> time estimates from software devs actually make sense if you think of them as "this is how long it would take me the second time if i did this twice in a row"
My university had essentially a weird-languages course to blow our minds. Smalltalk (everything is a message to an object, back when C++ was the new hotness), Lisp (everything is a sexpr that you can redefine), Prolog (everything is a clause to search for known axioms).
Later: Lisp again, because of closures and CLOS let you program how method dispatch should work and CLCS let you resume from just before an error. Haskell, because you can lazily consume an infinite structure, and every type contains _|_ because you can't be sure that every function will always return a value. Java, back when the language was poor but the promise was "don't send a query, send code that the backend will run securely" (this was abandoned).
Was that the case for you? I'm especially curious about the exams if there were any, because it's probably hard to keep the whole context in mind as a student and to evaluate students for the teacher.
When the mouse bricked itself, I was forced to switch to the trackpad. Maybe I could plug the mouse in, if I happened to bring my only cable from home (or maybe from the office) that day. But there was no way of knowing when the mouse was usable again short of interrupting myself to test it. Over time, I decided the mouse wasn't worth the mental overhead and switched full-time to the trackpad, which always works.
Plugging in a laptop is a nuisance, but at least I can do that every time I sit down to start working and it will manage itself all day.
The left don't like the rich, the right don't like remittances, it's going to be hard everywhere. I heard some from a team who gave up on getting licensed.
python3 had no compatibility mode, so everyone needed 100% of their dependencies to migrate. This was so painful that some teams abandoned their legacy python2 code and reimplemented in languages with better back compat stories.
Python 3 was announced way ahead and came with migration tools that worked pretty well. Besides character handling most of the stuff was pretty similar at the beginning and I never understood why apparently nobody or only a few projects made the switch early on.
That lead to a chicken and egg situation: if you depended on those libraries that did not migrate to python3 you where stuck at python 2 as well.
I believe being nice to the community and supporting python 2 for a long time was a mistake. They should have made a hard break and enforce the migration...
POSIX doesn't forbid new flags or options. It's up to the author to read the spec and test his portability, or else willingly rely on certain distros. Some GNU tools have had strict modes as a courtesy (they used to jokingly call it POSIX_ME_HARDER).
Information that is known to be useful is sometimes replicated. The https://en.wikipedia.org/wiki/Rosetta_Stone#Rediscovery was abused and damaged; only four incomplete copies were gradually found by luck after two millennia. There are entire academic fields trying to learn things about our history by digging up stuff that was thrown away or abandoned.
reply