We need to talk about funding
octoprint.org
octoprint.org
It has resulted in at least one new monthly contributor and hope others do the same.
I wonder if the decline in donations has anything to do with Prusa refusing to properly support Octoprint in favor of their own half baked, awful competitors that are no doubt in furtherance of their quest for SaaS rent seeking (BambuLabs fear? jealousy?).
The MK4 and XL are barely functional with Octoprint and Prusa has indicated they intend to keep them this way, as they did with the Mini. That's a large and growing market now not really supported by Octoprint.
They instead want you to use Prusa Connect, their currently free farm management software that requires internet access and all of your data to manage your local network of machines. No doubt this goes non-free the second any momentum is achieved.
I say 'want you to use' because their local network solution, PrusaLink, is a barely maintained skeleton of a project that replicates a tiny portion of Octoprint's functionality, poorly (4kb/s uploads! 27 char file name limits!), while missing many critical features, has no plugin support and many many bugs. No doubt the prioritization within Prusa is Cloud and not the local functionality you actually need and your machines are capable of.
By reducing Octoprint functionality to the level of their own rudimentary offerings on their newest flagship printers they certainly make the case for using Octoprint far less compelling and paying for it even less so.
Octofarm shutting down also probably didn't help things either.
Or Bambu Lab. I switched not long ago from a Creality printer to a Bambu Labs printer, and judging from the comments I see online I'm definitely not alone -- it's been very popular this year.
Even if you set aside Bambu and Prusa (who yeah, they use Marlin, but their fork is Theirs and they make little effort to contribute back), it's hard not to think that Klipper has developed better options than both Marlin and OctoPrint internally. And, from where I sit, it seems that the leading edge--the sort of folks who are likely to go spring for a SBC to drive their printers--are moving towards a Klipper-based stack at a nontrivial pace. The original reason I went to Klipper wasn't for speed or input shaping or anything like that--it was first for online printer config but a close second was Mainsail. Mainsail just works better than Octoprint - it handles multiple printers better because of how it's decoupled from Moonraker (Klipper's API layer), it's a nicer interface, and of course it gets comfortable with things like Klipper's printer.conf editing, too.
I can't speak to the Prusa stuff, I don't own any of their printers, but the Bambu experience inside the slicer is...actually pretty good, too. They figured out some good stuff. I'm pretty happy with my Klipper setups alongside my Bambu printers (though Bambu-the-company seems to come with some really loaded downsides, and that is a frustrating thing), and I just don't see a reason to use OctoPrint ever again.
(Now, the large manufacturers taking Klipper and not even upstreaming changes back--that is a hill I'll gladly charge, it's real gross!)
I went Klipper a little reluctantly at first when I built a semi-custom printer and the community seemed to all lean that way. As soon as I “got it”, I switched away from Marlin and Sailfish on my other printers to now be all-Klipper. There’s nothing “wrong” with octoprint per-se, but mainsail is indeed more usable for me in the Klipper ecosystem.
It really seems like Prusa should have done something with Octoprint than trying to reinvent the wheel on their own.
Now Octoprint is left in this shrinking space between Prusa/Bambu and Voron type printers. Bambu from what I can see is the best "click and go option" (albeit going against the open source ethos to a degree) then you go into Voron realm and you're definitely using the Klipper stack there.
When I'd gotten back into it I thought "am I really going to do put this all together, why not go an XL". 3 hours of YouTube videos later it was "no way.. sticking with Voron and going all in..."
They have, in fact, stated the opposite, that they are actively working to address the issues:
If Prusa machines only work well with Prusa connect, then I won't be purchasing any more Prusa machines.
I see you're seeking corporate sponsors. Awesome -- I think this can offset a significant amount of your expenses and pave the way to sustainability. IME it is much easier to sign on a handful of business sponsors than it is to get hundreds or thousands of users to commit to a few dollars a month. Your ROI will likely be much higher.
If you want to make it even easier for businesses to sponsor, can I suggest adding significantly higher tiers to your GitHub Sponsors offering? Most companies probably won't even pay attention to the other funding venues you list, but GitHub is likely already an approved vendor. Try to add a few tiers in the hundreds of dollars per month and the thousands of dollars per month.
You're doing great with this initiative. An earnest yet professional SOS is not a bad thing to do when a project is in peril.
For more inspiration, I hope this can help: https://matt.life/writing/the-asymmetry-of-open-source
A nonprofit that spends 1% of its funding on fundraising and barely scrapes by each year will be rated as more cost-effective than a nonprofit that spends 10% of its funding on fundraising and grows its donations by 15% a year, even though the latter is ultimately going to be able to devote far more resources to its programs.
But maybe they can't organize around your cause because the coordination and scale is too large. That's what large non-profits are good for. They burn efficiency on overhead, but can coordinate larger and grander projects.
It's a whole ecosystem. Experienced people are almost never comparing the cost-efficiency of a small organization and a large one for the same cause; they're using it to compare peers within whatever other criteria.
Raise a bunch of money, get an endowment going, and regularly fund medium size/medium usefulness projects both in a continuous way and on a grant-for-feature/limited-project way. Trying to focus on quality so most of the money goes towards people who actually get things done (as opposed to funding awareness, funding fundraising, and misc “not doing things” efforts) not that none of that overhead stuff would get done, but only a quite small portion.
I'm reminded of what happened to the creators of Dwarf Fortress and it's hard not to recommend some form of commercialization for projects that need financial support.
The problem usually is that this will cause a conflict of interest with the maintainer role. Leading to demise of the open-source/community side, and the rise of better alternatives while you're busy navigating this transaction and trying to balance the stick.
I'm just saying that I haven't seen much successful examples of projects that went down that route.
I'm not sure why we need to have holy wars over what is "open source" and rather recognize that it's a scale. Importantly, if an opensource operation can't be funded being unquestionably open source, then I see no issue making transitions like you suggested or the "free to user, cost to companies" method (how is that worth a holy war?). It's an extra shame given the average salary (current or future) of someone on this website. If you're not struggling, pony up, if you are, then I'm not sure how you don't have compassion.
After all, isn't the open source dream we all have about open communities, code, and everything? If we can't perfectly achieve that due to environmental constraints, let's not throw the baby out with the bathwater. That just discourages any other "open source" projects, by whatever definition you use. Mostly open is still better than fully closed, right?
[note to holy war people] I'll accept holy wars if you show me receipts that you fund >25% of the open source tools you use. That's a low enough bar, right?
What I haven't done yet is to earn a single dollars doing anything related to 3D printing and design work.
Mostly open is still better than fully closed, right?
The dream is that commerce and open source aren't in direct opposition to each other.
We do that by finding the right business models, not compromising our values.
I'll refer you back to my zealot note. The question isn't about if you're making money from the product but if you use the produce and have the means to donate. If you got a big tech job and you're only donating only $60/yr to open source projects then I'm not sure you have a horse to sit on. You probably pay a lot more for things you probably use a lot less. If not, I'm not sure what I said to that compelled you to defend yourself.
> The dream is that commerce and open source aren't in direct opposition to each other.
Yes, the dream is post-scarcity or being independently wealthy, which in either case one can just work for fun. But transitioning to post-scarcity is an incredibly disruptive process that has the capacity to hurt a lot of people before it benefits so many. But until then, people need money so I'm not sure what conversation we're trying to have here.
> We do that by finding the right business models, not compromising our values.
See the above where I mentioned someone previously try unquestionably open source and failed, as is the context around octoprint. The __only__ business model for that is relying on donations. You got an alternative that you'd like to enlighten us? One that doesn't "compromise values?" (I don't see the business pay option as compromising but it's not an uncommon discussion here. Hence the specific call out)
Let's be straight, it's not compromising values if you're forced into a position: that's coercion.
As I mentioned before, I paid thousand of dollars for Prusa products, which supports open hardware and support their slicer software. Though I have to note they seemed to waver a little on open hardware.
But fair enough, I paid 60 bucks a year currently for an open source software project. I should contribute more, and I have started supporting Octoprint before your reply. It's a pittance, but it's a start.
Yes, the dream is post-scarcity or being independently wealthy, which in either case one can just work for fun. But transitioning to post-scarcity is an incredibly disruptive process that has the capacity to hurt a lot of people before it benefits so many. But until then, people need money so I'm not sure what conversation we're trying to have here.
Not sure where you're going with that. Earning money with open source is the dream. I don't mean being supported by basic income or welfare, although I wouldn't be opposed to it, but in being able to bootstrap myself.
I don't have a sustainable open source career, and I am bad at business. I am working on that.
(I don't see the business pay option as compromising but it's not an uncommon discussion here. Hence the specific call out)
Businesses paying, corporate sponsorship, and support service, even software sale, are all perfectly fine and compatible with open source and free software philosophy...if it remains completely open source.
Let's be straight, it's not compromising values if you're forced into a position: that's coercion.
I am not sure what you mean by here. Open source developers doesn't have the right or expectation to expect being able to make a living, same as being a proprietary developer.
If you're not willing to make open source software for a living, then you're doing something else. Open source adjacent, open core, or just a plain old boring job at a proprietary software firm, whatever. You have the right to make a living under the current paradigm, but I'll keep searching for the business models that support my values.
Also I don't need Octoprint anymore (except for one older printer) since I all but one of my printers are now directly connected to Wifi.
Before I remembered that I was going to make a similar anecdotal point that I thought a lot of the market uses Klipper (not Marlin) these days and that there are other options more popular with Klipper firmware.
But if Octoprint usage isn't declining, then anecdata about usage is besides the point.