There’s a somewhat analogous case I often encounter (in my secret identity), along the lines of “You’ve shown evidence that programmers writing unit tests is beneficial to software projects in some particular cases, but until you show me published academic empirical studies saying that unit tests always save time and reduce bug counts, I’m going to keep writing code as before (no tests and trusting to hope as a method of proof).”
The absence of particular proof serves to dismiss even the readily available opportunities for self-experimentation which would allow the respondent to generate the very evidence they require.
Antipathy toward a particular self-proclaimed expert on self-help and akrasia around here shouldn’t be seen as a rejection of all techniques.
† I modified taw’s system to help me work more and waste less time on the Internet during the day. Briefly, from the time I wake up to the time I decide I’m done working for the day, I count time researching (or other productive active work) at 1 point per minute, listening to lectures at 1⁄2 point per minute, and wasting time on the Internet (including this site, sorry everyone) at −1 point per minute. I set my “good day” threshold at +180, and I’ve added a (somewhat complicated) reward system calculated easily on a spreadsheet. Then I can check LW and other sites guilt-free after the day is done. It’s worked pretty well!
†† This wasn’t the point of that post, but I’ve used it to my advantage with the add-on LeechBlock, which makes it just that much more inconvenient to stop working and start trawling webcomics/Twitter/LW.
I think at least most at SIAI do better. (I also think that attitude intuitively seems more common to me than it really is, simply because it annoys me so much.)
There’s a somewhat analogous case I often encounter (in my secret identity), along the lines of “You’ve shown evidence that programmers writing unit tests is beneficial to software projects in some particular cases, but until you show me published academic empirical studies saying that unit tests always save time and reduce bug counts, I’m going to keep writing code as before (no tests and trusting to hope as a method of proof).”
The absence of particular proof serves to dismiss even the readily available opportunities for self-experimentation which would allow the respondent to generate the very evidence they require.
Good call. Reminds me of a common attitude around here towards self-help/anti-akrasia techniques....
Not really, IMO. I should have posted my results before, but I’ve tried several of the methods suggested here, with some success, total failure, sustained success† and additional success††, respectively. I’d suspect many others have been trying various techniques as well.
Antipathy toward a particular self-proclaimed expert on self-help and akrasia around here shouldn’t be seen as a rejection of all techniques.
† I modified taw’s system to help me work more and waste less time on the Internet during the day. Briefly, from the time I wake up to the time I decide I’m done working for the day, I count time researching (or other productive active work) at 1 point per minute, listening to lectures at 1⁄2 point per minute, and wasting time on the Internet (including this site, sorry everyone) at −1 point per minute. I set my “good day” threshold at +180, and I’ve added a (somewhat complicated) reward system calculated easily on a spreadsheet. Then I can check LW and other sites guilt-free after the day is done. It’s worked pretty well!
†† This wasn’t the point of that post, but I’ve used it to my advantage with the add-on LeechBlock, which makes it just that much more inconvenient to stop working and start trawling webcomics/Twitter/LW.
ETA: Also, some success and marginal success.
Though I think that SIAI itself does a better job than that, don’t you?
I think at least most at SIAI do better. (I also think that attitude intuitively seems more common to me than it really is, simply because it annoys me so much.)