101–110 of 191 posts
In the end, devs probably need to stop being so easy going. Everyone probably needs to be a little more Linus.
Earlier quoted context omitted.
I think it's fine when you are doing open source for fun, and for other individual users. If a company is basing their livelihood on your library, they should (in my opinion) be supporting it somehow -- either by paying you, or paying someone to keep an eye on new versions. The other choice is they are trusting you to fix bugs and not be malicious, which isn't (in my opinion) reasonable, as you get nothing in return.
Nobody has any right to expect labor from an unpaid person. I get feature requests from people who make money with my OSS work (which I don’t begrudge), but unless I find the work inherently interesting to prioritize then I respectfully ask them to compensate me, make a PR, or file a GitHub feature request so that someone else can. I fix legitimate bug reports for free ASAP because I care about the quality of my work…
I'm very grateful for OSS projects and maintainers but this is an interesting question: is doing something for free a free pass for false promises (advertising)? If I commit to help someone for free I then have a responsibility to show up. If an OSS developer advertise their projects as production ready then they have a responsibility to honor what is promised. Otherwise it should be stated plainly that the project should not be used in production ("This is a hobby project, use it at your own risks"). Sure it's less attractive and can be incompatible with success and resume building, but it's not possible to have it both way.
Earlier quoted context omitted.
Whether a project needs those things is up to its author(s) to decide. If you don't agree with them you're welcome to fork.
If you want a real project then you need those If you just want to share code, that's fine as well, but that's not an OSS project (sure, it can be an open source "project" - but not a project that grows)
outside of a few specific areas; like compilers but that's mainly because hardware is constantly evolving.
Earlier quoted context omitted.
> We need the critical role of Open Source maintainer to professionalize. Do we? I do open source work and maintainership, and am happy that this is outside the control and influence of industry. Must even this that "makes the Internet special" also bow to industry needs?
In my opinion Filo is completely off base. It's the Americanization of Open Source. Notice some of the OSS conferences in the US are big sponsored events by O'reilly, to some extent big recruiting events. Contrast that to FOSDEM, which is completely different. This isn't just an offspring of early hacker culture. This is as offspring of how "socialist" european university systems were. A lot of bigger OSS projects we…
[1] Free Software originated with GNU which was started by Richard Stallman, an American at MIT. Open Source originated in Palo Alto, CA as an alternative (less political, more business friendly) approach to Free Software. There was freely available software prior to the formalization of terminology and licenses from organizations like FSF and OSI, but even then most of it was from developers in the U.S.
[2] Prior to the Internet, most open source projects were heavily concentrated in the U.S. as large scale distributed projects weren't a thing until the 90's with widespread availability of Internet access.
I think the "critical software industry" story is fascinating and if research does not already exist in these areas let's fund more of it: 1. Let's baseline the usage and spread first. What code is used, where and by whom. It ought to be possible to get complete pictures of this most large orgs scan their whole estate constantly - but publishing it academically? 1.a. A central "phone home" FOSS library that everyone…
The google magic words you want are "Biden", "Executive Order" and "Software Bill of Material" :)
Feel free to join one of the working group.
You can also look at what Tidelift does.
Earlier quoted context omitted.
Nobody has any right to expect labor from an unpaid person. I get feature requests from people who make money with my OSS work (which I don’t begrudge), but unless I find the work inherently interesting to prioritize then I respectfully ask them to compensate me, make a PR, or file a GitHub feature request so that someone else can. I fix legitimate bug reports for free ASAP because I care about the quality of my work…
> Nobody has any right to expect labor from an unpaid person. I'm very grateful for OSS projects and maintainers but this is an interesting question: is doing something for free a free pass for false promises (advertising)? If I commit to help someone for free I then have a responsibility to show up. If an OSS developer advertise their projects as production ready then they have a responsibility to honor what is prom…
… PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
I think it’s particularly clear, and that part is even in all capitals.
The article sums up to 'programming is hard', support FOSS devs. Which is hard to find a fault with. The tone is a bit chicken little given that everyone experiences burnout if they have complex jobs that people depend on. In the end, devs probably need to stop being so easy going. Everyone probably needs to be a little more Linus.
FOSS programming tends to be largely recreational or a labor of love, and a lot of the discipline and organization required to make good software is not fun, and will be actively avoided in recreational projects, especially for people who do the "not fun" parts of projects in their day job.
A really successful project has to have someone (or several someones) at its head who can organize everyone and keep things moving in the right direction. Part of this is being willing and able to say "no" even if it means someone may walk away from the project.
If you want to produce really, really good open source software, provide a stipend for the programmers so they don't need to keep day jobs and pay market rate for a small number of managers with the skill to organize, the charisma to get agreement from the team, and the knowledge of when to tell people "no".
Earlier quoted context omitted.
Nobody has any right to expect labor from an unpaid person. I get feature requests from people who make money with my OSS work (which I don’t begrudge), but unless I find the work inherently interesting to prioritize then I respectfully ask them to compensate me, make a PR, or file a GitHub feature request so that someone else can. I fix legitimate bug reports for free ASAP because I care about the quality of my work…
> Nobody has any right to expect labor from an unpaid person. I'm very grateful for OSS projects and maintainers but this is an interesting question: is doing something for free a free pass for false promises (advertising)? If I commit to help someone for free I then have a responsibility to show up. If an OSS developer advertise their projects as production ready then they have a responsibility to honor what is prom…
Maybe it's important to you and maybe people you support -- but that has nothing to do with your right to make demands against someone who put their work out there for free.
If you need a guaranteed support model, you either do it yourself or hire a company to provide that for you. I've used _supported_ commercial software that's garbage, and I've used FOSS software which has been absolutely rock solid. Also, commercial or not, just because it's supported doesn't mean the problem will be fixed or a feature added.
If you aren't directly paying someone to be there for you, then you're trusting goodwill and/or the community at large to fill that role for you. If you need that assurance, pony up to one of the consultancies like IBM which will support "X" for price "$Y". That support you're referring to is available, at a price -- but doesn't necessarily involve the original contributors.
I've had far better luck with FreeBSD on servers for the past 20 years than I have with Microsoft Windows, but I also understood the deal: "Fix it yourself, or submit a bug report and see if anyone else wants to work on it."
The corporate software method: "Pay us lots of money to look into the bug report, and maybe we'll fix it, maybe we won't -- but you'll need to pay hourly until we determine if its a bug. Then MAYBE we'll refund your money."
The primary difference is that OSS gives you the source and the ability to fix problems on your own, without any involvement from the original contributors/projects. You can also hire any number of other companies to provide support. With a traditional commercial model, you're entirely at the mercy of the company which originated the software, and you likely have no rights to the source or tools which would enable you to fix the problem yourself.
I think the "critical software industry" story is fascinating and if research does not already exist in these areas let's fund more of it: 1. Let's baseline the usage and spread first. What code is used, where and by whom. It ought to be possible to get complete pictures of this most large orgs scan their whole estate constantly - but publishing it academically? 1.a. A central "phone home" FOSS library that everyone…
[0]: https://en.wikipedia.org/wiki/American_Society_of_Composers,...