Hacker News Reader

Comment by hartator | Hacker News Reader

I will still think odd that’s someone who is a professional hasn’t found one bug in an open source lib that is worth a PR or even an issue in a decade of work.

View on HN

I don't know about a decade, but what percentage of people with >10 years of coding experience do you believe have submitted a public PR for an open source library?

Based on the people I've worked with (both great and mediocre programmers), I think that is highly uncommon.

Very true. I can't sadly think of a single instance where any of my colleagues in the last ten years would've contributed a bugfix upstream. It would always be a workaround or creating a local copy of the code in question to customize it.

Most software engineers will be on a commercial contract that doesn't allow contributing to open source projects. Even if you issued a PR or raised an issue, it would have to be on a throw away account and you wouldn't include it in a CV.

Most [citation needed]

As posted in one of the comments in this thread:

https://www.joelonsoftware.com/2016/12/09/developers-side-pr...

There are also a lot of software developer's that have contracts where they can contribute to open source but it has to go through legal. Which makes it to much of a pain to actually be worth doing.

And if they did, what value do you derive from it? Fixing a small logic error or one-off error gives you about as many data points about someone as them having absolutely nothing on github.

That they care enough to give back a little? Even from a pure selfish perspective, I would still open a PR or an issue to double check that my logic is sound.

You're selecting against employees working at Fortune 500 who are limited from participating in open source, unless filling some horrifying paperwork.

At my former company there was a three step process for contributing to open source:

- get approval from VP to spend time on that

- get approval from legal (they checked the license and if the company has any liability from that code not working, this alone took a couple months)

- get approval for release of code from a so called open-source committee (basically checked if your contribution doesn't damage company's reputation)

It was easier to set up a fake account and get things fixed this way. But no way I could do that on company hardware or time.

We're talking about contributions in one's own time, outside of work and not using their resources??

Are we?

> I will still think odd that’s someone who is a professional hasn’t found one bug in an open source lib that is worth a PR or even an issue in a decade of work.

It seems like the person is surprised how nobody has any contributions from just their professional work. Naturally unless I had no life outside of programming I would spend the vast majority of my time doing my 9-5 stuff.

Not allowed everywhere. Any code I write ay any point of time is owned by my employer legally and in an EU country. This is because I work at a large consulting company that touches on any and all business fields.

Free Github private repos has been a blessing for just storing modifications to existing projects.

In NA, it's common for software dev. employment contracts to prevent you from working on software in your own time, and/or to claim moral rights to anything you do work on, regardless of field or business interest overlap.

I have no idea about enforcement, and most coworkers I've had have ignored those clauses.

I've not heard of that before - is that a legal employment restriction. Not that it necessarily has to be to stop people, of course.

I'm going to guess you're in USA, is it legal to stop phone workers using the phone outside work. Seems like not allowing people to work on things directly related to their actual work (due to IPR claims by the employer) would be a thing, but no OSS at all?

https://www.joelonsoftware.com/2016/12/09/developers-side-pr...

I thought we were talking about contributions at work? I find bugs in OSS libraries from working 8 hours a day on software (including in glibc and python).

My last company did have a policy that you are prohibited from participating in OSS, unless getting a bunch of approvals per project or per patch. I'm risking my job sending a PR.

It's work related, it should be fixed at work. (I certainly don't want to patch things in my spare time because the company forbids me to work at work).

P.S. I am in London. USA would be worse.

I’m a decent developer. I’ve shipped code, led projects, and worked at scale. My employer owns everything I code. I’m not supposed to, legally, even write code during an interview if I’m looking for another job as they own it. I can contribute or create open source but everything must be approved by my org first. For many, that hurdle alone is too big to justify working on open source. I got a talking to for contribution to documentation on a Go project out of Facebook without proper paperwork being done and “exposing risk to my employer.”

I did find a bug in svelte but not sure if it was worth a PR. I cloned their repo, tracked down the issue, learned how their codebase worked, fixed it, wrote some tests for it, wrote an elaborate issue and pull request explaining the issue and nuances. It took around 4-5 hours over a couple of days even though the fix was a one line diff. The PR was ignored for 6 months. I was considering bumping it but then I read in another PR that it's bad manners and just decided to remove it instead.

This. I give people lots of room, I don’t even care if it’s in an ancient sourceforge repo, I still like seeing this kind of professional exploration and participation in new hires(like since 2010 at least). Not a requirement but a curious absence for me that suggests a more deeper interview is worth having.

I have worked in software engineering for over 15 years, and I have never contributed to open source as my employment contracts never allowed this. According to my contract, and this is typical for many commercial companies, any code I write is owned by my company, even if I would write it in my own time. If I would contribute to an open source project, there would be legal implications for both me and the open source project maintainer.

I hear you, and it’s why I also only consider it a curious marker and not a detractor. Nowhere did I say I wouldn’t employ someone, only that it sparks more curiosity in me.

Depending on your job, it might be a lot of paperwork to get permission to issue that PR.

When I find a bug in the tools I use, I will check to see if someone else has reported it and whether it'd be of any help to add my 2 cents.

It's almost always yes for the former and no for the latter.

Reply on news.ycombinator.com