Slashdot is powered by your submissions, so send in your scoop

 



Forgot your password?
typodupeerror
×
Programming The Military

Does The Military Need Agile Programming? (forbes.com) 141

OneHundredAndTen writes: According to this Forbes article, the Pentagon is worried that many in the USA's military nerve center claim to use Agile methods, when in fact, they aren't. Those responsible for these things at the Pentagon have therefore come up with a Detecting Agile BS document, so people can tell when they are doing Agile vs. when they are doing BS Agile. The implicit conclusion seems to be the usual "if it doesn't work for you, you are not doing it right."
The article was written by the author of The Age of Agile: How Smart Companies Are Transforming the Way Work Gets Done, a 2018 book arguing "An unstoppable business revolution is under way -- and it is Agile. Companies that embrace Agile Management learn to connect everyone and everything...all the time. They can deliver instant, intimate, frictionless value on a large scale." The book's author is Stephen Denning, who spent four years as Program Director of Knowledge Management during his decades of management at the World Bank.

His Forbes article this week warns "effective software development at DoD is not just a narrow issue affecting a few software developers. Questions of national cyber security and the integrity of the upcoming U.S. presidential election may depend on it... Fresh thinking and Agile mindsets are urgently needed."
This discussion has been archived. No new comments can be posted.

Does The Military Need Agile Programming?

Comments Filter:
  • get rid of up or out!

  • by Anonymous Brave Guy ( 457657 ) on Saturday September 28, 2019 @02:38PM (#59247384)

    ...The same way soldiers need high-tech weapons with complicated fire control systems, and pilots need jet fighters that will override their decision if they do something potentially dangerous in the middle of a combat situation.

    • Remember the US military recruiting posters saying: Uncle Sam Needs you!

      Now they should say: Agile Needs the US Military!

      I smell lots of juicy contractor jobs for Agile Consultants.

    • What they need is Functional Reactive Agile programming, also known as fragile programming.
  • by Anonymous Coward

    How to detect Agile BS is easy. If any of the following conditions are true, you have encountered Agile BS:

    * The term "Agile" is mentioned.

  • How about...... (Score:3, Insightful)

    by wolfie_cr ( 779921 ) on Saturday September 28, 2019 @02:44PM (#59247412) Homepage
    forgetting about this bs of thinking that adopting 'xyz' methodology is the magic pill???? anyone who has been working for 10-20 years knows how this goes, every 5 years someone comes with a new fad word...........its training, its books, its seminars/consultancies..........on how everything will be much better doing 'x' ...............I am turning into a cranky old fart that just sees this baby jesus people coming and leaving with a fat check and usually leave a half implemented 'solution' /bigger mess
    • Exactly. In fact the very first "DIB maxim" listed in that document is: "Competence trumps process". If you're getting stuff done, who cares if you're agile?
      • Wow, I'm going to take that quote and paste it on my wall. At least, I would if I had a wall. Cubicles are all gone around here. Competence is better than process.
      • by raymorris ( 2726007 ) on Saturday September 28, 2019 @05:15PM (#59247818) Journal

        > "Competence trumps process". If you're getting stuff done, who cares if you're agile?

        Certainly there is truth and value in that. ALSO there lay a trap. Individual competence does not in fact substitute for organizational process.

        I'm a nerd, who likes to study things. I enjoy continually improving in areas that I have some competence (which do not include music or sports). After 20 years of focusing on improving my skills in my chosen field of security-related programming, I've finally achieved a certain degree of competence in that particular field. At one time I thought that meant I didn't need to be concerned with process. My code worked without unit tests. Revision control? Why would *I* need that.

        I had to pick up projects that other reasonable competent programmers had done, figuring out what they had done and why - and why they hadn't done it the easy, obvious way. They had developed without the process elements of documenting the design, source control, and unit tests - so I had to guess at what it was supposed to do and hope that it continued to do it right.

          I had to work with other competent professionals, who made changes on their side of the system without a Request For Changes (RFC), and then react when their side no longer interacted and in quite the same way with my wide.

        Working with some very well done open source projects taught me the value of things like peer review and unit testing. I worked with people who were very competent individually - and that competent code was frequently fixed or improved through peer review.

        At my last long-term job I introduced a lot of process stuff which allowed more junior programmers to work effectively, and not make too many major mistakes, without having 20 years of study and experience. When I left, and they no longer had a senior engineer on the team, they survived just fine because process was in place to allow effective (though slower) work without demanding unusual skill.

        Even if you think that you personally can't gain anything by following a disciplined process, what about the people you leave behind when you of course get a better offer?

        The right process allows organizations to be consistently successful, even when the brilliant person leaves or goes on vacation.

        • by chthon ( 580889 )

          After being in the software field, professionally, for 25 years, this, this and 1000 times this.

          Electronics, electric and mechanical engineers make plans and schematics for much simpler things than software, because they understand that they need to document all their work so that people after them know why they took certain decisions.

          The problem with teaching software engineering: not learning people how to differentiate between process (good) and bureaucracy (bad).

    • Agile has been around for a long time now, at this point you can't call it a five year fad.
      • Extreme Programming: https://en.wikipedia.org/wiki/... [wikipedia.org]
        Was first talked about around 1995 and the first book published around 1999 ... in total roughly 25 years ago.

        However the /. crowd still thinks that "agile" is fad ... unfortunately agile only works if "competence trumps process" :P it seems that many people on /. work in software teams that are not very competent.

        Companies like Zalando are way beyond agile, as ordinary Scrum or XP is way to slow for them. But then comes the myth that CD is "tested by th

        • Re: How about...... (Score:5, Interesting)

          by ShieldW0lf ( 601553 ) on Saturday September 28, 2019 @10:22PM (#59248480) Journal

          When I started programming, you would do rigorous requirements gathering. You'd meet the client, you'd interview the executives, you'd interview representative members of each distinct group of users in the client company, then you'd go away and come up with a requirements document and that would form a contract of work. Then you'd architect a solution, you'd implement it, the client agree that it met the requirements of the contract, they'd put it into production and you'd get paid.

          But, there were some clients that were total fucking airheads, and they couldn't say precisely what they wanted, and they'd handwave away things that they thought were complicated and answer vaguely, and then later they'd tell you that this wasn't what they wanted, and you'd change it, and they'd think of additional things they wanted and you'd be expected to build them, and they were total messes. Sometimes these clients for some unknowable reason would have deep pockets full of money to spend, but you just knew you'd end up getting paid 100,000 for 200,000 worth of work by the time it was all said and done, or they'd run out of money when the work was half done and go around badmouthing you to other potential customers, etc. So, you'd "fire the client". As in, you're incompetent, I'm not working with you, go hire some other schmuck to follow you down the road to hell.

          Then Agile came along. It was a wasteful way to code, you never looked more than two inches in front of your face, you designed things in ways that you would never have done if you actually knew in advance what it was you were building, and it made you stupider the longer you did it.

          But it had one redeeming feature: It was an effective way to separate idiots from their money.

          That's Agile. A poor and inefficient programming methodology that is nevertheless very effective at draining money away from idiots who can't participate in requirements gathering well enough to tell you in advance precisely what it is that they need.

          I've done it. I have contempt for it, and I have contempt for the work I produce when I do it. But, hey, I like money.

          • Tell it, Brother.

            And might I interest you in a manifesto about this?

            I didn't write it.

            http://programming-motherfucke... [programmin...fucker.com]

          • Then Agile came along. It was a wasteful way to code, you never looked more than two inches in front of your face, you designed things in ways that you would never have done if you actually knew in advance what it was you were building, and it made you stupider the longer you did it.
            If that is your experience with "agile" then you obviously never did a proper agile project ... in other words: you did it all wrong :P

            well enough to tell you in advance precisely what it is that they need. That is obviously the

        • Only about 10% of the practicioners of any profession are truly competent.

          The rest are doing their best to not get discovered.
  • No. It does not (Score:5, Insightful)

    by pablo_max ( 626328 ) on Saturday September 28, 2019 @02:45PM (#59247414)

    What the military US military needs is to have its budget cut in half.
    Seriously, you Americans freak out out the thought of some poor person getting health care when he "didn't earn it", but mean while you spend like 20 times what everyone else in the world does on defense against some imaginary country who is apparently going to attack you at any moment. The ironic thing is that you do not even see that the military is the largest social and industrial welfare program in the US.
    NO ONE IS GOING TO ATTACK YOU!
    Maybe if you spent like 1 trillion a year less killing brown people you would not need to sell your house to pay for your cancer treatments.

    I never understood why this is such a point of pride for Americans. You hate the idea of giving any money to actually help anyone in your country or educate your populate, but you take pride if spending 20 times more than China on your Army to protect you from.... something.

    If you cut the military in half, you could end homelessness and give everyone in the US healthcare AND still have the largest Army.

    • You hate the idea of giving any money to actually help anyone in your country or educate your populate.

      Virtually all schools in the US are public schools. Nearly all the money for education comes from taxes.

      However, if by "help anyone" you mean "pay people's private expenses", that's correct.
      Defense is a legitimate function of a nation's government. Charity is not.

      • Re:No. It does not (Score:5, Interesting)

        by Uecker ( 1842596 ) on Saturday September 28, 2019 @03:16PM (#59247512)

        Why is "helping people in need" not a legitimate function of a nation's government? In all civilized societies it is.

        • In all civilized societies it is.

          "Civilized" is a word that is often used to mean, "stuff I like."

      • Re: (Score:3, Interesting)

        However, if by "help anyone" you mean "pay people's private expenses", that's correct.

        You do realize that investment is not the same as charity? If someone gets fucked by a series of unfortunate events they will be out on the street. People on the street do not generate taxes. If you put some state-funded "charity" towards fixing that person up again, he can go back to generating taxes. This is not a complicated scheme.
        If you help get drug addicts off the street, they might go back to generating taxes. If you help the large swath of people losing everything in 2008-2009, they could go back t

      • Why does it cost $3000 to see a doctor without insurance, when that insurance company actually gets paid when you go see said doctor through clawbacks (co-pays)??? Cuz insurance. Government backs those insurance companies so that you literally have to have insurance to protect you from medical bill gouging. How much did the doctor get for your visit? Hmm... Around $100-120... How much did an insurance company get? (Malpractice insurance is the number one cost doctors pay for practicing medicine)...

        Get rid o

        • I dunno, I don't have insurance and I don't pay $3000.

          Actually, I get a 5% discount off the list price because I pay at time of service.

          When I had a procedure at a hospital I got a 25% discount on much of it, for paying at time of service. Luckily I had the luxury of time to choose the hospital.

      • Virtual all schools in the rest of the world are: public schools, paid by tax money. Yet they don't need metal detectors, or special "anti weapons" measures. And in most countries the public schools give a real education.

        Defense is a legitimate function of a nation's government. Charity is not.
        So is health care ... well, in the developed world.

      • Defense is a legitimate function of the government. Offense is not. Close each and every foreign US military base, without exception.
      • Virtually all schools in the US are public schools. Nearly all the money for education comes from taxes.

        Are you sure about that? The latest numbers I could find say [ed.gov]:

        In 2015, private school students made up 10.3 percent of all elementary and secondary school students.

        I wouldn't call just under 90% "virtually all", but you might. For college the numbers are [statista.com]:

        In 2017, there were 14.56 million college students in the U.S. enrolled in public colleges and 5.1 million students enrolled in private colleges.

        That's 74% public.

        I haven't tracked down the numbers, but I'd bet my next paycheck the total spending per student is higher at private schools.

    • by Darinbob ( 1142669 ) on Saturday September 28, 2019 @03:05PM (#59247476)

      Ha, if the military went Agile, we'd lose in all future conflicts. Agile is the opposite of getting stuff done.

      Captain: I need you all ready to march at 06:00!
      Private: Sorry, we don't do waterfall anymore.
      Captain: You are out of line soldier!
      Private: I hear you, but we're already in the middle of a scrum and as a stakeholder you need to wait until the sprint is over before you provide new suggestions.
      Captain: wtf?
      Private: Right now I'm working on a new framework for squad formation, Private Smith is demoing how to polish your weapon, and Private Jones is gathering analytical data on the new beds.
      Sergeant: Private, you need to respect the chain of command, do I make myself clear!
      Private: As scrum master you are here to facilitate our progress, not to order us around.
      Captain: Well, at least I can still outsource to contractors.
      Mercenary: We do Kanban now.
      Captain: I picked a bad day to give up oxy.

      • A clear lack of focus on that team. They should schedule a retrospective weekly to figure out what went wrong and how they can all pull together for the next one.
    • by ceoyoyo ( 59147 )

      The US is one of the very few examples of a stable, long-lasting country formed by an armed revolution. It's also a frontier nation, founded by a bunch of rugged individualists. Or at least, those are the ones that survived.

      Those things combine to reinforce a national mythos of self-reliance, independence, and fear of outsiders.

      • No it is not one of the very few.
        China is one.
        Russia is one, too.

        Oooooops ...

        • "Stable". "Long-lasting". Neither Russia nor China of today have much resemblance to their past selves even 50 years ago.

        • Oops is right.

          Get your head out of your ass and look up how long those countries have been stable. Or even, the same country as now.

        • by ceoyoyo ( 59147 )

          China had a revolution in 1946. That's not exactly long lasting, as far as countries are concerned. Maybe one day. Even if you did want to count it, I said "one of the very few" not, "the only."

          Russia had a revolution in 1917, and then collapsed in 1991. Not exactly long lasting.

          Oooooops.

          • No, it had not.
            1946 is when the new peoples republic of china was founded.
            The revolution started 1880 or something ... kicking the european invaders out, one by one, and ended with dethroning the emperor 50 - 70 years later. You could even say the revolution lasted till 1980 as finally Vietnam kicked out the French.

            Russia did not collapse in 1991. It had *again* a revolution. The concept of revolution seems something you only grasp in the US context of an "independence war" which was not: a revolution.

            • by ceoyoyo ( 59147 )

              Even if we accept your contention that 70 years is "long-term," you've come up with a couple of examples in addition to the US. I said "one of only a few." So we've got a few. Notably, both China and Russia are also pretty militaristic.

              PS: Vietnam is not part of China.

        • by sjames ( 1099 )

          Russia had it's revolution in the early 20th century and that government already fell apart. So, not stable.

          China was mid 20th century, went through the leap forward and the cultural revolution and didn't become anything like stable until the '80s after multiple ideological purges ultimately jettisoning Maoism. Arguably not the same government that led the armed revolution.

          The U.S. happened in the 18th century. The Civil War was 19th century, net result, the states that seceded re-joined the union.

    • Re: (Score:3, Interesting)

      by EMB Numbers ( 934125 )

      The USA has treaty obligations to defend half the globe. European nations have treaty obligations for funding their own defense that they have NEVER met in 50 years because they know the USA will defend them anyway.

      Ask Ukrain if there are any threats of war in modern Europe.

      • Except even Ukraine calls it an anti terror operation, not a war.

      • Europe met the treaty obligations.

        But the americans forced a change on us a few years ago, and we struggle with them.

        Actually, we don't need those changes.

        Ask Ukrain if there are any threats of war in modern Europe.
        Why? They struggle in the aftermath of the destruction of the USSR. What has that to do with "Europe"?

        • The prevalence of idiots like you are why none of your militaries can even handle their own logistics in a conflict.

          It isn't just the tanks and guns that you need Americans for, or the airplanes and munitions. Even just simple things that don't require special technology, like cargo planes, you're unable to organize a commitment to funding, and you seem to believe you can begin investigating purchases after a conflict begins, which of course doesn't work.

    • Seriously, you Americans freak out out the thought of some poor person getting health care when he "didn't earn it", ...

      No, you were misinformed. About 18% or Americans are uninsured or underinsured. We are in favor of government intervening to help these people. What we freak out over is the government wanting to take over the entire system and screw up what is working for 82%. The private healthcare system in the US effectively pays for, subsidizes, the public system. 60% of hospitals report government payments on the public side are so low they are treating those patients at a loss, and that they make up the difference by

    • What the military US military needs is to have its budget cut in half.

      To start with, let me say I am generally against military use abroad. I would love it if the U,.S. pulled out of Afghanistan entirely, and took a bit more of an overseer role in world affairs instead of more direct meddling.

      However - the military does LOT of good around the world. The bases like it or not help prop up other countries economies, and engender goodwill.

      Furthermore in times of real disaster, a strong and competent military

    • The rest of the world is stable because of American military dominance, unfortunately. Our allies don't have to spend as much on defense because of the strength of the American military. That's a fact.
    • Every time the USA cuts back on military because "no one is going to attack us," someone attacks us. If you talk to Americans, we're mostly isolationists (because of the opposite: every time America gets adventurous, it goes badly. Teddy Roosevelt in the Philippines, for example. And Bush in Iraq). You leave us alone, and we'll leave you alone. That is the American ideal.
      • by sjames ( 1099 )

        Every time? How many times do you think we have been attacked?

        • It's been about once every 30-50 years. I will concede that the Cuban war wasn't the USA being attacked, but the American public perceived it that way. I won't even claim that the response of "build bases all across the world to prevent another world war" is effective, just that it's the reasoning behind it.
          • by sjames ( 1099 )

            We should also discount Korea and Vietnam. As for the WTC, that wasn't a nation state and the size of our military wasn't much of a factor before or after the attack. That leaves WWII. There aren't a lot of people left who actually remember that from an adult perspective. In WWI, the U.S. was not actually attacked though it was arguably sorely provoked and it could be argued that an attack couldn't have been far away. The Civil war is questionable since that was an internal matter. A larger army would have

            • There aren't a lot of people left who actually remember that from an adult perspective.

              Oh come on, if you want to argue that WW2 doesn't color many aspects of our culture, then I'll roll my eyes. What is the greatest evil of all time? Hitler.

    • I totally agree. We Americans should withdraw from NATO forthwith and remove all our troops and equipment from Europe without delay.

  • I actually use SAFe Agile in my team to help estimate actual times and get buy in. It's interesting watching Soldiers get asked what they think they should do for a change to accomplish the mission. The key factor in if it is appropriate is the types of projects. This definitely wouldn't work for infantry.
  • "if it doesn't work for you, you are not doing it right."

    ... about socialism.

    • Re: (Score:1, Insightful)

      by Tablizer ( 95088 )

      ... about socialism.

      Apparently we aren't doing capitalism "right" either: our middle class is shrinking.

      GOP's solution: more tax-cuts for the rich. They are surrounded by screws and they only carry hammers.

      • Our middle class is shrinking because more people are moving UP than moving down.

        - https://www.aei.org/carpe-diem... [aei.org]
        - https://www.forbes.com/sites/t... [forbes.com]
        - https://www.latimes.com/opinio... [latimes.com]

  • Sweet Jesus, stop! (Score:4, Interesting)

    by Nova Express ( 100383 ) <[moc.liamg] [ta] [nosrepecnerwal]> on Saturday September 28, 2019 @03:11PM (#59247496) Homepage Journal

    Stop with this "Your agile is not pure enough! Repent heretic!"

    Agile is not a panacea, not a magic bullet, not The One True Programming Methodology. It's a methodology that works for a lot of shops a lot of the time for a lot of projects. But I've worked at six or seven different places that did agile, all of which did it differently. The very worst case of trying to use agile I've seen was a company that was trying to do it in it's most pure sprint groups form, but most groups couldn't make any headway because the semantic database that was the heart of the system wasn't ready yet. It doesn't make any sense to build try and build GUI controls when it couldn't control anything, and for which the GUI primitives hadn't even been defined yet.

    Agile is a methodology, not a religion. Everyone can and should adapt it so that it meets the project and organization needs, regardless of how "pure." It is.

    One of the supposed virtues of agile doesn't even apply in the case of the military, since it doesn't need to be "close to the customer" since the military is it's own customer. Rapid iteration probably works a lot better building a web app than it does for controlling a multibillion weapon system that can blow up and/or kill people due to a bug.

    Could we please have fewer postings from The Agile Purity League?

    • by Junta ( 36770 )

      I am generally jaded on "Agile" but the pieces that I have some semblance of respect for are the ones that admit up front that these are all suggestions and do whatever your team feels comfortable with.

      On the flipside this is also the part where Agile evangelists take credit for any success no matter how that team actually executed while at the same time calling any failures "not true Agile"

    • by vadim_t ( 324782 )

      Your example seems to be more of an organization problem than an Agile problem. Regardless of the methodology, how can anybody get things done if they're assigned to work now on something that has a dependency that's not ready?

      Something else is wrong there. Either the people doing the GUI should be working on the database too, or they should be doing some completely unrelated project meanwhile, or they should be doing training, or if there's nothing useful at all for them to do perhaps they shouldn't have b

    • by Cederic ( 9623 )

      Have you read the document? Which of the questions it suggests asking are bad questions to ask? How would you answer them for your current work?

      One of the supposed virtues of agile doesn't even apply in the case of the military, since it doesn't need to be "close to the customer" since the military is it's own customer.

      You're opining on development methodologies and you can't even understand how the customer is? Oh dear.

  • by Tablizer ( 95088 ) on Saturday September 28, 2019 @03:22PM (#59247526) Journal

    Agile takes a lot of things to go right in order to be a net benefit. It's not forgiving of a missing link or two in the chain. The problem is that humans are good at hubris.

    • by cheesyweasel ( 5072497 ) on Saturday September 28, 2019 @03:51PM (#59247604)
      I agree with this. Agile requires strong teams and proactively followed processes. Without it, things fall back into traditional hierarchical, managerial feudalism and bullshit jobs. It's not easy.. it requires strong communities of practice and a collective desire to fix a broken environment (or prevent it from descending into chaos).
  • They can be the first place to actually do it correctly.
  • While the topic question appears a bit silly the question of speed of development is pretty important for an effective military. The F35 started development in 1992. Supposing that the critical problems get fixed in the next few years you have something functioning as intended 30 years later. Then the planes would need a lifetime of another 30 years to make them remotely worth their investment. In the meantime the environment they have to work in changes quickly.
    This lifecycle is a virtual guarantee for ma

  • ..."An unstoppable business revolution is under way -- and it is Agile. ...

    Unstoppable business revolution? Good grief. Get over yourself.

  • Everyone knows the military likes (silver) bullets.

  • Absolutely not. (Score:4, Insightful)

    by Qbertino ( 265505 ) <moiraNO@SPAMmodparlor.com> on Sunday September 29, 2019 @01:37AM (#59248750)

    Disclaimer: Agility advocate and Scrum person here.

    Agile software development processes are a good thing. A huge part of using them correctly is knowing when *not* to use them.

    Programming military devices and systems is one of those things. Medical systems, nuclear systems, spacecraft, airborne systems are some of the others.

    Agile software development processes are for using standardised software development pipelines to build software for people who don't know what they want but know exactly what it may cost and when it needs to be finished. In short, it's building sophisticated software for people who for their life can't think two steps ahead.

    If you find someone like that in charge of a military project: run. Or do consulting for obscene rates. But avoid doing anything productive in such projects. They are always a disaster.

    You're welcome.

    • by DeBaas ( 470886 )

      Hmm, there are two reasons why I partially disagree (and I'm not even a Scum fan)

      1) the military is a very large organisation. They probably have loads of processes governed by i.e. Excel sheets and/or Access databases. Replacing some of these with (rapidly build and not over engineered) webapps often leads to those processes being much more efficient and secure. The existing spreadsheets or MS Access apps are often great prototypes of what is needed. But they don't have to be 'military grade', this are per

    • by flink ( 18449 )

      Disclaimer: Agility advocate and Scrum person here.

      Agile software development processes are a good thing. A huge part of using them correctly is knowing when *not* to use them.

      Programming military devices and systems is one of those things. Medical systems, nuclear systems, spacecraft, airborne systems are some of the others.

      Most military software isn't for anything as exciting as nuclear control systems or aircraft firmware. It's mostly logistics, training materials, information dissemination, mission planning, payroll, etc. All the same crap that big enterprises deal with, with some extra highly specific security requirements and government bureaucracy sprinkled on top.

      The main obstacle to agile for military contracts is that most of the ones I've encountered that were not DARPA or pure R&D is that they were fixed price

  • What was wrong with how things were being done during the past 60+ years?

    Change for the sake of change is stupid, and it opens up too many new problems and risks.

    Fighter jets and other critical systems were always made with tight, fault tolerant code. Meaning the software running the plane must be 100% reliable. Of course, this means no fancy frame works, or graphics for the HUD, or any of that crap.The software Just Works(tm)

    We don't need fighter jets to become the next Boeing 737 M

  • whatever works for you is the best development model.

  • Agile is just as useful for creating prototypes as for finished products. If a military project has the time to engage in rapid iterative development to establish its proof of concept, it might be beneficial to use Agile in that fashion, then using the prototype to define the specifications for the "real version" of the product, to be developed in a non-Agile fashion.

There's no sense in being precise when you don't even know what you're talking about. -- John von Neumann

Working...