4 ms·
Another jewel from that author https://web.archive.org/web/20110702162929/https://zedshaw.com/essays/programmer_stats.html https://web.archive.org/web/20110702
by 29athrowaway 3y ago
Another jewel from that author
https://web.archive.org/web/20110702162929/https://zedshaw.com/essays/programmer_stats.html https://web.archive.org/web/20110702162929/https://zedshaw.c...
- cmiles74 3y agoI remember reading this when it was first posted (probably to HN). It's been a long time but I think Zed still has a point. I've also had to revisit these issues with product managers or junior developers.
- 29athrowaway 3y agoOnly junior developers? Developers across all levels of seniority can suffer from these problems.
- SuperNinKenDo 3y agoThat was cathartic. Understanding statistics is a curse, because as the author states, you often understand that you don't know something. You can prove to somebody that their way of measuring something is complete horeshit, but if you can't come up with an alternative (because ypu know better), then you'll find people telling you "well we have to go with what ee have". Like dude. I literally showed you you don't have anything, you can't measure the thing you think you can, this measurement is lying to you. Sometimes institutional pressures account for this, someone has to show their manager something even if it's bullshit, but you get the same reaction when that isn't the case. People really can't accept the idea of not knowing something, that's just not a concept they can grok.