Hacker News Reader

Hacker News

Comment by dannyr | Hacker News Reader

Parent Full thread

dannyr

Do you really have to ask "why"?

View on HN

I do. It sounds really weird to me. The whole concept of pull requests is that anyone can make them.

Yes. Anybody should be able to make pull requests.

But incentivizing people to make pull requests is a different matter.

How so? Incentivising people to help (via PR) is great. It's like incentivising people to donate blood or similar "everyone benefits" scenarios.

(Yes, of course, blatant spam doesn't help anyone and is disgusting, I'm talking about in general)

The problem is people aren't incentivized to help, they're incentivized to get free stuff. So people who have no interest in helping are doing the bare minimum to get free stuff, which is a much larger number than the people actually wanting to help.

In your scenario, it would be like incentvizing "donating a vial" and hoping it will be filled with blood, but 99% of people are donating an empty vial.

Are you unfamiliar with the problems this has caused?

People who donate blood are pre-screened to ensure their contribution won't be harmful.

Have you been following the events of this year's Hacktoberfest?

Should companies ask for permissions before they employ someone to send pull requests to public repos?

I think the problem was not as much incentivizing pull requests but incentivizing low effort pull requests and outright spam.

"Anyone" does not scale infinitely: it only specifies who might participate, not how many can. Each project will collapse — lose work for one day or more — under a certain sheer quantity burden of pull requests. When people say "anyone can", they don't always mean "everyone do so all at the same time", as happened here. The term "anyone" makes no statement about the frequency of pull requests, and everyone is generally upset about the frequency of pull requests associated with a third party action (Hacktober). Rightfully so: their time is being wasted and their expectations of reasonable behavior are being violated.

I find it hard to consider a flood of pull requests ‘wasting my time’. Don’t look at them. It’s not as if you’re obliged to do anything with them.

Others have a different experience than yours, that is both negative and not improved by the “ignore them” process that works for you. I encourage overcoming the difficulty you describe.

I would argue that having your repository on github is a strong way to say "feel free to contribute". Not saying spam is OK, just that having to ask repo maintainers their authorisation seems superfluous.

Feel free to "contribute and get nothing" is completely different then "contribute get tshirt".

The people who participate are different and pull requests are different.

How are they different as long as they are not spam? Good contributions are good contributions and help, regardless if there's a crappy t-shirt or not in exchange for them

You're clearly not aware of what's been happening the past few days. It's not good contributions or help.

It's the incentive structure. It's effectively Goodhart's Law in action: When a measure becomes a target, it ceases to be a good measure. In this case, the measure is "number of pull requests", and the result of making it a measure is that people created an insane number of PRs that simply adjusted white space in a README or slightly reworded phrases but kept the exact same meaning. Because they wanted the T shirt, because they felt that it validated them as a developer (ignoring that good contributions would have been a much better validation).

Reply on news.ycombinator.com