Live data from Hacker News

What makes teams successful? 150,000 GitHub projects analyzed

research.gigaom.com

1–10 of 16 posts

Re: What makes teams successful? 150,000 GitHub projects analyzed

#2
This sounds like they are reversing the likely cause-and-effect:

"Even in larger teams, high-impact teams have core and support cadres. A small number — sometimes just one — of the team do the majority of the work, while other, non-core members act in support roles."

Successful projects will tend to bring in a larger number of contributors, most of whom will only make a few contributions. So successful projects will tend to have this pattern. But it is the success of the project that creates that pattern, not the other way around. Lots of projects are not successful, so no one ever contributes to them.

Re: What makes teams successful? 150,000 GitHub projects analyzed

#3
post #2

This sounds like they are reversing the likely cause-and-effect: "Even in larger teams, high-impact teams have core and support cadres. A small number — sometimes just one — of the team do the majority of the work, while other, non-core members act in support roles." Successful projects will tend to bring in a larger number of contributors, most of whom will only make a few contributions. So successful projects will…

Yeah, this whole thing looks like an advert for confusing correlation with causation. Social science can be better than this.

Re: What makes teams successful? 150,000 GitHub projects analyzed

#5

In the teams where one person does more than all others combined, why is that described as "a successful team" rather than "a successful person who succeeds despite a bunch of hangers-on"?

I think that's looking at it the wrong way - you have one person who's talented and/or hard-working; contributing to the majority of the project and the others are the supporters - they're function may be primarily to allow the core contributor to do what he does best and while they take care of the more mundane tasks, provide their knowledge / experience and so forth when needed.

A little bit of encouragement and support goes a long way.

Re: What makes teams successful? 150,000 GitHub projects analyzed

#7
post #5

In the teams where one person does more than all others combined, why is that described as "a successful team" rather than "a successful person who succeeds despite a bunch of hangers-on"?

I think that's looking at it the wrong way - you have one person who's talented and/or hard-working; contributing to the majority of the project and the others are the supporters - they're function may be primarily to allow the core contributor to do what he does best and while they take care of the more mundane tasks, provide their knowledge / experience and so forth when needed. A little bit of encouragement and su…

IME it's usually the other way around about who's doing the mundane tasks. Occasional contributors are most motivated by getting in their one pet feature; it's the project leader who takes on the dull task of cultivating an environment where other people can bolt on that feature successfully.

Re: What makes teams successful? 150,000 GitHub projects analyzed

#8
post #2

This sounds like they are reversing the likely cause-and-effect: "Even in larger teams, high-impact teams have core and support cadres. A small number — sometimes just one — of the team do the majority of the work, while other, non-core members act in support roles." Successful projects will tend to bring in a larger number of contributors, most of whom will only make a few contributions. So successful projects will…

Yes, and double the issue of cause-and-effect with a huge lack of controlling variables. How does the language of the repository and/or number of developers with experience in that language correlate with number of contributors or stars? How about size of the codebase?

I suspect both larger repos, and more popular language choices dominate the "high-impact" projects.

Re: What makes teams successful? 150,000 GitHub projects analyzed

#9
post #2

This sounds like they are reversing the likely cause-and-effect: "Even in larger teams, high-impact teams have core and support cadres. A small number — sometimes just one — of the team do the majority of the work, while other, non-core members act in support roles." Successful projects will tend to bring in a larger number of contributors, most of whom will only make a few contributions. So successful projects will…

Yes, I had the same reaction. I don't think the 'more people participating' observation is so useful.

But I find the related -- and almost opposing -- one to be useful and thought-provoking, and matching my experience and suspicions:

> Even in larger teams, high-impact teams have core and support cadres. A small number — sometimes just one — of the team do the majority of the work, while other, non-core members act in support roles...

> High-impact teams are more likely to be ‘dominated': where the lead member contributed more work than all the other contributors combined.

I think sometimes we think to have success, we have to widely distribute the responsibility/authority. This apparently shows that that's not neccesarily the case, a succesful project can still have a very small number of people (often just one) doing the bulk of the work and with authority (formal or informal) for the direction of the project.

I'm not sure that means you should try to create this as a means of success. On the other hand, it can be hard, but possible, to have a clear direction and goals and stay consistent to them, with "too many cooks".

That successful/influential projects have lots of people participating is not surprising -- success draws contributors. What may be counter-intuitive is that even after this happens nonetheless there are still usually only a few (or often one) person providing the heavy lifting.

Post reply on HN