How I build software quickly (evanhahn.com)
from learnbyexample@programming.dev to programming@programming.dev on 18 Jul 11:53
https://programming.dev/post/34114002

#programming

threaded - newest

ExLisper@lemmy.curiana.net on 18 Jul 12:23 next collapse

Cut corners, remove features and sacrifice quality. Pretty much what I do in my hobby projects.

iamtherealwalrus@lemmy.world on 18 Jul 14:09 next collapse

How a lot of software development ends up in real life too.

muhyb@programming.dev on 18 Jul 18:21 collapse

That’s… too realistic.

dukatos@lemmy.zip on 18 Jul 12:37 next collapse

So, go agile?

A_norny_mousse@feddit.org on 18 Jul 12:40 next collapse

c/thanksimcured

MonkderVierte@lemmy.zip on 18 Jul 14:41 next collapse

This is a developer perspective that doesn’t consider the big picture with security holes. And one of the reasons why software has gotten worse.

daniskarma@lemmy.dbzer0.com on 18 Jul 17:40 collapse

Don’t go crazy about abstraction.

Sometimes it’s better to write some code twice than expending two weeks trying to make a fit-all new class with 50 interfaces.

Sometimes you should do that, but for many projects that level of abstraction is not needed.

Ephera@lemmy.ml on 18 Jul 19:56 collapse

Well, the writing-part isn’t the bad part about duplicated code. It’s the maintaining of it. In particular, if you duplicate logic, it happens all too quickly that you make modifications to one, but not the other, or you make differing modifications to both.

Eventually, you’ll end up with two wildly different versions, where you won’t know why certain changes were made and not applied to the other version. Similarly, if you do need to make a similar change to both, you might now need to implement it two times.

I guess, I do agree that it isn’t *always* worth it, but in my experience, it is far more often worth it than one might think.