11 ms·

A story about planning for Desert Storm

76 points by bibliographer 4y ago

jhgb 4y ago
Quite interesting. I though this would be about DART (https://en.wikipedia.org/wiki/Dynamic_Analysis_and_Replanning_Tool https://en.wikipedia.org/wiki/Dynamic_Analysis_and_Replannin...), but I take it that this is more about front line stuff.
- jamessb 4y ago
  "Introduced in 1991, DART had by 1995 offset the monetary equivalent of all funds DARPA had channeled into AI research for the previous 30 years combined" Thanks for the link - someone at BBN had told me a similar story, but I didn't know the name of the program.
- dayofthedaleks 4y ago
  Amateurs talk about killbots. Professionals train their AIs on logistics.
  - imglorp 4y ago
    I think it depends, right? If you have a small number of high cost, high impact UAVs, you will target infra and supply columns. If you happen to have a cargo plane full of cheap, mass produced head-popper slaughterbots which you can dump over a town, they will remove all the humans and leave you with intact buildings, vehicles, and infra. https://www.youtube.com/watch?v=HipTO_7mUOw https://www.youtube.com/watch?v=HipTO_7mUOw
    - rsecora 4y ago
      Parent comment was jokingly paraphrasing General Bradley words: “Amateurs talk strategy. Professionals talk logistics."
- buescher 4y ago
  I didn't know about DART. Developed and deployed in eight weeks [edit: less optimistic sources say 23 months]! Is there a good retrospective somehwere with some technical detail?
  - monkeybutton 4y ago
    Seconding. I'd be interested in reading more into it.
bannedbybros 4y ago
numlocked 4y ago
Wow, I have never heard this before: “Few people remember it, but Gorbachev was negotiating with Saddam and had successfully convinced him to pull out of Kuwait. The agreement they came up with would give the Iraqis three weeks to pull out. At this point, it had become a major goal to eliminate the Republican Guard and we didn’t want them to pull their head out of the noose, so President Bush turned down the compromise and ordered the ground forces in.” Is that true?
- AlgorithmicTime 4y ago
- soviettankie 4y ago
  - sudosysgen 4y ago
    It would probably be better to do a simple fact check before going on a dubious grand narrative : https://www.nytimes.com/2011/01/20/world/middleeast/20archive.html https://www.nytimes.com/2011/01/20/world/middleeast/20archiv... > After prodding by Mr. Gorbachev, Mr. Hussein had offered to withdraw Iraqi troops from Kuwait in 21 days. But the United States appeared to be moving ahead with its land campaign.
    - antattack 4y ago
      21 days seems like a delay tactics since it took Iraqi troops just two days to invade.
      - waffleiron 4y ago
        It probably took them longer to prepare than just 2 days.
      - teh_klev 4y ago
        Iraq was able to "complete" its invasion and decapitation of military command & control in two days because the Kuwaitis were massively under prepared. After that you then need to consolidate your position and that would be days if not weeks of deploying troops and their equipment deeper into the country. So after seven months of occupation there's probably a heap of logistics required to extricate and decamp your forces & equipment that likely isn't going to happen even in 10-15 days.
techdmn 4y ago
I'm shocked at the lack of humanity in this story. Every once in a while an "enemy" will start a war, and we'll get a bunch of coverage about orphans and mothers and fathers and horror, and I think maybe just maybe people will realize that all wars are terrible. Except when we start them, then I guess it's mostly kind of funny because our intelligence wasn't as great as it could have been?
- bpodgursky 4y ago
  The US didn't start the war that Desert Storm ended. https://en.wikipedia.org/wiki/Iraqi_invasion_of_Kuwait https://en.wikipedia.org/wiki/Iraqi_invasion_of_Kuwait
- deburo 4y ago
  All wars are bad and the US doesn't go to war because it thinks it's funny. Unrelated people may joke about it, as usual.
  - AtlasBarfed 4y ago
    Why were we in Iraq the second time? Why were we in Vietnam? Why did we stay in Afghanistan for so long? The military industrial complex does indeed find war funny. It's called laughing all the way to the bank.
- xyzzy21 4y ago
  This is what war is all about. This is why many want cooler heads to prevail in the Russia-Ukraine war now! Those who have lived through war know how horrific and uncontrollable it is and how sociopathic you need to become to wage it successfully. Intelligence is the first problem. Then the model to fit it to. Then the conclusions to take from it. Each layer is HIGHLY prone to error. Just an example: precision munitions largely eliminate the odds of NOT killing to zero but then it's the intelligence and interpretation that becomes the weak link: are you really hitting the right target for the larger purpose, or are you doing everything wrong hitting the wrong target because you are measuring success wrongly (e.g. body count - as we did in Vietnam and then returned to in Iraq and Afghanistan - simply because it was the only thing we could measure; classic "streetlight fallacy") https://en.wikipedia.org/wiki/Streetlight_effect https://en.wikipedia.org/wiki/Streetlight_effect https://en.wikipedia.org/wiki/McNamara_fallacy https://en.wikipedia.org/wiki/McNamara_fallacy
Arainach 4y ago
This seems to have flown entirely over my head: QJM == Quantified Judgment Model. Other than what the acronym stands for, I'm still not sure what it is or why it's better, as this article never explains. The best short explanation I've found is "The QJM consists of two submodels whose interactions represent several battlefield intangibles such as leadership, morale, and training." [1] WEI/WUV (that's one thing, not two, and if you try to search for just WEI you'll get nowhere) == Weapon effectiveness index/weighted unit value. [2] It seems to me that this 2017 post from the same blog [3] is all but required prereading, with the problem that many of its linked sources (such as the one that might explain WEI/WUV) are dead and point to .mil domains that do not currently exist. [1] https://calhoun.nps.edu/handle/10945/23101 https://calhoun.nps.edu/handle/10945/23101 [2] https://www.jstor.org/stable/26553096 https://www.jstor.org/stable/26553096 [3] http://www.dupuyinstitute.org/blog/tag/wei-wuv/ http://www.dupuyinstitute.org/blog/tag/wei-wuv/
- wrp 4y ago
  PDF for [2] at https://info-gap.net.technion.ac.il/files/2018/05/mil-eff006.pdf https://info-gap.net.technion.ac.il/files/2018/05/mil-eff006... Christopher Lawrence took over after Dupuy. He explains the current state of their project in War by Numbers (https://www.goodreads.com/book/show/32824088-war-by-numbers https://www.goodreads.com/book/show/32824088-war-by-numbers). He brings this methodology to bear on some recent conflicts in America's Modern Wars (https://www.goodreads.com/book/show/22224145-america-s-modern-wars https://www.goodreads.com/book/show/22224145-america-s-moder...). Reviews have been very skeptical.[1,2] [1] https://journals.sagepub.com/doi/10.1177/0968344517747141k https://journals.sagepub.com/doi/10.1177/0968344517747141k [2] https://www.miwsr.com/2016/downloads/2016-039.pdf https://www.miwsr.com/2016/downloads/2016-039.pdf [PDF]
Animats 4y ago
This reads like an ad for the QJM model. The DePuy Institute had a proprietary model for figuring out who wins a battle. This was invented by General William DePuy, one of the leading planners of the US loss in Vietnam. He came up with a Quantitative Judgement Model for evaluating military strength. It's mostly adding up weapons values (sword = 1.0), although there are other features of the model. There are tables, now used mostly by war gamers. DePuy looked at a large number of historical battles, and came up with weights which seemed to work. The result was supposedly that the side with a 2:1 advantage almost always won. Within the 2:1 range, it was a tossup.
- mandevil 4y ago
  Two different people with differently spelled last names: General William DePuy (who was MACV C-o-S, BigRedOne-Six and then ran TRADOC after the war, and is considered the architect of the rebuilding of the Army after Vietnam) is unrelated to Col. Trevor Dupuy. Both of them fought in the US Army in WW2, and have confusingly similar names, but Dupuy is the one who founded the Dupuy Institute to do his numerical analysis of combat with all of his models and fudge factors.
  - Animats 4y ago
    Oh! Thanks.

76 points by bibliographer 4y ago

jhgb 4y ago

Quite interesting. I though this would be about DART (https://en.wikipedia.org/wiki/Dynamic_Analysis_and_Replanning_Tool https://en.wikipedia.org/wiki/Dynamic_Analysis_and_Replannin...), but I take it that this is more about front line stuff.

jamessb 4y ago
"Introduced in 1991, DART had by 1995 offset the monetary equivalent of all funds DARPA had channeled into AI research for the previous 30 years combined" Thanks for the link - someone at BBN had told me a similar story, but I didn't know the name of the program.
dayofthedaleks 4y ago
Amateurs talk about killbots. Professionals train their AIs on logistics.
- imglorp 4y ago
  I think it depends, right? If you have a small number of high cost, high impact UAVs, you will target infra and supply columns. If you happen to have a cargo plane full of cheap, mass produced head-popper slaughterbots which you can dump over a town, they will remove all the humans and leave you with intact buildings, vehicles, and infra. https://www.youtube.com/watch?v=HipTO_7mUOw https://www.youtube.com/watch?v=HipTO_7mUOw
  - rsecora 4y ago
    Parent comment was jokingly paraphrasing General Bradley words: “Amateurs talk strategy. Professionals talk logistics."
buescher 4y ago
I didn't know about DART. Developed and deployed in eight weeks [edit: less optimistic sources say 23 months]! Is there a good retrospective somehwere with some technical detail?
- monkeybutton 4y ago
  Seconding. I'd be interested in reading more into it.

bannedbybros 4y ago

numlocked 4y ago

Wow, I have never heard this before: “Few people remember it, but Gorbachev was negotiating with Saddam and had successfully convinced him to pull out of Kuwait. The agreement they came up with would give the Iraqis three weeks to pull out. At this point, it had become a major goal to eliminate the Republican Guard and we didn’t want them to pull their head out of the noose, so President Bush turned down the compromise and ordered the ground forces in.” Is that true?

AlgorithmicTime 4y ago
soviettankie 4y ago
- sudosysgen 4y ago
  It would probably be better to do a simple fact check before going on a dubious grand narrative : https://www.nytimes.com/2011/01/20/world/middleeast/20archive.html https://www.nytimes.com/2011/01/20/world/middleeast/20archiv... > After prodding by Mr. Gorbachev, Mr. Hussein had offered to withdraw Iraqi troops from Kuwait in 21 days. But the United States appeared to be moving ahead with its land campaign.
  - antattack 4y ago
    21 days seems like a delay tactics since it took Iraqi troops just two days to invade.
    - waffleiron 4y ago
      It probably took them longer to prepare than just 2 days.
    - teh_klev 4y ago
      Iraq was able to "complete" its invasion and decapitation of military command & control in two days because the Kuwaitis were massively under prepared. After that you then need to consolidate your position and that would be days if not weeks of deploying troops and their equipment deeper into the country. So after seven months of occupation there's probably a heap of logistics required to extricate and decamp your forces & equipment that likely isn't going to happen even in 10-15 days.

techdmn 4y ago

I'm shocked at the lack of humanity in this story. Every once in a while an "enemy" will start a war, and we'll get a bunch of coverage about orphans and mothers and fathers and horror, and I think maybe just maybe people will realize that all wars are terrible. Except when we start them, then I guess it's mostly kind of funny because our intelligence wasn't as great as it could have been?

bpodgursky 4y ago
The US didn't start the war that Desert Storm ended. https://en.wikipedia.org/wiki/Iraqi_invasion_of_Kuwait https://en.wikipedia.org/wiki/Iraqi_invasion_of_Kuwait
deburo 4y ago
All wars are bad and the US doesn't go to war because it thinks it's funny. Unrelated people may joke about it, as usual.
- AtlasBarfed 4y ago
  Why were we in Iraq the second time? Why were we in Vietnam? Why did we stay in Afghanistan for so long? The military industrial complex does indeed find war funny. It's called laughing all the way to the bank.
xyzzy21 4y ago
This is what war is all about. This is why many want cooler heads to prevail in the Russia-Ukraine war now! Those who have lived through war know how horrific and uncontrollable it is and how sociopathic you need to become to wage it successfully. Intelligence is the first problem. Then the model to fit it to. Then the conclusions to take from it. Each layer is HIGHLY prone to error. Just an example: precision munitions largely eliminate the odds of NOT killing to zero but then it's the intelligence and interpretation that becomes the weak link: are you really hitting the right target for the larger purpose, or are you doing everything wrong hitting the wrong target because you are measuring success wrongly (e.g. body count - as we did in Vietnam and then returned to in Iraq and Afghanistan - simply because it was the only thing we could measure; classic "streetlight fallacy") https://en.wikipedia.org/wiki/Streetlight_effect https://en.wikipedia.org/wiki/Streetlight_effect https://en.wikipedia.org/wiki/McNamara_fallacy https://en.wikipedia.org/wiki/McNamara_fallacy

Arainach 4y ago

This seems to have flown entirely over my head: QJM == Quantified Judgment Model. Other than what the acronym stands for, I'm still not sure what it is or why it's better, as this article never explains. The best short explanation I've found is "The QJM consists of two submodels whose interactions represent several battlefield intangibles such as leadership, morale, and training." [1] WEI/WUV (that's one thing, not two, and if you try to search for just WEI you'll get nowhere) == Weapon effectiveness index/weighted unit value. [2] It seems to me that this 2017 post from the same blog [3] is all but required prereading, with the problem that many of its linked sources (such as the one that might explain WEI/WUV) are dead and point to .mil domains that do not currently exist. [1] https://calhoun.nps.edu/handle/10945/23101 https://calhoun.nps.edu/handle/10945/23101 [2] https://www.jstor.org/stable/26553096 https://www.jstor.org/stable/26553096 [3] http://www.dupuyinstitute.org/blog/tag/wei-wuv/ http://www.dupuyinstitute.org/blog/tag/wei-wuv/

wrp 4y ago
PDF for [2] at https://info-gap.net.technion.ac.il/files/2018/05/mil-eff006.pdf https://info-gap.net.technion.ac.il/files/2018/05/mil-eff006... Christopher Lawrence took over after Dupuy. He explains the current state of their project in War by Numbers (https://www.goodreads.com/book/show/32824088-war-by-numbers https://www.goodreads.com/book/show/32824088-war-by-numbers). He brings this methodology to bear on some recent conflicts in America's Modern Wars (https://www.goodreads.com/book/show/22224145-america-s-modern-wars https://www.goodreads.com/book/show/22224145-america-s-moder...). Reviews have been very skeptical.[1,2] [1] https://journals.sagepub.com/doi/10.1177/0968344517747141k https://journals.sagepub.com/doi/10.1177/0968344517747141k [2] https://www.miwsr.com/2016/downloads/2016-039.pdf https://www.miwsr.com/2016/downloads/2016-039.pdf [PDF]

Animats 4y ago

This reads like an ad for the QJM model. The DePuy Institute had a proprietary model for figuring out who wins a battle. This was invented by General William DePuy, one of the leading planners of the US loss in Vietnam. He came up with a Quantitative Judgement Model for evaluating military strength. It's mostly adding up weapons values (sword = 1.0), although there are other features of the model. There are tables, now used mostly by war gamers. DePuy looked at a large number of historical battles, and came up with weights which seemed to work. The result was supposedly that the side with a 2:1 advantage almost always won. Within the 2:1 range, it was a tossup.

mandevil 4y ago
Two different people with differently spelled last names: General William DePuy (who was MACV C-o-S, BigRedOne-Six and then ran TRADOC after the war, and is considered the architect of the rebuilding of the Army after Vietnam) is unrelated to Col. Trevor Dupuy. Both of them fought in the US Army in WW2, and have confusingly similar names, but Dupuy is the one who founded the Dupuy Institute to do his numerical analysis of combat with all of his models and fudge factors.
- Animats 4y ago
  Oh! Thanks.