from Hacker News

Wicked

by thunk on 8/1/11, 3:28 PM with 16 comments

  • by zb on 8/1/11, 8:11 PM

    I was in total agreement up to this point:

    there really is no 'right' formulation and no 'right' answer. These are problems that cannot be engineered.

    On the contrary, engineering is 100% about addressing these kinds of problems (not all engineering problems match all 10 criteria, but most match at least some of them). Those people who think the engineering approach to problems is "define it, decompose and scope it, solve it, implement it" - or, as we call it, the Waterfall Method - have mistaken homework problems for engineering problems. They really have no idea what engineering is.

    Then there's this:

    our biggest challenges are ... issues of communication, coordination, and cooperation. These are, for the most part, well-studied problems that are not wicked.

    That's the most ridiculous thing I've ever heard.

    Incidentally, although he inexplicably doesn't link to it, Ritter and Webber's original paper "Dilemmas in a General Theory of Planning" is quite readable and well worth the time: http://www.uctc.net/mwebber/Rittel+Webber+Dilemmas+General_T...

  • by dodo53 on 8/1/11, 5:24 PM

    it feels a bit like: damn those scientists/engineers trying to solve complex problems with a one one-size-fits-all analytical/reductionist toolbag; some problems are really hard! What we need is to use our one-size-fits-all Structured Dialogue Design!

  • by msluyter on 8/1/11, 5:55 PM

    "What we need, in other words, is a Facebook for collaborative decision-making: an app built to compensate for the most egregious cognitive biases and behaviours that derail us when we get together to think in groups."

    I think what he's referring to may have already been proposed by Robin Hanson in the form of "Futarchy," a system of government that uses prediction markets to enact laws:

    http://hanson.gmu.edu/futarchy.html

  • by AndrewO on 8/1/11, 6:03 PM

    Anyone else having trouble finding a good definition of "Structured Dialogic Design"? Apparently it's been used in the Cypriot peace process, but I can't find a good description of how it differs from "getting a bunch of people in a room to talk about a problem".

    I'm hoping he threw that term out as a teaser for part 2.

  • by hendler on 8/1/11, 6:59 PM

    Some projects I've done failed precisely because of a desire to tackle wicked problems. When I heard that I wanted to "boil oceans", it was an apt criticism. However, I learned a lot.

    The article need not be taken as a criticism of engineering or scientific reductionism. These approaches to problem solving are correct for certain phases of tackling a problem (like implementation).

    The problem we collectively have with wicked problems is that they are vastly interconnected, and so many small moving parts rapidly changing that we collectively can not keep up. Even if climate is slow moving, all the parts that affect it are not, and we are not fast enough or smart enough to keep up.

    It is our self-righteous stance against nature that helps us survive, but admitting that a problem is bigger than us isn't ... natural.

    My belief/hope is that computers will increasingly tackle wicked problems.

  • by mchusma on 8/1/11, 6:20 PM

    I actually like this idea of an app to help craft arguments, but needs to be implemented on a smaller scale. Just a tool to help identify common logical fallacies would be pretty useful. Might be a way to ultimately leverage IBM's Watson technology, analyzing semantics to determine likely logical fallacies or cognitive biases.

  • by praptak on 8/1/11, 7:23 PM

    "[...] surely we can build a simple app that everybody can use that does even one useful thing, like, say, mitigate the Erroneous Priorities Effect when you're attending a meeting."

    This, I believe is a good old engineering problem. The hard (maybe not wicked) problem is getting people to use it. Ah yes, and if you really want to help solve the wicked problems, it can't just be any people but rather those who make/influence important decisions.

  • by samlevine on 8/1/11, 4:55 PM

    >Climate change is a great example of a wicked problem: Quick, somebody tell me what the acceptable maximum amount of CO2 in the atmosphere should be, in parts-per-milion! Provide me with the answer to that question, and you win a pony!

    We don't know. At some point 450ppm was considered acceptable, but at this point the maximum acceptable level may well be 350ppm (pre-industrial revolution it was around 275ppm).

    Right now we're at around 390ppm, so whatever it ends up being we have to reduce emissions from their current levels on at least a per-capita basis if not overall.

  • by indrax on 8/1/11, 6:07 PM

    Web search is the same way.

    This sounds like a lot of meta-level confusion.