Open mrjoelkemp opened 8 years ago
@rnicholus There are likely a few paths to long-term open source development. The most familiar approach is to make a living off of one's work.
The ideal, in my mind, is for an author of a successful project (should they desire to continue working on that project in the long term) to make a living from that project.
We need to avoid the current trend of working a full-time job and maintaining increasingly popular projects. Open source projects become more and more demanding – putting strains on the author's personal and professional life. When (not if) that author burns out, the project, users, the community around the project, and the author (and his/her family) suffers. For "prolific" authors, the damage is much greater as their creative genius is gone – no longer delivering value and advancing the state of the art.
An alternative to seeking profits is for an equally shared load of work spread across many authors. Realistically, this is a pipe dream; most projects seem to have less than a handful of devs willing to lend a hand.
Any thoughts on alternatives to financial concerns that allow authors to work on projects for the long haul?
Getting a project into an open source foundation.
@mikesherov For those devs that want to work on their project for the long term, would the foundation sponsor them?
Or would the foundation only help keep the project afloat by dedicating foundation devs to chipping in?
More generally, how does a foundation make development more sustainable?
EDIT: also, how does one get their project into a foundation? Is there a set of characteristics shared about projects in the jQuery foundation (as a concrete example)?
the current trend of working a full-time job and maintaining increasingly popular projects
This is not sustainable, but I wonder if this is more of a character flaw than a problem brought on by the rigors of maintaining a popular project. In my case, I suspect the former is true. Had I never been exposed to OSS development, I would have found something else to obsess over.
an equally shared load of work spread across many authors
This does seem like a pipe dream, as you went on to say. But why? Is it because software developers don't want to help out? Is it because software developers don't want to work on OSS? I believe the answer two both questions is commonly "no". Instead, many developers may want to maintain a work/life balance. Work on software as part of their job, and enjoy their life during off hours. I assert that many OSS projects are maintained by those who are comfortable spending most of their free time working on tech projects (until they burn out, at least). And I don't think we need to encourage the "balanced" developers to work on OSS during business hours. They'd probably love to do this. The problem is with managers, directors, PM, and CEOs.
The article you shared with me on Twitter makes some great points on this front. Non-developers that "call the shots" are often not aware of the importance of OSS. Maybe they don't realize how quickly their flagship products would crumble if the open source building blocks disappeared. Maybe they don't understand how much of what they produce is, more or less, glued together open source libraries. It seems that the inculcation of the importance of OSS is left mostly to the development community. We're preaching to the choir! Instead, we need to put together compelling presentations and business cases for the CEOs, directors, PMs, and managers. Then, we might see a shift in the landscape.
Typically, open source foundations do several things:
jQuery Foundation projects are really anything in the jQuery ecosystem, the dojo ecosystem, the browser ecosysyem, and the tooling ecosystem. Really, anything that helps devs get stuff done.
The Foundation doesn't invest in projects. They have projects join, and separately, have donors join.
Getting a project into the jQuery Foundation involves starting a conversation with the Foundation about joining. It gets more complicated from there but always just starts with a conversation. Esprima joined as a result of a chance meeting between myself and Ariya at a conference where I was trying to pitch CST to him. We got started talking about how Esprima really needed a neutral home. Rest is history.
I would also say that lots of projects burn out because the necessary work isn't being done to attract new contributors to projects. Lots of projects run as BDFL (benevolent dictator for life), and then crash when the BDFL realized "for life" is a really long fucking time. Successful projects figure out ways to transition, like jQuery when Resig stepped aside, and Babel when sebmck stepped aside.
Maybe we need a section dedicated to OSS foundations. I'd like to see if @hlship would be willing to share his experiences and thoughts based on his long-time association with Apache (via Tapestry). I suspect he has some interesting things to say.
I added a section on Software Foundations and a link to Mike's answer in the FAQ. Thanks @mikesherov for the clarity around foundations.
I'll respond to BDFL and outreach later tonight.
On Tuesday, August 9, 2016, Ray Nicholus notifications@github.com wrote:
Maybe we need a section dedicated to OSS foundations. I'd like to see if @hlship https://github.com/hlship would be willing to share his experiences and thoughts based on his long-time association with Apache (via Tapestry). I suspect he has some interesting things to say.
What a relief now that I've stepped back.
I could write quite a bit about mistakes I made but the summary is: I treated the effort as being about the code but it's about the community.
I still like the theory of the ASF but I, for one, am not cut out for the process.
The key challenge is that everyone has a strong opinion, poorly expressed (via hard to follow e-mail threads) and nobody wants to do the work.
I'm much happier with my stable of one+ person little libraries. Help accepted but not needed, and no governance or restraints.
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/mrjoelkemp/awesome-paid-open-source/issues/2#issuecomment-238753793, or mute the thread https://github.com/notifications/unsubscribe-auth/AADNtOY82Do5E_73WdoclIofxto3bwYuks5qeT_lgaJpZM4JgqzY .
Howard M. Lewis Ship
Senior Mobile Developer at Walmart Labs
Creator of Apache Tapestry
(971) 678-5210 http://howardlewisship.com @hlship
@hlship Good stuff Howard. I followed the Tapestry mailing list for years, and I suspect you have enough perspective on the ASF, especially due to your work with Tapestry, to merit a long-form article on the subject. I'm sure a lot of people would be interested in this, if you are willing and able to write candidly about your experience.
Nope, I wouldn't do that. I don't think I could do a good job of it, and it would not be fair to the ASF otherwise.
Being inside the ASF makes things harder ... it's dozens, perhaps hundreds, of cooks; people who'll drop in uninvited to tell you that you're doing everything wrong.
In addition, I have qualms with the concept of absolute meritocracy because everyone's contributions and insight are NOT equal. Effectively, the benevolent dictator expresses an opinion, and the remainder of the PMC naturally falls in line. 90% of the time that's a good thing, but sometimes the benevolent dictator needs to be reigned in, especially if (like me) he loves the code TOO much.
My intent with Tapestry was to do the heavy lifting of the "engine" and let the rest of the community and committers work on everything else. That did not work out, and my 60/40 hindsight (see, I'm not the right person to write that article) informs me of a number of problems, some technical but largely social, for why that happened.
At this point, Tapestry is not the technology I'd use for the kind of web-based UIs I might build: I'd be using Clojure on the server and ClojureScript w/ React on the client.
On Wed, Aug 10, 2016 at 10:49 AM, Ray Nicholus notifications@github.com wrote:
@hlship https://github.com/hlship Good stuff Howard. I followed the Tapestry mailing list for years, and I suspect you have enough perspective on the ASF, especially due to your work with Tapestry, to merit a long-form article on the subject. I'm sure a lot of people would be interested in this, if you are willing or able to write candidly about your experience.
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/mrjoelkemp/awesome-paid-open-source/issues/2#issuecomment-238946996, or mute the thread https://github.com/notifications/unsubscribe-auth/AADNtBIGMZ-_iTpf2cAXcahUCSyB-f_7ks5qeg8hgaJpZM4JgqzY .
Howard M. Lewis Ship
Senior Mobile Developer at Walmart Labs
Creator of Apache Tapestry
(971) 678-5210 http://howardlewisship.com @hlship
I think you've already provided a lot of useful thoughts. Thanks Howard.
I'm participating over at https://github.com/sfosc/sfosc which is a community which offers a discussion platform and resources surrounding Sustainable Free and Open Source Communities. In the SFOSC view, funding is only a part of whether such a community is sustainable.
To name a few other aspect being discussed:
So in my opinion, all of this is important to consider. The monetized open source databases going head to head with cloud providers right now is a great example of why a profitable/sustainable business can still be a massive risk for people in the community by being dependent on the survival and benevolence of a single for-profit company.
Like discussed here https://github.com/sfosc/sfosc/issues/63 And here https://github.com/sfosc/sfosc/pull/42
From https://github.com/mrjoelkemp/awesome-paid-open-source/pull/1#issuecomment-238726783