Dunning-Kruger applied to Dunning-Kruger, research on happiness, and type safety: http://danluu.com/dunning-kruger/
-
Show this thread
-
Amazing example of failure to citation/fact check: http://journalofpositivesexuality.org/wp-content/uploads/2018/08/Failure-of-Academic-Quality-Control-Technology-of-Orgasm-Lieberman-Schatzberg.pdf …pic.twitter.com/1PH7cTe41V
1 reply 26 retweets 49 likesShow this thread -
This reminds me that I have a draft post where I read all the papers citing a handful of obviously bogus papers to see if any citations would mention the issues. Never published b/c I couldn't figure out how to write this without seeming like I was picking on particular papers.
3 replies 2 retweets 15 likesShow this thread -
Periodic reminder that many (most?) citations are bogus. This happens to be about the claim that 3.5 hours/day of deliberate practice is the right amount, but I find this level of accuracy when I check most citations:https://acesounderglass.com/2019/06/21/epistemic-spot-check-the-role-of-deliberate-practice-in-the-acquisition-of-expert-performance/ …
5 replies 17 retweets 57 likesShow this thread -
So, about that desperate covid-19 related shortage of COBOL programmers in NJ that I've seen written up in six different publications...pic.twitter.com/gOMFJo1hL2
5 replies 275 retweets 482 likesShow this thread -
Replying to @danluu
Governments calling for volunteers they can afford to bid on is clearly bogus. Offer $1000/hour and the best contractors across the US will come calling. They need how many, 100 COBOL programmers? That's $200m/year. NJ budget is $60B/year.
2 replies 5 retweets 29 likes -
Replying to @davidcrawshaw
Relatedly, I've thought about writing up a post titled "Bad career advice". For 20+ years, I've been hearing that you should specialize in COBOL et al. since you'll rake in the $ when people retire. That seemed highly implausible 20 years ago and turned out to be totally wrong.
1 reply 0 retweets 5 likes -
Replying to @danluu
I have got a lot of value out of stable tech that is useful in many projects, e.g. the Unix sockets API. There is also clearly value in short-lived tech if you jump on the train at the right time *and* remember to jump off. Niche stuff is much trickier.
2 replies 0 retweets 3 likes -
Replying to @davidcrawshaw @danluu
I bet long-lived niche expertise like designing high-frequency analog circuits is very lucrative. So career advice needs to distinguish between skills like that and skills like COBOL.
1 reply 0 retweets 3 likes -
Replying to @davidcrawshaw @danluu
I feel like basic meta career advice should be something like — what do you think is the the risk / reward / enjoyment probability distribution (qualitatively)? How would you know if you were wrong? Find a smart friend with contrary perspective and discuss with them.
1 reply 0 retweets 1 like
IMO, a major related issue is that the "standard" sources of initial career advice (guidance counselor, professor/advisor) mostly have no idea what industry is like for any industry let alone multiple industries. There are exceptions, but none that I encountered in HS or college
-
-
For example, at least once a year, I see a prof say you should go into academia for N reasons, where half the reasons are things that are similar in industry, a couple are better in industry, and a couple are worse. Maybe you should go into academia, but not for those reasons!
0 replies 0 retweets 3 likesThanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.