this post was submitted on 26 Jan 2024
100 points (96.3% liked)

Fediverse

27740 readers
474 users here now

A community to talk about the Fediverse and all it's related services using ActivityPub (Mastodon, Lemmy, KBin, etc).

If you wanted to get help with moderating your own community then head over to [email protected]!

Rules

Learn more at these websites: Join The Fediverse Wiki, Fediverse.info, Wikipedia Page, The Federation Info (Stats), FediDB (Stats), Sub Rehab (Reddit Migration), Search Lemmy

founded 1 year ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
[–] [email protected] 7 points 7 months ago (1 children)

I'm all for competition within the Fediverse (that's a big strength of the Fediverse if you ask me). I just wish we'd move on from Java and other exception-based languages.

[–] [email protected] 5 points 7 months ago (3 children)

What's wrong with exceptions?

[–] [email protected] 7 points 7 months ago* (last edited 7 months ago)

Exceptions opt out of the type system. The problem with them is the same problem as null. Here's a video of the creator of null explaining: https://www.infoq.com/presentations/Null-References-The-Billion-Dollar-Mistake-Tony-Hoare/

Null and errors are just values at the end of the day, and should be treated as such. Doing so means your code is far easier to reason about.

Rust takes this approach and is one of the many reasons people love it so much.

[–] [email protected] 5 points 7 months ago

What @[email protected] said, but also they kind of have the same problem as goto. The control flow becomes very complicated and you can jump from one place in the code to another extremely far away.

Spooky action at a distance, basically.

[–] [email protected] -2 points 7 months ago* (last edited 7 months ago)

Because exceptions are old and the new (recycled) kids are much more fun to play with? Or people yearn to implement the low level switch-case pattern error matching mechanism all over again, which try-catch-exceptions were solving.

I think there is no moving on from a paradigm as long as it has a function.