Tug of war continues over Army intel system

Army testers say the service's main intelligence network in Afghanistan is difficult to operate, prone to crashes, and vulnerable to hacking and that an alternate system would be better, reports Wired's Danger Room blog.

After limited testing of the multibillion Distributed Common Ground System-Army (DCGS-A) earlier this year, the Army Test and Evaluation Command
said the system is “effective with significant limitations, not suitable, and not survivable,” according to the story.

DCGS-A has been the subject of a major controversy in the military community since it was established in 2010. Various individuals in the military services and intelligence community have been trying to get it replaced with software from a Silicon Valley start-up Palantir. However, their efforts have been blocked by bureaucrats determined to protect DCGS-A, the story said.

Supporters of DCGS note that the system accesses hundreds of data sources for millions of reports, while Palantir only handles a small number, the story said. In an effort to simplify the cumbersome DCGS system, the Army has signed a cooperative agreement with Palantir to import some of the best attributes of its software.

The House Oversight Committee is looking into the matter, the story said.

Reader Comments

Fri, Sep 21, 2012 Rob Savage, MD

Funny what a difference a few weeks makes. Palantir's shameless campaign of attempting to discredit a system long under development and following a governance cycle has failed. The ironic part is that their tactics will not be forgotten by the government going forward, and since their products do not share code and cannot be properly integrated under information assurance guidelines, they will quickly fade from the landscape of support contractors for major government systems development. There are many other good business intelligence companies out there that can do what Palantir can. Lesson learned for other vendors? Don't bite the hand that feeds you. Learn the agency's governance rules and abide by them while trying to participate, solve problems, and add value.

Fri, Aug 17, 2012 OB1 Far Away

The comments above are all incorrect. DCGS-A is no different from any DOD system. Open framework?? Security requirements?? LOL! Perhaps you have heard of the company SAIC? They developed a lot of the ingestion software used on DCGS-A and they guarded it like Ft. Knox. Also, they were notorious for changing the baseline configuration in theater without the govt's consent. Furthermore, DCGS-A never provided any mechanism by which other systems could pull their data. The ARMY wants people to come to the so-called source. The funny thing is that DCGS-A really has no authoritative data. It just pulls data from other systems. The warfighter could survive without it. As for Palantir... who cares. DARPA has a couple programs in theater as well that have been deplored on a waiver as well.

Mon, Aug 13, 2012 Roger MA

Your article is very misleading. What "testers" are you referring to. The test and Evaluation Comand gave the system a thumbs up. The "bureaucrats" you refer to are program management professionals with many years of sophisticated systems development experience under their belts. We can disparage DoD systems development efforts, and there's a lot of room for improvement, but these developers don't let the public do their beta testing the way Microsoft does. They try to put a functioning product out with real security features. DCGS-A has had the extra challenge of being in development while concurrently accomplishing a real-world mission. NAme another system that has done that. Palantir will never be infused into DCGS-A. It doesn't meet DoD security requirements, in large part because its developers refuse to share their proprietary code.LAst I looked, government was (wisely) trig to avoid proprietary code and use open source and shared code (within a strict security framework). It makes sense. Problem is,Palantir doesn't play well with others. If DCGS-A wants to blend a good GUI or other good feature into its toolkit, there are dozens of other market intelligence tools that can do what Palantir does, only do it cooperatively as a part of a team.

Mon, Aug 13, 2012 withheld

Your story may be a bit upside-down. Bureaucrats from the vendor home state are trying to push Palantir, not protect DCGS-A, as you suggest. Palantir features are those of a police-oriented product. It seems unlikely to be capable of the range of DCGS-A warfighting functions without customization, augmentation, and integration. That's what afflicts DCGS-A today. If not addressed it defeats the purpose of replacing DCGS-A with Palantir, and could put Palantir into the same boat as DCGS-A today, but with even less history, and slipping schedules by years. A major issue is: many vendors don't understand the Systems Engineering changes needed when dealing with COTS integration, versus custom code, GOTS re-use, or single-vendor product customization. There is no indication that Palantir has such experience or a truly open framework so other vendors can assist with the significant work to be done, and Palantir is not likely able to do all the work themselves due to the range of expertise required. I have no "skin in the game" on either side of the debate, but would caution anybody who thinks Palantir is a slam-dunk replacement for DCGS-A to do their homework. The key to effective DCGS-A delivery is not to change horses mid-stream, but instead to understand and control the full range of Systems Engineering issues that come with projects of the size and complexity of DCGS-A.

Thu, Aug 9, 2012

This is a weak story, just a rehash of what Wired reported. Would have thought Defense Systems would at least advance the story by digging into a few details. Palantir could never replace DCGS-A. What Palantir does is only a small fraction of the capabilities in DCGS-A. In fact, when I was in Afghanistan, all the data Palantir used was pulled from DCGS-A. So without DCGS-A, Palantir would be useless. That may have changed is past 1.5 years, but I doubt it. Second, Palantir at that time was not accredited. DCGS-A had to go through some very stringent testing to get a PL-3 accreditation to operate on classified networks. Palanitir on the other hand was installed and operated on classified networks with a local waiver. Strikes me that this story is just a rehash of the Wired story, and the Wired story appears to be a rehash of a Palantir press release.

Please post your comments here. Comments are moderated, so they may not appear immediately after submitting. We will not post comments that we consider abusive or off-topic.

Your Name:(optional)
Your Email:(optional)
Your Location:(optional)
Please type the letters/numbers you see above