2/ It's one thing to be wrong occasionally (as in software engineering) and to swallow your pride, correct course and move on. But having that be the norm can really be demoralizing.
Show this thread
1/ For every idea I have that actually works when researching and developing DeOldify, many, many more fail. One of the most important skills I've had to develop over the past year which is very different from software engineering is getting used to being wrong as the norm.
-
-
-
3/ It's tempting to fall in love with ideas and to start looking for evidence that you're right (confirmation bias) instead of trying to uncover the very likely scenario that you're wrong. This is a constant battle for me. I'd imagine this is true for others as well.
Show this thread -
4/ I'd say the value of the "emotional intelligence" component of research and venturing into the unknown isn't something I've heard talked about much but it's very real to me.
Show this thread -
5/ In fact I'd say most log jams in my progress in research revolve around allowing ego to take over rather than being disciplined about following the evidence. In particular, clever theories of mine are particularly difficult to discard when the evidence just doesn't pan out.
Show this thread -
6/ I'm just putting this all out there in case anybody feels weird/alone/stupid about running into these same situations.
Show this thread
End of conversation
New conversation -
-
-
Yeah the way I educate my engineers and other stakeholders on this is wrapping up this work into a separate risk category with hypotheses experiments, and results producing insights. It's actually very similar to how I've evaluated new tech and present it like that to my team.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
- Show replies
New conversation -
-
adage my dad used when people asked him how he achieved "I've failed more times that you've tried"
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
The process of hypothesis testing is one of the things that makes junior programmers advance the faster. If you teach them to never trust their feelings when debugging, they become outstanding programmers pretty fast. Worth it outside of research too.
-
Yes great advice! I always tell new developers: Collect the evidence FIRST, then narrow down the possibilities BEFORE you start rattling off theories of why your program isn’t working. Otherwise you’ll waste a lot of time.
- Show replies
New conversation -
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.