[drewdevault]

Sorry, Armin, but this is such nonsense.

Do you really think that Prusa going with non-free firmware is going to dissuade Bambu from undercutting them? The point isn't the firmware is free or non-free, hell, 3D printer firmware is rather easy to make. Prusa's was already GPL'd and there's no evidence that Bambu used it in their product, so making it non-free is pointless. The material issue is that Bambu is undercutting them on hardware costs. Now that Prusa is non-free, they have lost both the price, the hackability, and the moral advantage. It's an utterly braindead move. If Prusa were smart they would be looking for ways to reduce production costs to get down to similar price points, investing in usability improvements, and doubling down on the free software approach. Let's face it, Prusa got complacent and someone at their lunch.

You can monetize free software without breaking the social contract. I'm quite fine with your licensing approach with FSL, but it's not free software nor open source, and it does not align with the social contract, and that's all well and good so long as you can acknowledge that -- and live with the consequences, such as heavily disincentivizing third-party contributions. But you can make money with genuinely free software, social contract and everything, if you're smarter about it than barking up the VC tree and promising infinite growth and monopolization using a tool (open source) which is not suited for that purpose by design -- that was Sentry's mistake. Honestly, I'm glad that FSL is working out for you, but by no means does it justify some call to action to water down FOSS nor is it indicative of some fundamental problem with FOSS.

[punkybr3wster]

> reduce production costs to get down to similar price points

How on earth is anyone supposed to compete with near slave labor in China?

[drewdevault]

Certainly you can't get a similar product made at the rock-bottom price points that China will ship it for. But, you can get a lot cheaper than Prusa, and there are vulnerabilities in the Chinese market you can exploit.

And, to be honest, Chinese manufacturing is not the open slave market it used to be. They've cleaned up their act quite a lot, and the quality of life in Shenzhen isn't that bad. China isn't winning just because they have cheap labor, but also because they have made numerous innovations and smart investments in manufacturing. The poverty rate is only 6% higher in Shenzhen than in San Francisco.

[zaphar]

The reports are that their is still plenty of de-facto slave labor out there. The Uighur's for instance are heavily used as slave labor in re-education camps. I think it very much matters which industry and what commodities you are sourcing. Also we are pretty sure that the official reporting on this from China is heavily doctored. I'm not sure where you got your numbers from but I'm pretty sure I don't trust them.

[aragonite]

Both the average income and the cost of living is significantly lower in China than in a typical Western country, which inevitably translates into lower production costs, regardless of whether/how much "slave labor" was involved.

[sverhagen]

Just wondering aloud, here, could it be that the open source company comes from a place of feigned or real enlightenment that makes it hard to compete on hardware cost, as the lowest-possible hardware cost may require unethical practices in sourcing or production? Even something like "manufactured in the USA", if that were something you're into, I have to imagine, might make it hard to compete on hardware cost, compared to China.

[drewdevault]

Yes, quite likely. This is not really a matter of FOSS or not, but it is something that Prusa absolutely needed to have foreseen and made a viable business plan to address. Chinese manufacturing was always a huge vulnerability in their business model and they definitely should have planned for that.

That could have taken many forms. They might have invested in a much stronger brand, aiming to make Prusa (needs a better name tbh) synonymous with 3D printing, and developing a product line which covers a wider range of needs, from entry-level highly-usable stuff up to hackable mid-tier things and possibly high-end workshop printers as well, all of which should have had a shared brand identity. China can't really compete on reputation, so that's where to focus. Moreover, they could have invested more in services surrounding the product, such as the aforementioned cloud printing services, perhaps a thingiverse-equivalent with a models marketplace and a small margin off of that, etc. Auto-detect when your filament is running low and have a subscription service to ship you a new one -- maybe the cloud printer notices when you don't have enough in stock for a model you want and suggests you order more.

Just making a new model every 2 or 3 years and hoping China wouldn't notice was a very stupid strategy, and the free or non-free nature of their firmware has nothing to do with that.

This is such a common failure mode in open source companies -- they think that the code is 90% of the battle for the market. It's not, it's more like 5% at best and because it's open source the rest of the market can make use of it. This can be very powerful if you're smart about it, but "write the code and the rest will work out" is not that.

[chii]

> Chinese manufacturing was always a huge vulnerability in their business model and they definitely should have planned for that.

none of those plans would work - at best they're a gamble.

> China can't really compete on reputation

nor will they. The competition is always going to be on price. It's why counterfits exists as a market, but not only counterfits but also white-labeled copies.

The only thing that china cannot clone is technology they cannot get access to - such as high tech semi conductors, and that's thru a lot of losses copped by the shareholders by fiat from the US gov't.

[drewdevault]

Yes, it's a gamble. But having exactly one 3D printer model at a high price point and doing nothing else is also a gamble, and a much stupider one.

Western manufacturing is at odds with China for the reasons you stated, but free software eliminates this competitive advantage entirely and you need to plan for your output to be reused by your competitors as a fundamental part of your business model.

[the_mitsuhiko]

> nor will they. The competition is always going to be on price. It's why counterfits exists as a market, but not only counterfits but also white-labeled copies.

I’m not convinced that this is true. DJI, Roborock, Dreame, Xiaomi are all brands that compete to some degree on brand reputation.

[the_mitsuhiko]

> Do you really think that Prusa going with non-free firmware is going to dissuade Bambu from undercutting them?

I don’t have an opinion on that. I know that Prusa is clearly keeping some firmware closed (for now?) that they previously developed in the open. They also talk about their disadvantages of being open. Does it change the undercutting? Probably not, but I don’t believe that many more Open Source players will enter the race now.

> But you can make money with genuinely free software, social contract and everything, if you're smarter about it than barking up the VC tree and promising infinite growth and monopolization using a tool (open source) which is not suited for that purpose by design -- that was Sentry's mistake.

Our funding did not motivate the license change. I can say this with clean conscience. I understand that I have biases here but I also do not want Open Source to turn into something that is exclusively for plumbing.

I don’t want Sentry to be my last adventure and I also don’t want to give up on Open Source. But I don’t know a better way either. When I start something new I’m not sure I’m any more the wiser when it comes to balancing Open Source and commercial interests.

[drewdevault]

>Our funding did not motivate the license change. I can say this with clean conscience. I understand that I have biases here but I also do not want Open Source to turn into something that is exclusively for plumbing.

Here's the thing: you are no longer working in open source. If that's right for your product and your company, then so be it, I wish you success. FSL is an interesting development in this space. But, you opted out. You did give up on open source. Okay, fair enough, but to square your conscience I think you need to acknowledge that.

If you want to chat a bit more about the problems of open source and commercialization, share your insights from Sentry, and hear some thoughts from someone who might have done differently, feel free to send me an email, [e-mail]. Would be interested in talking about it with you further.

[the_mitsuhiko]

> Here's the thing: you are no longer working in open source.

That’s just not true. I personally and Sentry too write open source software every day. Not every thing we create is FSL licensed.

But I will take you up on the offer.

[drewdevault]

Sorry, I'm aware of that nuance but I worded myself poorly.