> I have been part of the free and open source movement for the entirety of my 25+ year career in software. In the late 90s, I was one of the first people to advocate in my org for switching from proprietary tools like the Intel C compiler to open source alternatives like gcc.
I’m happy to give the author the benefit of the doubt on their contributions to FOSS beyond what’s stated in the article, but taken at face value, they are trying to establish a tone of authority by establishing that they championed… using someone else’s free or open source software rather than paying for something? That might be interesting if led to a frank discussion on the value and cost of FOSS vs contribution or funding vis-à-vis general sustainability, but it’s not.
Second, the term “open source startup” makes my skin crawl. If you’re a business focused on making profit, you’re just a commercial entity that happens to put out some open source stuff, using it to lure customers to your actual business. While personally I’m wary of this model as a potential trap, whatever, it’s fine. But this is not the only way FOSS gets done, and I don’t see any rethinking going on in the article.
> they are trying to establish a tone of authority by establishing that they championed… using someone else’s free or open source software rather than paying for something?
At that time, there was a lot of mistrust of FOSS (and the GPL in particular) in commercial organisations. That extended to even _using_ FOSS, and particularly something like a compiler, whose output "might be contaminated" by the FOSS licence.
That changed, slowly, in part because there were dedicated advocates who addressed people's concerns, showed them the actual licence text, etc, etc. It was a valuable contribution to the success of the concept of FOSS beyond just developers.
From today's perspective, I understand your concern, but perhaps this might help explain the OP's claim.
> At that time, there was a lot of mistrust of FOSS (and the GPL in particular) in commercial organisations.
I know, I’m an old enough programmer to have gone through the same thing in my career. Nonetheless, it’s a weird and kinda irrelevant credential to pull out in an article that purports to rethink open source licensing.
It's a rethink from the starting point of "everything should always be open source"
I think what's mentioned here is a pragmatic approach for software-centric businesses who want to be open-source but have realized that it doesn't really work without opening yourself to getting "Jeff'd" (ripped off by Amazon).
I was going to say that no one said everything had to be only open source, but of course one person did say that, but hardly anyone else.
And getting Jeff'd is not a problem. If you care about getting Jeff'd, then you simply want to sell software instead of create software. So go forth and sell, and don't pretend all the people who wrote those open source licences and live by them just somehow don't get the realities of life. They do. In fact they get things people like these guys don't. It's fine to have that weakness and just decide to sell under the traditional 80's terms, but don't pretend it's required and forced upon you and anyone else who chooses not to do the same somehow doesn't also live in reality.
We haven't talked enough about the second-order effects of open source on the economics of programming as a profession. I'm glad to see that starting to change[1].
Open source makes it harder to make a living as a programmer (at least an author of libraries, tools and infrastructure). Keeping server utilities closed source while still allowing clients to have full access to the source for client tools and self-hosting utilities seems like a pragmatic approach to the user-centric principles of open source.
This way, companies can continue to maintain some competitive advantage with their own investment into closed source code while still enabling users to own, control and contribute to software that runs on their machine.
The crucial element here is maintaining low switching costs by allowing a marketplace for _real_ competition on quality and cost of compatible services (as well as allowing self-hosting). This is the kind of software freedom that actually benefits users.
For a short time I was all about being a FOSS zealot using M$ on email signatures and such, because I had a confortable university student life where most bills were paid by someone else, then I realised what it means to have to pay bills for ourselves.
People are realising that PD and Shareware make more sense when one wants to make a living from software, the alternative being putting everything behind SaaS walls.
Not every use case can be done selling hardware to go alongside the software, consulting, conference talks, selling books (which are often pirated by those that should donate to FOSS)
The "die-hard free software enthusiast" inside me made it through the whole article, and was already low-level triggered, and then really got shot into the stratosphere with this down the bottom.
I'm not saying I've a better idea for how to reconcile software freedom, noxious capitalism, and the necessity of people who love software development and related areas getting rooves and bread, but...
The article does read a bit like an oil company telling you they love the planet, and that's why they have to keep mining it for oil as they work hard on renewable energy on the sides. "We're sorry".
No. But it's definitely much closer to what's described in the article than binary blob drivers.
Complaining that a company only makes available the source to everything you'll ever want to run in your computer, but not the tools certify it for avionics (SQLite), or to scale it horizontally to 100s of nodes (Fermyon Spin), is not the same as not giving you the source to display drivers.
I'm not complaining about anything (other than the article, but not because they want to sell their software).
But is a fact that a lot of companies that want to straddle both sides of the fence, the open side is kinda crap. It's also true that a lot is not crap, but so what?
Another random example I happen to be familiar with, a long time ago a company essentially took over hylafax, and what do you know, ever since then the open parts got practically no useful or valuable advancement, and ifax commercialized such trivial to implement but end user valuable nicities as being able to associate a fax job with a db record. hylafax+ was forked and has been the good useful version ever since.
ifax didn't really hurt anyone since the fork was possible and actually happened, and has been great, but they didn't help either. they provide essentially no value to the open source ecosystem in trade for their ownership and commercialization of hylafax. I don't know if they ever tried to write a blog post like this one though. If not, then they were at least more honest, ish. It's not that honest to claim to be managing and caring for some open source project, while for some mysterious reason not accepting prs for features that compete with ones you sell in your commercial version of the same thing, for 20 years.
I observe that there is crap, or that a certain set of incentives produces crap. So what?
What makes it not complaining is I didn't write a blog post about it to try to convince anyone of anything, and I'm just doing my own thing and don't care about them doing their thing, until they tried tondo more than that and also tried to convince everyone else to applaud them for it.
But if the question is raised to evaluate the situation as part of a discussion, this is my evaluation. That isn't complaining even if it is critiquing.
Here is what the real conundrum is about open vs proprietary software:
No one has to sell software to have roofs and bread, even while working on software as your living. You can live very comfortably, better than 80% of everyone around you, and all without any blown out knees, backs, ears, lungs, etc. Just doing support and custom work. You can have all the homes and cars and vacations and health insurance and kids educations you want, and no stress with cash in the bank and complete security.
But getting paid for work instead of paid for copies of work doesn't scale. In fact the fact that it doesn't scale is ecactly why there is enough to go around and everyone can have that life.
But you can never get a $billion unless you can do one days work for a cost of $1k and sell a million copies of it for $5k each.
No one needs that, and no one deserves any pity that they want that and can't get it, EVEN IF someone like amazon can, using something you wrote and gave away no less.
Amazon can make a billion off of the same open source thing that you can't, but so what? Amazon also makes a billion off of driving their trucks on the road out in front of my house that I paid for with taxes. It doesn't change the fact that it's incalculably valuable to me that the road is there and usable by anyone. It has to be that wide open or else it's actually not valuable to me either, both directly and indirectly.
It's not a perfect analogy because they pay taxes too, except then again no they don't really, like all big companies they pay a tiny fraction of what they should and what you and I pay. And the maintenance of the road is more like the servers and stuff, the core value software would be the engineering knowledge how to build roads and maybe the road laying machines and the formulation of asphalt. Whoever did that work was just plugging away at their regular job and got a days pay for it. And so what?
> I have been part of the free and open source movement for the entirety of my 25+ year career in software. In the late 90s, I was one of the first people to advocate in my org for switching from proprietary tools like the Intel C compiler to open source alternatives like gcc.
I’m happy to give the author the benefit of the doubt on their contributions to FOSS beyond what’s stated in the article, but taken at face value, they are trying to establish a tone of authority by establishing that they championed… using someone else’s free or open source software rather than paying for something? That might be interesting if led to a frank discussion on the value and cost of FOSS vs contribution or funding vis-à-vis general sustainability, but it’s not.
Second, the term “open source startup” makes my skin crawl. If you’re a business focused on making profit, you’re just a commercial entity that happens to put out some open source stuff, using it to lure customers to your actual business. While personally I’m wary of this model as a potential trap, whatever, it’s fine. But this is not the only way FOSS gets done, and I don’t see any rethinking going on in the article.