undefined | Better HN

0 pointsonion2k6y ago0 comments

That doesn't matter though. The point of plussing is about how you word your criticism so people don't feel the need to be defensive or antagonistic.

If there's a problem that can't be solved for a while there's world of difference between saying "Your code is wrong. This is a problem we can't solve for weeks." and "You've made a great start and this code is going to be really useful as a foundation to build on, but there's a new challenge we didn't foresee so we'll need to make a couple of changes when we have more data."

I've never met a technical problem that couldn't be solved better with good people skills. Knowing how to get good work from people is key for any tech business.

0 comments

brankoB6y ago

What if the code is absolutely terrible and needs to be scrapped ASAP?

onion2kOP6y ago

Then your hiring process, the team lead's leadership skills, and whole team's requirements gathering and code review processes all need some attention.

asdfman1236y ago

There is nothing that you can't express in a kind and constructive manner. Speak objectively, keeping judgments, blame and "you" statements out of it.

"It's been difficult to add new features to the code base because it's grown so much and I think we need to consider a serious refactoring project. The best approach in my opinion is to take the lessons of the past and start over for X and Y reasons."

Will everyone agree with you? Maybe not. Will it work? Maybe it won't. But that's also true if you say "your code sucks," and you won't have to be bogged down in savage internecine code warfare.

logfromblammo6y ago

"It works well enough now to be left alone entirely while we develop a parallel codebase to replace it."

jamesrcole6y ago

It sounds like you're agreeing with me that criticism is important. I never advocated for poorly communicating that criticism, even though that seems to be what you're implying.

To repeat the point of the comment you replied to, it can be much more effective to first work on identifying exactly what the problem is, and if you always expect you can immediately identify solutions it'll hamper your ability to address tough problems.

j / k navigate · click thread line to collapse