[jcoffland]

HN is pretty good at moderating discussions. Which is the main reason I come here. Still, I've seen plenty of beef only thinking here and I too have been guilty at times.

It can be quite frustrating when you make an observation about someone's comment only to have them automatically assume you were in disagreement. It's good to assume a generous interpretation. Since tone is so hard to gauge on the Internet, discussions quickly devolve otherwise.

[mrspeaker]

HN is now my last un-deleted social media account... and even here I end up removing 50% of my posts, and regretting another 30-odd% after realizing they were low-value emotional negative "beef" post.

Why is "if you can't say something nice, don't say anything at all" so hard in practice? Why do I just NEED to throw in my 2 cents?!

[ggreer]

The vast majority of people do follow that rule, but there are so many people reading these comments. If even 1% of them have a moment of weakness (or are kinda dickish personalities), they'll drown out the nice comments.

[humanrebar]

> if you can't say something nice, don't say anything at all

Because pleasantness isn't a virtue in every context? Some unpleasant things need to be done and said sometimes. In fact, saying no in certain contexts is good, though it can often be unpleasant.

That being said, some people find it pleasant to be unpleasant. Certainly self-censor if you sense that urge arising. Getting drunk on anger or righteousness isn't a good look.

[disqard]

To maintain type safety, function subtypes are contravariant in the input type and covariant in the output type.

"Be liberal in what you accept, and conservative in what you expect others to accept."

Maybe human communications also has to adhere to basic principles to guarantee robustness.

[paganel]

> I've seen plenty of beef only thinking here and I too have been guilty at times.

Most of the beef-related discussions around here involve either Tesla, Apple vs Google (when it comes to their phones/mobile OSs), FAANGs vs the rest of the world, some futuristic AI fields (like self-driving cars) and I think that's about it. There used to be a beef between supporters of static-typing vs dynamic-typing (I personally was in the latter group), but the static-typing supporters mostly won that debate.

Other than that most of the topics on this website are pretty civilised (with a few exceptions that confirm the rule).