted babysitters to help them. If the software is not intuitive enough for a GUI end user to figure out in about 30 minutes, then we shouldn’t buy it…PERIOD!

To the best of my knowledge none of the US Army’s proprietary software would pass this test. None of them share functionality with popular off-the-shelf or web based programs or apps. They all seem counter-intuitive and confusing.

And I believe that is by design. Contractors make a fortune by increasing Soldier dependency on continued, intimate support and maintenance.

Shocker a Army General who want to buy crap instead of real systems because he likes his plan better.…. Shocker!!!

Operations research is here to stay, and things like the German tank problem can’t be solved by push-button systems. Not sure how much DCGS is meant to do, but presumably routine tasks with empirically optimized parameters can be push-button. Things that need careful handling will probably still need a data guy.

Real time info, distributed to all echelons, guaranteed not to malfunction at the most critical times.… Automation, the key to reducing the force structure.… Nothing like having a good back up system now is there? Problem is, no one remembers the manual systems to gather, interpret and distribute.… Everyone wants the computer to do it.….. HAL, Open the Pod Doors.

Like shooting blanks.

The generals pushing this sub par software probably have jobs lined up with the company that is making this software when they retire. Business as usual.

Hang the people who made and sold this fake war toy.

If you’re using it right now, why are you posting on an unclassified network in the same location?DCGS –A operates classified and if you are truly sitting in front of a DCGS-A, you are violating several security regulations by being on an open source and unclassified network in the same location. Ever heard of spillage.

Can you get onto the open web from SIPRnet? Considering Iranians can get into the NMCI, anything vaguely connected to the internet is open season.

I appreciate your support of DCGS-A. While DCGS-A may not be perfect, if you check off which DCGS-A requirements Palantir meets, you would find there are a lot requirements it does not meet.

I think your observations about investors losing money with Palantir is on the mark.

I say if this LTG, sitting in her nice,soft office chair, thinks it is such a great program put HER out on point with it. It has only killed 10 line Soldiers what is one extra LTG.

I have never observed a Gen who can function on the battle field. They are good behind a desk but when it comes to understanding the problems of people out in the battle, they are clueless. Anytime you get too dependent on technology you set yourself up for trouble.

Just like Obamacare and the american education system. Brilliant.

Not all generals are combat arms. And not everyone who makes it to general does so because they are the next Alexander the Great.

It is inproper to state that DCGS-A killed 10 line Soldiers. I could just as easily state that the CPOF was responsible for killing these 10 soldiers. Please hold in reverence the death of our soldiers, and place the blaim where it should be, on those that killed them.

Wow, never heard of Command Post Of the Future until now.

DCGS another vaporware system for which SAIC will receive $47 million tax payer dollars over the next two years. Combine DCGS with Trailblazer, City Time, Virtual Case Fiile, AHLTA and several more that I won’t mention.…..Time for a Congressional investigation into SAIC wouldn’t you all say?

Every politician who works in DC has a right to become rich. If you don’t agree simply look at the laws they passed for themselves.

Incorrect syntax. The proper command is “Open the pod-bay doors, HAL.”

Tru dat…

SAIC created LEIDOS (LDOS) and LDOS just fell $7.00 a share.…. A 20% share price loss in one day.!! Perhaps if the government will take no action against SAIC and LDOS. Well certainly Wall Street might…LDos was spun off in October 2013 and has lost 30% of it’s value.….

There are unclassified workstations sitting right next to classified workstations all over the world. There are even classified machines sitting next to unclassified machines sharing the same KVM all over the world. There are even some multilevel secure machines where people can operate SIPRNet and NIPRNet on the same workstation at the same time. So you are incorrect and behind the times.

If it’s not SAIC, it’s MITRE or GD or one of the other players. MITRE with it’s permanent contract to screw the government through classified operations is the company they really ought to investigate. But your very partial laundry list of failed intel programs is certainly telling. Army MI has never been able to see any sort of transformational program through from start to finish; they are always lots of money down the contractor drain. They should stick to incremental development of tools that work for the Soldiers doing the job.

You realize that the corruption is with the senators/congressmen who push through more funding for a project that gets terrible reviews from the users, because if they admit they were wrong and kill the money-pit project then it will affect the bottom line from their constituents and they might not be able to buy that vacation home in Aspen [this year]. Or the corruption when top-tier brass shake hands with these congressmen and get promised a political career after they retire from the military, so without any regard for the well-being of their soldiers they mandate use of a certain tool (conveniently the one their BFF in DC supports) and prohibit the use of another tool that the DoD has already footed the bill for. Really, $6,000,000,000 to build an analyst software system? That’s great you haven’t had a negative experience with this tool, but you’re incomprehensively delusional about where the corruption lies.

Used it in garrison. It crashed 6 times during 10 days of instruction (April 2014). One crash was so bad the instructors had to recall a technician to get it back up and on-line (and this is with training modules of a limited bandwidth already installed).
Once again, Generals, who do not understand or use a system, make decisions based on their “opinion” or interpretation. Its just like Fox or CNN calling a retired General to make a statement on Afghanistan or Iraq. They are talking heads, they only made decisions off of what the “real intelligence soldiers” gave them.
Look at and ask the MI Capts, Sgts, and Chiefs if it works.

NOTE: Comments are limited to 2500 characters and spaces.

By commenting on this topic you agree to the terms and conditions of our User Agreement

AdChoices | , and join us on Google+
© 2015 Military Advantage
A Monster Company.