The open-source hardware business landscape is no doubt a tough one, but is it actually tougher than for closed-source hardware? That question has been on our minds since the announcement that the latest 3D printer design from former open-source hardware stalwarts Prusa Research seems like it’s not going to come with design files.
Ironically, the new Core One is exactly the printer that enthusiasts have been begging Prusa to make for the last five years or more. Since seeing hacker printers like the Voron and even crazy machines like The 100 whip out prints at incredible speed, the decade-old fundamental design of Prusa’s i3 series looks like a slow and dated, if reliable, workhorse. “Bed slinger” has become a bit of a pejorative for this printer architecture in some parts of the 3DP community. So it’s sweet to see Prusa come out with the printer that everyone wants them to make, only it comes with the bitter pill of their first truly closed-source design.
Is the act of not sharing the design files going to save them? Is it even going to matter? We would argue that it’s entirely irrelevant. We don’t have a Core One in our hands, but we can’t imagine that there is anything super secret going on inside that couldn’t be reverse engineered by any other 3DP company within a week or so. If anything, they’re playing catch up with other similar designs. So why not play to one of their greatest strengths – the engaged crowd of hackers who would most benefit from having the design files?
Of course, Prusa’s decision to not release the design files doesn’t mean that they’re turning their backs on the community. They are also going to offer an upgrade package to turn your current i3 MK4 printer into the new Core One, which is about as hacker-friendly a move as is possible. They still offer kit versions of the printers at a discount, and they continue to support their open-source slicer software.
But this one aspect, the move away from radical openness, still strikes us as bittersweet. We don’t have access to their books, of course, but we can’t imagine that not providing the design files gains them much, and it will certainly damage them a little in the eyes of their most devoted fans. We hope the Core One does well, but we also hope that people don’t draw the wrong lesson from this – that it does well because it went closed source. If we could run the experiment both ways, we’d put our money on it doing even better if they released the design files.
Imagine if the next Bambu flagship announced for beginning of next year is also easy to repair… The difference between both would become the features and the Core One already has less than the X1C (no flow calibration, no camera check of first layer, etc.).
For many hobbyists “closed source” = “closed wallet”. For my hobbies I try FOSS first. Let professionals pay big bucks.
any little bit of copy friction helps because clone manufacturers are lazy. I agree with Prusa on this.
Funny given Prusa’s “new” design is basically a copy of bambulabs x series.
not sure why that is “funny” when that is how business functions (I do not condone violating intellectual property protections unless they are reasonable improper), e.g., Android and iPhone are close enough “copies” and they each have enough patents for detente.
One of my previous products was copied by several Chinese manufacturers. We did not open source the software and that is what kept our product lead for 10 years.
By that argument the Bambu stuff is all ‘basically a copy of..’ as the community have been building similar concepts to all their stuff for seemingly eons, played with many of the QOL features etc.
When in reality the level of difference between these devices is actually pretty significant in how they are built and with all the effort put into user experience in various places – there is serious R&D costs to making them so smooth to use even though the concept is very comparable. As they are not just clones of each other or any of the communities various projects they rather resemble.
Seems like the Bamboo stuff is exceptionally noob friendly but will have a much higher total cost of ownership than the Prusa in how spare parts will work out. And the Prusa still looks to be very open to user customisations for their own personal needs and experiments.
NB I haven’t had hands on with either companies machines, so there might be more copied homework than it seems. But
Joseph Prusa pulled a Bre Pettis. lol.
I think it’s an unfair comparison in fairness… Makerbot had contempt for opensource hardware from day one IMO, I visited Hack In The Box Amsterdam round 2009. Chuck Pettis was there demonstrating the Makerbot Cupcake to a room of people, most of whom who had never seen one before. Basically claiming to have invented it solo, I stepped up and asked if he’d heard of the Reprap Open source 3D printer (Adrian Boyer was an investor of makerbot) and he straight up said said no, that makerbot was designed and built completely in house. I pointed at the silkscreen on the exposed PCB and asked why it said Reprap on it??? Chuck froze for a second, turned and walked away from his own stand, and literally hid until I was gone.
Pre Pettis waited until his keynote at OSHW conf to deliver his FU speech at the height of his success, while simultaneously firing all his engineers loyal to reprap and going fully closed patent troll for Stratasys – the company that kept 3D printing away from consumers for decades. This would be the equivalent of Jean-Baptiste Kempf doing a keynote at FOSDEM to announce that VLC Player will including mandatory Ads and data collection effective immediately.
Prusa has been OSHW to the core for nearly 2 decades, My first 3D printer was a Mendel Prusa that i modded and frankensteined to death over many, many years. I bought a Bambu A1 mini this year because i need a more of a tool and less of a hobby. Prusa no doubt has seen Bambu take a bite out of his market share with a closed design that takes the best of opensource communities (orcaslicer, klipper) and frankly gives little back. He’s got hundreds (thousands?) of employees to take care of and handing over the design files of non hobbyist machinable components is a world away from Pettis’s open contempt of his OSHW origins.
If Prusa put out a competitor to the A1 mini at a competitive price, people would be all over it on decades of good will alone. this all or nothing PURE OPEN SOURCE mind set is very Stallmannesqe and infighting does nothing to progress OSHW. When people are doing their best for as long as Prusa they deserve a little slack IMO.
i have written both open and closed source libraries but i dont make the open stuff closed.
If it’s open it stays open.
All of this being said if everyone rips you off and you dont get paid it’s frustrating and the engineers starve to death.
i have my own closed source code as well but i dont pretend it’s open.
Open Source takes us back to where the industrial revolution started from – a world where enthusiasts benefited from a common wellspring of new ideas, and were moved to return their own contributions to that well in a virtuous circle. But inventing quickly became a full-time job, and the only way to get paid for it was to jealously guard your discoveries, or someone richer would exploit them first.
The modern IP regime was a direct response to that problem, entitling creators to the proceeds of their work precisely so they didn’t have to keep it secret. This is what’s frustrating about Open Source fundamentalism, especially when it comes to hardware; it ends up opposing the exact principles it claims to be founded on. Because 90% of the zeal isn’t coming from creators, but from consumers who just don’t want to pay for stuff.
I think a mature version of Open Source would look more like a cooperative movement, where creators patent and copyright the shit out of their work, but assign those rights to a common trust. All creators have equal access to the knowledge, and when money is made from open-source designs, it pays every creator’s wages.
That way a company like Prusa can develop new designs, and publish all the schematics, because if a factory in Asia wants to churn them out, they’ll have to pay for a license.
Yup.
The fundamental economics of needing revenue to eat and live makes Open Source an ideology, maybe a marketing ploy, and less of a viable business model.
Even when the cost of manufacturing and iteration is near zero, as in software, people still need money to eat.
Hardware takes real time to iterate, real money for each iteration, and carries real risk of loss of capital. Open source hardware projects were quickly copied and sold for cheap by factories looking to fill downtime on their machines.
A large chunk of consumers just want cheap and low cost and don’t actually care about open source. E3D hot ends started open source and went closed source because consumers spoke with their money, they didn’t care open source principles they cared about price.
Open source should be considered a hobby and of you commit to open source project it should be for fun . Expect nothing in return and don’t put time in it that you can’t afford to do . Secure your own bag first before the giving away time and money to others.
I think Open Source hardware is doomed in that the creator loses a lot of money- which is required spending making hardware (as opposed to software). You’re buying PC boards, materials, CNC work, CADD software etc. all for the good of humanity? Never mind the many hours toiled.
There needs to be some way for a creator to at least break even so they can continue making things. When our favorite low cost country copies and mass produces it for profit, is a creator supposed to feel happy and proud about that? No way.
I also see someone will rip off the design by taking source files, making a small change and claiming it to be their own on their blog etc. for fame and glory. Is a creator supposed to feel happy and proud about that? No way.
I think Open Source needs fundamental changes but not sure how to do it. Protecting IP from Asia is costly as well. Good reads here: https://harris-sliwoski.com/chinalawblog