people in this discussion seem to be like perl cod...

# datascience

[jimn]

people in this discussion seem to be like perl coders blissfully satisfied with thier methods and progress in accomplishing their goals, generally the question is "why do we need any less restrictive syntax"

[altavir]

er... no. I remember only one discussion about less restrictive syntax and most of people do not like the idea.

[jimn]

i work on expression solvers from time to time, more than a few times at various points in history. after a few go-rounds you get a feel for what is going to increase your transparency with a problem domain, and when a particular language is going to obliterate all clarity. defintiely would still hold to bash > python > kotlin > java

[jimn]

you can talk about taking the data science crown with kotlin, but as long as the focus is on type-safety and propeller head internals being the factor that will sway a community, it will remain hopeful, but quixotic

[jimn]

dissolving kolmogorov complexity from a problem statement and an implementation strategy is somehow lost on anyone who does anything with kotlin language. Arrow-kt is taking a flawwed premise forward with workarounds, and apparently isnt especially well received in here.

[jimn]

was there ever a moment you checked out an antlr codebase and had an "ah-ha" moment?

[jimn]

you can take DSEL's a lot farther with kotlin than a lot of other languages, but fundamentally it'sa coincidental occurence when smething fits, and an off-limits discussion when it doesnt

[altavir]

If you want expression analysis, you should stick with perl/bash. No questions about that. The rest of the discussion seems pointless so far.

[jimn]

and I'll be stuck with python and pandas for a loooong time to come while those points are missed