Comment by hoaw
7 years ago
> Though Rich is right, it pains me to read this because it is indicative of some disputes in the clojure community.
I am genuinely curious why you, or anyone else, would think he is right. I can see why people would agree or why he wants to do things a certain way, but to be right you have to have arguments backing up what you are saying. I don't see that in this post. Am I missing something?
> I am genuinely curious why you, or anyone else, would think he is right.
Because Rich has the rights to do what he wants with his time and money. (Like everybody else.) It is completely within his rights to establish his own governance and contribution model for his version of Clojure. At the very least, I don't see an argument why it's NOT within his rights to do so.
Of course, there are several corollaries. The first is that nobody is forced to use or contribute to Clojure. The second is that, given that Clojure is Free Software, it's possible for someone else to pick up the baton, make a fork, and run it their own way. (egcs and XEmacs both come to mind as examples where someone did this and it wound up improving the original.... so there _is_ precedent.)
I wouldn't feel terribly bad if someone made this fork, but I'm not going to do it myself. Unlike Rich, I'm not willing at the moment to make the personal sacrifices that'd be required to make this sort of contribution. Neither, apparently, is anyone else.
So Rich gets to make the rules. Because he (and Cognitect) have paid the dues. Whether that's good for Clojure in the long run remains to be seen. (But honestly, my semi-informed take on the ecosystem is that it needs library maintenance much more than it does core language improvements. The core language has been an excellent choice for many purposes for years.)
I don't know much about the Clojure community... but this article says "Open Source Is Not About You." If this is the generic "you," ie the random user that's coming in and complaining about something, then sure.
If this is a veiled criticism at a major contributor to the community, then I think it's disingenuous. If you're trying to foster a community around your OSS project/product (and Cognitect is doing just that) then you take on a set of obligations to that community, especially those who contribute code to it in good faith.
I understand that this is his position, what I don't see is his reasoning.
If everyone is free to do what they want with their own time and money, they are also free to condition their participation in the project. As far as I can tell that is not what he is saying though. What he is saying is that you are not entitled to anything, meaning that you can't expect anything for your participation.
If he had just said that he runs the project how he want and people can "take it or leave it" that would be another thing. But as I read it he is specially questions the right of others to question the project. Essentially saying that they don't have that right. It is this position I don't see him backing up with an argument.
> Because Rich has the rights to do what he wants with his time and money. (Like everybody else.)
This feels like a non sequitur. Having a right to do something is completely orthogonal to whether you should do that thing.
> Having a right to do something is completely orthogonal to whether you should do that thing.
Yeah... there are really several conversations that could be had:
1. Rich deserves credit and respect for personally taking on a cost that ranges into the hundreds of thousands of dollars to fund Clojure. Very few people do this, and even fewer bring the taste and judgement Rich brought to Clojure.
2. Rich/Cognitect currently own the Clojure governance model and they can do with it what they want.
3. Maybe the current governance model isn't what's going to bring Clojure the next 20-30 years of viability.
My take on it essentially this:
1. Yes. I wouldn't/couldn't do it myself. (And I've started down the path at least once.)
2. Yup. Maybe it will be what the platform needs to grow and thrive. Maybe not. Maybe it will be attractive to users and developers. Maybe not.
3. At some point, this will probably be true. Given the history of companies and projects, this shouldn't be a surprise.
I mean, he is right about open-source only being a binding license for the source, and in this case, the Eclipse 1.0 license in no way entitle users too any kind of support, decision making authority, opinion, voice, or any other such thing.
This is just fact. You can go read the license to learn more about this: https://opensource.org/licenses/eclipse-1.0.php
That part of what Rich is saying is pretty much indisputable.
The second part, is related to what is best for Clojure. And the argument is simple, Rich says what is best for Clojure is a thorough review process of all changes to its core and standard libs, with a very high bar towards contributors and their contribution. His argument is that this has worked so far, and has created the solid piece of software that is Clojure today. Thus its own success is proof that it is a good enough process and is good for Clojure.
The argument against is that contributors find it too hard and too much work and thus very little contributors make it through the process. Though I didn't really see them mention any alternative process suggestion. It seems they were mostly wanted their patch to just be merged in without challenge.
Because when you invite me to your house, I do not gain the right to demand you rearrange the furniture. You built that, you put things in their place. If I want something for it I can only make a case for it, and that case hinges on convincing you.