Article

2012-07-17 Afghanistan

US Military has to fight to purchase lauded IED buster

Archived material is restricted to Rantburg regulars and members. If you need access email fred.pruitt=at=gmail.com with your nick to be added to the members list. There is no charge to join Rantburg as a member.

Posted by Besoeker 2012-07-17 00:00|| || Front Page|| [11136 views ] Top

#1 Palantir

Posted by Slusosh Thrineck4715 2012-07-17 00:17|| 2012-07-17 00:17|| Front Page Top

#2 At long last, a fraud revealed!

Palantir, an effective, relatively easy to learn COTS (Computers Off the Shelf) system is just part of the Army software (SW) contracting, and systems acquisition problem.

Background: Digital Common Ground System (DCGS) (pronounced Deesigs) is an all encompassing Army system of the mid to late 1990's. It was designed to bring together operations, logistics, medical, and intelligence into one central digitized software. One system, one answer...what could possibly go wrong?

The only real Army takers for DCGS was the intelligence community (forced upon them by the way) where DCGS hitched it's wagon to a highly successful existing program called the Joint Intelligence Operating Center-Iraq (JIOC-I). The JIOC was the brain child of BG Keith Alexandera, then commander of the US Army Intelligence and Security Command at Fort Belvoir, VA. JIOC-I enabled soldiers to access massive amounts of data from archived DB's in the forward areas called "brains". JIOC-I was a simple program that permitted the soldier to access a 'flat network' containing dozens of diverse, existing systems, such as the highly successful, easy to use legacy program 'Analyst Notebook (AnB). Log on one time, use anything, the soldier's dream. JIOC-I integrated integrated newly developed SW licensing agreements as is normally done everywhere. JIOC-I also came with highly trained contract mentors and former soldiers, Marines, Airmen who assisted the soldier at the tactical levels. The DCGS program at first embraced and integrated the mentor concept, but is now reducing the mentoring cadre dramatically.

DCGS has taken bits of other programs and integrated them into DCGS hybrid Science Applications International Corporation (SAIC) promoted variant software. The rub came initially with the Palentir Software licensing costs which were seen as prohibitive. Palantir also created a problem by making other existing DCGS software redundant. Decisions were made by the Program Manager (PM) to NOT use the very popular Palantir or the legacy AnB, and to continue to steal bits from existing and new SW applications to morf and prop up the SAIC DCGS system.

DCGS quickly became and continues to be the bane of the Army analyst. As you might expect the outcry from the Army analyst was muted as he or she were told by Army leaders to forget about A&B and Palentir and drink the DCGS kool-Aid.

Worked for the Army, but not the USMC, SOF or Joint Community....who fortunately do NOT report to an Army program manager, his Army Pentagon bosses and funders or leaders.

This SW systems war is not just about Palantir, but proprietary systems and contracts in general. The controversy was a long time seeing the light of day, but it has finally been revealed. This is a fight I hope the Marines and our fine Army analysts win.

Hoorah Marines!

Posted by Besoeker 2012-07-17 03:10|| 2012-07-17 03:10|| Front Page Top

#3 Republican States should buy Palantir to look at voter fraud patterns...

Posted by Bright Pebbles 2012-07-17 05:01|| 2012-07-17 05:01|| Front Page Top

#4 Democrats should buy Palantir to look for right wing extremists.

Posted by bman 2012-07-17 09:56|| 2012-07-17 09:56|| Front Page Top

#5 Fred could buy Palantir to look for trolls...

Posted by manversgwtw 2012-07-17 10:51|| 2012-07-17 10:51|| Front Page Top

#6 Palantir has quite a history and has quite a capability.
Bank of America and the U.S. Chamber of Commerce wanted Palantir to foil Wikileaks. When that was exposed people were suspended at the company. It's capability in regards to artificial intelligence is as stunning as Stuxnet.

Read more here.

Unencumbered by legacy software languages it most likely works in multi-threaded mode on the software side with up to 12 core processors on the hardware side. Current can connect to any database regardless of the manufacturer and schema that older, slower legacy systems can't without a lot of coding on the development side or environmental reconfiguration on the client in the field.

The end user is King. Listen to them. Always. "The more sweat you put into training and preparation, the less blood you loose on the battle field."

Posted by wr 2012-07-17 18:57|| 2012-07-17 18:57|| Front Page Top

#7 Besoeker, I am very glad to hear that you appreciated AnB. I worked with the team that built and deployed portions of it. I agree it was a good system with the tools available at the time.

I don't agree though that DCGS is bad. It has some good features, but is hard to use and is not Palantir, which is sweet. Also, DCGS' bad features are certainly not due to the development contractor. To the extent that DCGS is bad, it is because the PM for it is making poor decisions for the user, and the contractor, SAIC, is obliged to make what they are contracted to build by the PM.

Posted by rammer 2012-07-17 20:04|| 2012-07-17 20:04|| Front Page Top

#8 Palantir was built with money from In-Q-Tel (the IC's venture fund) as well as from Peter Thiel's VC fund. While it has significant visualization and other capabilities, it works best with very large amounts of data. The DSGS program on the other hand must serve, among others, tactical operations centers and other lower-echelon analysis groups which often have limited bandwidth in theater.

In addition, DSGS is aimed at more than human-centric analysis. It intends to fully integrate the INTs (imagery, signals, sensors etc.) as a first step to full fusion across C4ISR. The program managers therefore end up making decisions that de-optimize one task (IED intel analysis, for example) in order to try to move forward with the larger mission.

FWIW - and yes, I could criticize elements of DSGS-A with gusto. But I understand the PMO's challenges and having a publicity campaign pushing them to adopt the product of a well-funded and politically well-connected product doesn't make their job easier.

Posted by lotp 2012-07-17 21:15|| 2012-07-17 21:15|| Front Page Top

#9 Palantir isn't compatible across the community and still can't handle large data sets. Like ANB, there are propietary issues between Palantir and ANB. To me it reminds me a bit of AxisPro and all that folks tried to incorporate with that SW. There were plenty of issues which stemmed from the "brains" when they were first deployed, but the concept is sound. Using SW to help sort out meaning from vast data sources. Even so, we still have fallen way short because the individual analyst is encumbered with sorting out data and doing a triage of unstructured and structured data... that's something Palantir and most other SW programs still can't resolve consistently with large, diverse data sets.

Posted by Squinty Chaith8856 2012-07-17 21:25|| 2012-07-17 21:25|| Front Page Top

#10 Well then, it is settled. I completely agree with the last couple comments; although, we would benefit from a clearer definition of terms like large data, which can't really turn up in this forum.

And having fought for better intel over many years, am also surprised by the wisdom and very careful consideration being applied to this topic here. Yet again, Rantburg U is delivering a masters thesis in the form of commentary.

PM DCGS ought to crowd-source his requirements and design alternatives in this forum prior to settling on a design solution for the next increment.

Posted by rammer 2012-07-17 21:51|| 2012-07-17 21:51|| Front Page Top

00:01 Redneck Jim
23:58 crosspatch
23:44 JosephMendiola
23:26 Anguper Hupomosing9418
23:24 JosephMendiola
23:16 JosephMendiola
23:04 JosephMendiola
23:02 JosephMendiola
22:21 Nimble Spemble
22:16 Thing From Snowy Mountain
21:51 rammer
21:35 trailing wife
21:25 Squinty Chaith8856
21:15 lotp
21:08 Procopius2k
20:58 European Conservative
20:55 trailing wife
20:53 trailing wife
20:29 Frank G
20:21 Raider
20:14 gromky
20:04 rammer
20:03 JosephMendiola
19:59 Rob Crawford