Comments on: A rigorous friction model for human-computer symbiosis http://blog.palantirtech.com/2010/06/02/a-rigorous-friction-model-for-human-computer-symbiosis/ Articles from the Engineering Group at Palantir Technologies Tue, 24 Jan 2012 09:51:18 +0000 hourly 1 http://wordpress.org/?v=3.2.1 By: Bacony Oatmeal http://blog.palantirtech.com/2010/06/02/a-rigorous-friction-model-for-human-computer-symbiosis/comment-page-1/#comment-823 Bacony Oatmeal Tue, 30 Nov 2010 20:59:09 +0000 http://blog.palantirtech.com/?p=1344#comment-823 the final formula only represents the tool set half of the whole problem. "a" is directly proportional to the data set and the tool set combined. a good tool set (as predominantly reflected in a low interface friction to my value system ;] ) with a marginal data set would render "a" to a minimum as would a very complete data set with a marginal tool set. overall analytic capability has to account for both factors. :? the final formula only represents the tool set half of the whole problem. “a” is directly proportional to the data set and the tool set combined. a good tool set (as predominantly reflected in a low interface friction to my value system ;] ) with a marginal data set would render “a” to a minimum as would a very complete data set with a marginal tool set. overall analytic capability has to account for both factors. :?

]]> By: Zach Travis http://blog.palantirtech.com/2010/06/02/a-rigorous-friction-model-for-human-computer-symbiosis/comment-page-1/#comment-714 Zach Travis Mon, 02 Aug 2010 02:02:26 +0000 http://blog.palantirtech.com/?p=1344#comment-714 Neat post. As you write, if h and c are high, then decreasing interface friction is the most effective way to increase analytical capacity, but as you might expect, at times it might make sense to improve h or c instead. For example, as a quick approximation, if (1+f)*(potential change in h) <- say, by changing analysts or users is greater than h*(potential change in f), then it makes more sense to upgrade the human factor, rather than your interface... likewise for computer upgrades. This could probably be fleshed out with some quick estimates of varying human/computer analytic capabilities, friction coefficients (maybe with the model system of chess programs in the earlier post?)...might be interesting! Neat post. As you write, if h and c are high, then decreasing interface friction is the most effective way to increase analytical capacity, but as you might expect, at times it might make sense to improve h or c instead. For example, as a quick approximation, if (1+f)*(potential change in h) <- say, by changing analysts or users is greater than h*(potential change in f), then it makes more sense to upgrade the human factor, rather than your interface… likewise for computer upgrades. This could probably be fleshed out with some quick estimates of varying human/computer analytic capabilities, friction coefficients (maybe with the model system of chess programs in the earlier post?)…might be interesting!

]]> By: Henrique Abreu http://blog.palantirtech.com/2010/06/02/a-rigorous-friction-model-for-human-computer-symbiosis/comment-page-1/#comment-579 Henrique Abreu Mon, 07 Jun 2010 12:21:26 +0000 http://blog.palantirtech.com/?p=1344#comment-579 Intuitively I'd also say that if the friction is really bad, I mean when the parts hinder themselves, it compromises the result in way that it would be better for them to work separately. Or at least there's no gaining at all, I'd write something like lim f->reallyBad (a) = Math.max(h,c)-badFrictionDecrement But it really needs a harder thought to express mathmatically. Anyway, great article, I loved it! Intuitively I’d also say that if the friction is really bad, I mean when the parts hinder themselves, it compromises the result in way that it would be better for them to work separately. Or at least there’s no gaining at all, I’d write something like
lim f->reallyBad (a) = Math.max(h,c)-badFrictionDecrement
But it really needs a harder thought to express mathmatically.

Anyway, great article, I loved it!

]]> By: Max http://blog.palantirtech.com/2010/06/02/a-rigorous-friction-model-for-human-computer-symbiosis/comment-page-1/#comment-577 Max Mon, 07 Jun 2010 00:34:01 +0000 http://blog.palantirtech.com/?p=1344#comment-577 Do you know this work? http://books.google.com/books?id=7lo0ag-noh8C It may seem like slightly too big a hammer, but take a look at 4.1, 7.4. Do you know this work? http://books.google.com/books?id=7lo0ag-noh8C It may seem like slightly too big a hammer, but take a look at 4.1, 7.4.

]]>