"It writes boilerplate faster."

Is that all we aspire to now? Not abstraction or templating or better language design, just... output boilerplate faster.

It's depressing.

@xgranade the only fucking metric anywhere and everywhere is lines of code.

Agile! Nope, lines of code.
Feature Driven Development! Do I even need to.

Measuring anything of actual value is hard. wc -l is easy.

@xgranade I think the people who use this argument do not know that some software engineers know how to avoid having to write boilerplate day in and day out.
@xgranade This chat bot could have been a hotkey. Or a preprocessor macro.
@feonixrift @xgranade and it would actually be deterministic and not cost money on every single run.
@paulshryock @feonixrift @xgranade management has always had an intrusive effect on software development but this takes the cake.
@xgranade yay it makes more of the thing we want less of
@xgranade i equally blame the incurious dolts and the CS education system that only teaches how to make more code
@xgranade and it isn't even good at that! we had boilerplate generation with m4 in 1987! "worse than m4" takes work to get to
You can choose a language that has less boiler plate.
*** THis is basically a non-starter in a business setting,
BUT probably the right call in my opinion in lots of cases.

OR you can use this chaos clamping device to write boilerplate
for you (THAT HUMANS WILL HAVE A HARD TIME READING).

Of course they pick the latter.
How can we begin to stop them having that power?
How can we get the bad actors away from these levers of power.
(*power does not mean good)
@xgranade we probably also tend to overstate how much boilerplate there really is. a lot of the common infrastructure is not actual boilerplate, but slightly varies by task, eg defining the properties of a class is something you do on every class, so the *typing out code* is routine, but the actual deciding on types isn't. but we lump both of those into the "boring boilerplate" category even though one of them is tedious routine and the other is *the fundamental work of software engineering*
@xgranade Not once in my career has boilerplate ever been the most time-intensive part of programming. And when it's necessary, you can just like... use a template or a hotkey or something. Some languages even have really nice systems for this, like Lisp with macros. You don't need to boil the ocean to use what is effectively Jinja most of the time.
@theorangetheme I've definitely hit cases where boilerplate was a bit of a bottleneck, and where I was working on contexts such that macros weren't really an option, but never has it crossed my mind to just... make the problem worse. Like, if you absolutely have to, code generating to files that you then .gitignore is a thing?
PEP 638 – Syntactic Macros | peps.python.org

This PEP adds support for syntactic macros to Python. A macro is a compile-time function that transforms a part of the program to allow functionality that cannot be expressed cleanly in normal library code.

Python Enhancement Proposals (PEPs)
@xgranade @6d03 help I can only abstract from specs :)
Music code story , including human speech nearly any major good or bad artform is abstraction at the end of the day
-- a human
Fwiw - I ve seen claude fail at crawling ( which is fair) , getting the specs wrong despite feed the manual , knowing my machine and me ( for good)
But yes #bigtech #socialmedua is hellbent on making it work

@xgranade The real problem for me is the existence of so much boilerplate (without tooling for it) that anyone is even tempted to reach for a chatbot.

Does anyone have an example for that? Because I don't. And I've done this shit for more than 30 years now.

@xgranade When people started using that line on me my response was always, if your programming is requiring that you write so much boilerplate, there's probably a better way of going about what you're trying to do in the first place.

But that argument never gains any traction with them.

@xgranade we need to generate more slop!!