Literate Programming - We Forget That Open Source is Made of People • Hacker News Board

Re: Literate Programming - We Forget That Open Source is Made of People

#11

Post by bonzoesc » Fri, Aug 12, 2011, 5:45 PM UTC

mauricio wrote: post #4

sstephenson wrote: post #2

>> "But guess what: competition sucks. Cooperation is much better. One of the biggest problems with competition is that it means that somebody has to lose. For every winner, someone is a loser. That person is going to be hurting. In the case of open source, someone has spent (possibly) years of their life working on something, and now, all of that is gone. That shit stings." A thoughtful post by Steve, but I don't ag…

What did Wayne say when you proposed to make these changes to RVM itself? It seems like if you wanted to make the pie bigger, and you just want ideas versus egos to rule then you could've proposed to improve RVM.

Open-source software is about making software, not about staking claims to product areas and stroking egos. Linux Torvalds didn't ask for Andrew Tanenbaum's opinion before releasing Linux, and John Resig didn't send Sam Stephenson a pull request for Prototype before releasing JQuery.

Re: Literate Programming - We Forget That Open Source is Made of People

#12

Post by joshsusser » Fri, Aug 12, 2011, 5:48 PM UTC

sstephenson wrote: post #2

>> "But guess what: competition sucks. Cooperation is much better. One of the biggest problems with competition is that it means that somebody has to lose. For every winner, someone is a loser. That person is going to be hurting. In the case of open source, someone has spent (possibly) years of their life working on something, and now, all of that is gone. That shit stings." A thoughtful post by Steve, but I don't ag…

I agree. Look at nginx vs apache - competition with nginx has done a lot to improve apache. Merb did more to improve Rails than anything else has. New projects with different design centers give you the flexibility to try out approaches that wouldn't be possible in the old project.

The thing that sucks is people taking sides and getting all bent out of shape over things. Even the people behind "competing" projects can cooperate in their work. Ruby MRI, JRuby, Rubinus, Maglev - all these supposedly competing projects created collaboration on RubySpec and other shared pieces of technology. When John Trupiano and I collided over Rack::Rewrite vs Refraction, we just talked on IRC and had a friendly chat about our different approaches and motivations, and how features in one project influenced the other. Nobody got mad or hated anyone over it. Even major political rivals can be close friends - look at James Carville and Mary Matalin (or don't, if you value your peace of mind).

Re: Literate Programming - We Forget That Open Source is Made of People

#13

Post by petercooper » Fri, Aug 12, 2011, 6:13 PM UTC

mauricio wrote: post #10

petercooper wrote: post #6

Earlier quoted context omitted.

Why assume he did? There's no law or even expectation that you can't create new projects. Not everyone is gifted at the diplomacy or negotiation required to work on existing projects. And, of course, many people want to have their "own" projects. There's nothing wrong with that at all. If everything was just an improvement of something old..

There's no law for many things in this world, but there certainly are expectations. There are many projects with widespread use that have become core projects in the Ruby community (Rails, bundler, RVM, etc). It's reasonable to expect that issues or design decisions get discussed on the project mailing list before you try and divide the community.

It's at least no more divisive than to have Passenger vs Thin vs Unicorn vs Heroku vs WEBrick vs Pow for the serving end of things. Or RSpec vs Test::Unit vs MiniTest vs Shoulda vs Riot for testing. Good things have come out of all of these additions and they were not all fuelled by discussions.

In any case, to defend against claims like yours (which are common, it seems), it does seem wise for anyone wishing to create something new to post to existing mailing lists with their ideas. Once the maintainers say "No" to your idea to rip out 90% of their system, you can then create your new project and have the valuable mailing list link to say "Look, I tried."

Re: Literate Programming - We Forget That Open Source is Made of People

#14

Post by saurik » Sat, Aug 13, 2011, 1:15 AM UTC

joshsusser wrote: post #12

sstephenson wrote: post #2

>> "But guess what: competition sucks. Cooperation is much better. One of the biggest problems with competition is that it means that somebody has to lose. For every winner, someone is a loser. That person is going to be hurting. In the case of open source, someone has spent (possibly) years of their life working on something, and now, all of that is gone. That shit stings." A thoughtful post by Steve, but I don't ag…

I agree. Look at nginx vs apache - competition with nginx has done a lot to improve apache. Merb did more to improve Rails than anything else has. New projects with different design centers give you the flexibility to try out approaches that wouldn't be possible in the old project. The thing that sucks is people taking sides and getting all bent out of shape over things. Even the people behind "competing" projects ca…

Most "competing" projects are not friendly: marketing materials put out deride the alternative as "slow", "buggy", or what-have-you (some of the stuff put out about my project simply makes me want to quit it is so nasty).

Interesting, then, that, for whatever reason, I do /not/ see this kind of behavior with the various Ruby implementations: most people I deal with in that community actually seem to enjoy the various versions existing and work together. In this case, I wouldn't even call it "competition": I'd call it "cooperative experimentation".

But this argument breaks down when you look at the nginx vs. Apache example: proponents of nginx tend to just pummel Apache with pain, even when many of the claims aren't even true (which is extra stupid as I don't even think it is correct to believe that nginx and Apache are trying to solve the same problem).

Re: Literate Programming - We Forget That Open Source is Made of People

#15

Post by saurik » Sat, Aug 13, 2011, 1:18 AM UTC

sstephenson wrote: post #2

>> "But guess what: competition sucks. Cooperation is much better. One of the biggest problems with competition is that it means that somebody has to lose. For every winner, someone is a loser. That person is going to be hurting. In the case of open source, someone has spent (possibly) years of their life working on something, and now, all of that is gone. That shit stings." A thoughtful post by Steve, but I don't ag…

The pie gets larger, but the use cases do get fragmented: if someone is really excited about some new feature or implementation mechanism, it is usually not the case that whatever "stalwart" project that is out there is unwilling to include it, and yet people assume this constantly and start a new project from scratch, or (a greater insult) fork the other project, to include that feature. Users and developer alike now have to choose, and that really does suck.

Re: Literate Programming - We Forget That Open Source is Made of People

#16

Post by saurik » Sat, Aug 13, 2011, 1:22 AM UTC

One of the things posted in this article is a comment on GitHub's popularization of the concept of "fork": to me this was a mistake made by those developers, and they should have used the actual term from git, "clone" (which, in fact, Google Code does in their implementation of this same concept).

The thing that makes git great is not that it is easier to /fork/ other people's work: we don't think of all the different versions of Linux out there maintained by different people in the hierarchy (Andrew Morton, Alan Cox, etc.) as "forks", and no one considers them "competing".

But now, GitHub is teaching this new breed of open source developer that there is no difference between using a distributed version control system to collaborate on a project and forking someone else's efforts to compete with them: that really does suck. :(