When open becomes opaque: The changing face of open-source hardware companies
blog.adafruit.com
blog.adafruit.com
It's a very, very sad story. It sounds like open-source hardware could have thrived if it weren't for China subsidizing local companies and enforcing bad IP claims for its domestic companies (which was really IP stolen from other countries but filed for patent first in China by Chinese companies).
i swear it's like the collective west just infinite-loops through the stages of grief when it comes to china even existing.
and to those of you who think "well we have to do something about it" -- yeah, the west has been "doing something" about it for hundreds of years. various tactics, strategies, wars, colonialism, both pro- and anti- whatever regime is in power. none of it works long term.
Chinese cultural identity, centered around Han dominance and collectivism, has been a unifying force for quite some time. HK was a major exception because of British influence, but I have a hard time seeing the rest of the country willingly breaking up rather than just slowly forcing through change.
Low-end semiconductor manufacturing is one of the last essential things the world imports from China, but much of this is already moving to Vietnam. Medium to high-end semiconductor manufacturing largely takes place in Japan, Taiwan, and the United States - all of which are allied together and will respect export restrictions. We will see Germany (another ally) step in once their fabrication facilities are complete and operational, maybe by 2026.
Calling their wide-scale demographic collapse as merely “challenges” is disingenuous. This is a looming complete and total catastrophe that will spark a humanitarian crisis, the likes of which the world has not seen before. Their short-sighted one child policy they had into place from 1979 to 2015 has robbed them of more than half of their 30-40 year olds, people at their prime working age, just as their parents are entering retirement. The legal obligation of children to take care of their parents in their old age had put additional burden on the single child of each parental union. These adult children are putting off having a family of their own due to this. This cycle continues.
And this is just the stuff we know about. We likely won’t even know how many people died of COVID-19, what are the figures of major events that we don’t even know about? Disinformation campaigns stamp out people speaking ill China, but when all this comes to a head, it will seem sudden but will the crescendo of compounding events, falling onto the next like dominos.
2023 PRC isn't capturing $8 per iphone 3g in assembly anymore. She captured $100 / 25% BOM since iPhoneX. Current trend of PRC+1 reshoring (derisking) is moving low end industries / labour to cheaper labour regions while PRC moves up intermediate goods and value chains. Hence why PRC exports to Mexico increase proportional to Mexico exports to the US. Mexico is finishing PRC intermediate goods to evade tariff. What we have seen is PRC becoming more and more indispensable and integrated throughout the global value chain.
Actual low end semi production is expanding massively in PRC. Medium end (12nm+) expansion is going to be mostly in PRC, because it's capturing market share in worlds biggest semi market that western fabs will increasingly lose due to export controls. What you're seeing is west strategic expansion of fabs that will lead to bloc overcapacity. Most of those fabs will only be viable via state subsidy. This isn't even going into capital goods / machinery that RoW are increasingly dependent on the PRC to try to re/industrialize at all. It's not what PRC makes but how much of everything that she makes that positions PRC to be indispensable trade partner for at least decades.
Repeating the demographic collapse narrative from Zeihan tier analysis is disingenuous. It's a manageable challenge / transition because PRC demographers aren't retarded, this is one area where they actually had a long term plan and are mostly meeting expectations. There's a reason PRC society has ~90% (87 urban / 96 rural) home ownership and some of the highest household savings rate. Old cohorts who saw QoL expand hundreds of times, simply do not have expectation of massive state welfare program, or increasingly being taken care of by kids. The entire dependency ratio / confucius obligation is orientalist projection that overlooks the fact that in aging countries like JP, old people just kind of rot and die unceremoniously. That’s where cultural expectations are currently trending, elderly don’t want to be filial drag on the next gen and are going to die quietly in their house with their nest egg. And what happens when they die? Up to 8 household wealth transfers to a couple who have increased consumption and family planning options. Will some families get fucked, of course, but you know which cohort will see large intergeneration wealth transfer dividend, the educated/skilled ones whose pushing PRC up the value chain.
Meanwhile PRC family planning prevented like 200m-300m excess births, when west is meming about PRC youth unemployment, which at 20% still means 80% of a very large number of skilled workers are being integrated into PRC economy. Much more than US integrates via immigration. It’s very difficult to make 6-8m new jobs for youth, especially high skilled every year. Nevermind double that. See PRC still has 600M farmers + low skilled workers stuck in the informal economy. They’re productivity drags. Reality is avoided adding another 200m-300m mouths that system wasn't scaled to deal with. Current TFR lower than planners want, but right now PRC is where roughly demographers were aiming and planning for.
PRC is also food self-sufficient in caloric terms. Agri imports mostly bulk animal feed and luxury items that don't undermine absolute food security. Not that it matters since she's shifting away from western agri to BR/RU/global south. And really this entire obsession with PRC food imports is dog whistle for PRC food insecurity, in which case, the US has exactly as much food and energy security as PRC because any disruption of PRC energy SLOC is going to be met with disruption of CONUS energy infra. Fortress America is over. PRC can starve US as much as she can starve PRC.
The entire idea that the PRC is unknowable is stupid. Incredible amounts of info on PRC is available via one of the largest western diaspora in the world. The problem is people choose to double down on the dumbest PRC collapse narratives from analysts with piss poor track records or outdated info that’s no longer relevant because US gov literally spending 100s of millions to seed disinfor about PRC while useful idiots wonder why PRC is not collapsing due to XYZ stupid reasons.
I took an upper-level course from a China expert in 1989. At the time, a big question was exactly how long China's government could maintain national unity unity in the face of the many contradictory forces pulling at it. Turns out the answer was: longer than a lot of Westerners thought.
They're not quite communist but they don't protect individual property rights to the same degree that we do. If we stole their designs they'd probably just shrug, but we'd still have to contend with our own laws and domestic competitors and international trade agreements with other Western nations, while they just continue on their merry way, growing and growing while securing logistics and supply routes from a bunch of small countries and mines.
At the end of the day, their system can more efficiently organize national production around some shared goal. Ours allows more independent experimentation and a faster pace of innovation. But once they have their minds set on something, collectively, I don't think our market would be able to keep pace unless we keep being able to technologically leapfrog them. But that can only last for so long, especially as our education continues to weaken...
That sounds like propaganda. Oligopolies are historically hilariously inefficient, and China's brain drain hardly seems like it portends well for them:
https://www.wsj.com/articles/chinas-brain-drain-threatens-it...
https://web.archive.org/web/20170225132236/http://www.nytime...
https://www.independent.co.uk/news/world/asia/china-s-millio...
That's why US/west copying PRC industrial policies, because PRC industrial policy is initially "inefficient", and then they become absurdly efficient, saturating domestic market at first, then export massive surplus to globe, and eats away at western shares. Analysis of PRC moving up value chain and global export share on PV, telecom, android smartphone, display panel shows reduction in operating margin of leading companies by 50-90%. Hence US is trying to stop PRC semi. And I'm guessing west on PRC EV soon. Because the efficiency of PRC indy policy is also measured in making western companies less productive.
As for PRC brain drain, Chinese academics/talent in west floating back to PRC in record numbers. Those are good portends vs statistically insignificant % of millionaires trying to hide their money abroad. Which entire ignores the fact that PRC generates so much domestic talent now that west can't meaningfuly braindrain even if they wanted to. PRC is in process of milking the greatest high skill demographic divident in recorded history. It took PRC 20 years to build up ~15M STEM talent to catch up in west in most sectors except extremely difficult integration projects like aviation and semi. She's generating that much talent every 3-4 years now.
E: over post limit
Yet the west copies PRC, an "unserious" regime, none the less. From PRC content moderation, to surveillance, to industrial policies, there's more convergence than divergence. Because PRC is prescient. Enough so to be labelled as only US competitor who is capable of reshaping global order.
China's massive digital divide is shameful, as is its neglect and brutalization of ethnic minorities. China's sexism is notorious the world over, as is its noxious pollution and the utter disregard its companies have for environmental laws. There's little for the West to copy, and even less the West should copy.
I'm not entirely convinced by this, of course. But it's important to note that headwinds are not brick walls. And even the inevitable next collapse of the Russian Empire will probably not happen for years, if ever. There are levels of survival that the elites who run sovereign states are willing to accept.
[0] On the timescales of normal family planning, the difference between home ownership and a ninety-nine year lease with no rent payment is immaterial.
China has its problems and its risks, to be sure, but it would be dangerous to underestimate their hybrid model of quasi-capitalism, quasi-communism. What is freedom limiting at the small scale (why would I innovate when the government owns me) can become absolute beasts at sufficient economies of scale. When they decide something is of national priority, they can top down refocus much of the national productivity (both money and people) towards it and then spin up a multinational logistics chain behind it across Asia and Africa. Then they scale up and dominate. It's happened to many industries already and just saying "China bad" isn't going to stop it.
It's like a game of Factorio played well, except with a billion workers and a lot of money that can be moved around at will. We see similar effects in the West during controlled wartime economies (when production can be directed to a particular cause), or government subsidies and money injections (like during covid, or with the infrastructure bill). It's also how our military is run, for example, and most of our big companies. Top down control economies have certain efficiencies (and inefficiencies) of scale that a bunch of small businesses don't have. But when it comes to commodity manufacturing, it's hard to beat their hybrid approach.
https://theworld.org/stories/2014-02-18/us-complains-other-n...
Of course, the obvious solution would be to not respect international patent priority from China, but that is tantamount to fighting a trade war, which has significant strategic implications. You can't sue the US to force them to de-normalize trade relations with China for the same reason why you can't sue the US to force the army to nuke China.
I stopped working on OSHW because it can't protect hardware. It has no legal teeth. Better to just give it away (public domain or whatever).
It does, it specifically protects chip die masks and by implication and case law protects PCB layouts and circuit diagrams.
China can do whatever the f..k they want. The problem is our own governments: they could and should have sanctioned the country to oblivion... had they not foolishly tied their entire nations to the success of China.
We all allowed, hell we welcomed Chinese products in our markets because our population got a decade or two of cheap Chinese made crap products which helped to hide wage stagnation (and rich CEOs getting ever richer). Then we got addicted, a ton of jobs got shipped off to China, here in Germany they stripped an entire mining facility and sent it overseas [1], and far-right parties fed themselves fat on the resulting economic devastation. And then we went even more foolish and sat idly by as our car companies completely ignored the domestic market and only focused on growing in China... with the predictable fuck-up that China learned how to make cheap EVs and now our industry is at the curb of collapsing against cheap Chinese cars and no competent competition bar Tesla.
[1] https://www.nd-aktuell.de/artikel/205443.vom-ruhrgebiet-nach...
20 years ago it "could" be justified by selling the idea that American or European labor is light years away in terms of quality compared to Chinese or any developing country's labor.
And that may still be true, but the gap has shrunk in size by a lot. To the point that corporations may not want to pay 120k or 200k a year for something that can be made in a developing country for 20k a year and with ~80% of the quality.
The thing is, the comparatively high labor cost in Western countries is so high because we can actually have a life on it and feed a bunch of uber rich people with our labor and taxes (because guess what, people like Warren Buffett and his ilk pay a ridiculously low tax rate [1] compared to legitimately employed working class people). In contrast, China, India, Thailand or Vietnam can offer very cheap labor because for the people there even utter pittances and absolutely ridiculous exploitation are better than the life these people had before.
While I do support the efforts China and India both have committed to lifting literally a billion of people out of utter poverty, it has at the same time brought disastrous consequences on our own society. I'd be happier with globalization if we had forced importers of any good to make sure that wages, labor conditions and environmental impact were on par with domestic regulations because the status quo is exploitation on all levels to benefit Western oligarchs (and imagine how the life of Chinese factory workers would be with Western wages!).
[1] https://www.cnbc.com/2021/06/08/bezos-musk-buffett-bloomberg...
The objectively fair solution would be for the developed world - particularly the countries that have benefitted from slavery, colonization, theft and other crimes the most - to pay for actual restoration and to meaningfully and sustainably help these countries.
Instead, the status quo is that Western countries exploit cheap labor and lax environmental laws to the detriment of the source countries in the usual rat race to the bottom - say China bans something toxic to the environment, there's always a poorer country willing to let that happen just to get an inflow of hard dollars/euros. On top of that we dump our second-hand clothes or surplus food especially in Africa under the disguise of "aids", which has completely wrecked local industries.
The EU is beginning to do at least something with the supply chain monitoring regulations, but the opposition to them is fierce - it's a pure show of how profitable exploitation (or its ignorance) still is.
If first world labor is living an extravagant lifestyle how much would you suggest dialing it down to? Should it be totally equivalent to the state of Chinese or Indian labor today?
The accusations of unsustainable lifestyle and over valued contributions can easily be hurled in the other direction. From a standpoint of pure self interest this should be concerning - by definition numbers are on labors side.
I am a very libertarian person and I acknowledge that the low prices of my daily living are in part due to the exploitation of others. This is unacceptable morally and I change my spending habits to compensate, despite it not being in my best financial interests.
The gap in price between Chinese labor and first world labor has also decreased.
If labor is too expansive, then nobody gets the goods. This is a feature of capitalism, not a bug.
Now that scaffold has no real foundation, save for a handful of gigantic corporations holding on as tightly as they can, and punishing anyone who dares contradict them.
China has been depending on industrial espionage in order to keep up for decades now.
Punish them for their malfeasance.
The way that the west works is far from ideal, but limited term intellectual property protection provides a significant portion of the motivation for the creation of novel products.
Without those protections, and the products that depend on them, there won't even be designs for China to steal.
Of course, there is a very valid argument that we don't actually need a lot of that shit, regardless of who's making it.
Oh, that's been done plenty. They don't care. Now what...escalate?
That's been done, too. Hell, that was probably Trump's biggest campaign promise. Obviously it didn't work.
So how far should we keep escalating? Do we declare war over intellectual property? Fuck that. It's time to wake up.
We are the ones being called out on our bullshit. China called our bluff, and we are left holding the cards. There is no sense in waiting around: we should fold ASAP.
I refuse that and actually would go to war over it.
Because without that, you just reward war-like tactics in business, thievery, deception, etc, until nobody wants to create ideas anymore, and thinkers are selected out of the gene pool until we're a hive of drones sucking the toes of our queen bee influencers.
Where can I see the names of the people who designed my phone? Is it printed on the box? Or those who wrote a service I use like GitHub?
Inventors have been dispossessed of their authorship for the last hundred years.
> thinkers are selected out of the gene pool
Right now having contributed to openbsd makes a guy the sexiest dude in every party, right?
Honorable, idealistic, yeah. I deeply value the people that work on OpenBSD. I wish they had fewer struggles in life. They should start by valuing their time and effort more, like lawyers, who charge 10s of thousands of dollars to read PDFs.
It is absolutely morally debased and corrupt to advocate violence to resolve intellectual pursuits. I want to exist in a society that shares ideas since they are not scarce and we can argue over the actually scarce things we care about.
It's just personal preference, like all politics. We create the world we want to live in. Yes, I once idealized your society, but after years in the corporate world I learned there will always be plenty of snakes out there. I'd prefer not structuring society in a way that rewards them even more.
this is known as impotent range.
Either way, your comment really added nothing at all to the discussion - no counter points, just a poorly considered insult.
So here's the thing - China is a self deluding autocracy that has repeatedly purged competency from its institutions and government (the extent to which this is true far exceeds that of any western nation), I don't need to exert any control in order for that titanic economic system to fail - I just don't.
The entire system is built on lies told to superiors - exaggerated, repackaged and passed on to the next level, all the way up to Xi.
Everything they build depends on stolen technology, but they lack the tacit domain knowledge required to correctly implement what they steal; Cha bu duo simply is not good enough.
Not only can they never compete on merits, they can't even reach a baseline of competency.
That system will fail all by itself, and the more nationalistic they become (and that's a feedback loop) the faster that western support will erode, and the faster the collapse.
The ground falling away like a Ponzi.
What need have I for... range?
just like the US federal reserve, the US federal debt, the US stock market... it's all going to collapse any day now.
give me a break. both the US and china doomsayers are perennially WRONG. they're all fueled by the same stupid handful of youtube channels that spout the same shit over and over and over again.
Come on, China was pretty irrelevant to the outside world until the US started pumping money into it in the 70s. That's exactly what the West should stop doing.
literally everyone has tried to take over china (modern translation: "access their markets"), from the mongols to the british to the japanese. i bet even the romans had some half-assed plan they were working on.
China was an Empire because they tried(and succeeded) taking over neighbours by force, including Mongols and Japanese.
To portray Chinese like saints or victims is not knowing about the thousands of years you talk about.
Uhm... You clearly have no idea about China.
First, it was _itself_ conquered by Mongols ( https://en.wikipedia.org/wiki/Mongol_conquest_of_China ). Who were then simply assimilated.
Second, China has never conquered Japan.
Overall, China has been remarkably non-aggressive.
They're no saints. Anybody can see that. But by my estimation 90% of the discourse around China in the west is just pure consent manufacturing.
OKAY BUDDY.
you are literally just making shit up.
you're going to upset both the china ccp tankies and the japanophile weebs at the same time. truly a remarkable feat. something i didn't think was possible.
I wish Western countries had import bans on goods manufactured by Chinese companies that did that.
The bigger problem seems to be that the i3 printers are table stakes now, while Prusa engine seized up and forward progress went to pot. The Prusa XL is not vaporware but it's close and preposterously expensive, while the MK4 released without input shaping in their extremely customized port of Marlin. Those "Chinese knockoffs" are shipping Klipper-based printers, with SBCs, where you get that for free.
And then you have the Bambu printers, which are OSS-compliant where required (their slicer is a derivative of PrusaSlicer) but built an in-house OS (not actually that shocking, and almost certainly not pirated; 3D printers just aren't that hard) that has given them a printer that's conservatively 18-24 months ahead of everybody except maybe Prusa with the XL. I really don't like Bambu's patent saber-rattling, that sucks, but they just shipped a better mousetrap, too.
Open-source 3D printing is on the ropes because it isn't making stuff as good as the closed-source stuff. Not because of cheap cloners. I didn't want to buy a Bambu printer so held out for the Creality K1 (and they are being bad citizens, they haven't released their Klipper source) and it's not good, I waited for Prusa to sell me on a Prusa XL but to get one I'll be waiting until 2024 at the earliest and spending $500 on that "preassembled" option to boot, and...Bambu just sold me a P1S with an AMS for $1000 flat. C'mon.
This is the crux of the problem. Where Prusa is openly sharing, you have companies that are benefiting from that without reciprocating. Part of the tax you're paying when buying an i3MK4 is the continued investment in the open source/hardware contributions of the company, not just the end product. Shelling out $1k for a Bambu is your prerogative, but it does cast a vote with your wallet for a company that is more predatory than collaborative.
So what, exactly, is your point here? Bambu is compliant and is reciprocating; OrcaSlicer derives from Bambu Studio and works great. Their printer firmware is not as far as anyone can prove derived from GPL software (and I tend to think that by now somebody would've found it), so they keep that.
Let's get to the brassest of tacks: one can talk about "contributing" until one's blue (or orange) in the face, but Prusa can't or won't ship a functioning, assembled-before-you-get-it multimaterial CoreXY for less than $2,500. Bambu came out the gate with one, not reliant on Marlin or Klipper, and it actually works.
I genuinely can't believe I'm having to point this out, because I think Bambu does suck as a company and sniffing about patents well-and-truly sucks, but they remain, and are the only, such bastard-coated bastards that can actually ship something usable at a price somebody can afford. $2500 for a Prusa XL would be more expensive than my CNC. I actually do need multimaterial, not for models but for tooling, so continuing to limp along with my collection of Klipper printers isn't reasonable. So what's your actual solution for a quality-outcomes tool at a reasonable price?
That being said, everyone I've interacted with at/from DJI have been fellow tech-nerds of the highest order, and I mean that as a compliment. I've enjoyed talking to people who are excited about building things.
Why is this China factor even a problem for open-source? The open source community have always been threaten by closed-source copycat. Even western companies also do that without much repercussion. The real threat is, as the article pointed out, the trend of `open source` companies going closed source because of profit motive.
I am more annoyed that Prusa is doing a bad job of shepherding open source and shipping bad products than that Bambu has a closed-source better one.
Bambu lab on the other hand came up with something pretty good, very well integrated that has basically taken 0 from the prusa designs. So it's not really closed source profiting off of public or open source work. Maybe for the slicer, but that's it.
Just compare the abysmal performance and quality of prusa's MMU that still really sucks almost half a decade after they originally released the product. Even if they are super expensive too! While on the bambu printers... It just works.
In other words, is this a fear of something that might happen or something that is currently ongoing?
Or, to put it another way, is the simply rationalization for decision to close source made for other reasons or an actual reason?
If a Chinese company simply makes and sells a version of an open source design are they doing anything wrong? This is allowed by the licensing terms.
I have only marginal sympathy for Prusa when they aren't shipping things people want to buy.
The problem is rather the customers who buy China stuff instead of products from companies from western countries.
"The upstart nation was a den of intellectual piracy. One of its top officials urged his countrymen to steal and copy foreign machinery. Across the ocean, a leading industrial power tried in vain to guard its trade secrets from the brash young rival.
In the late 18th and early 19th centuries, the rogue nation was the United States."
I do like Limor's response "I’m going to keep shipping open source hardware while you all argue about it,” She's fighting the good fight as always.
I've been designing hardware for decades. I've come to learn that it's more about staying ahead of competitors technically than keeping them from copying you. There will always be copies, you just need to be selling the next better version while the copies are of your previous version. There is no "make a thing, profit for 20 years". If companies like prusa or sparkfun stay knowledge leaders, people will be willing to pay a few extra dollars foe their product over a clone just to have the improved support, documentation and quality, also to support what they want to support. Making this change makes these companies no different than the clones now. This move takes away incentive for me to order products from these companies and I believe will actually cause them to loose more business than they are expecting. Their whole sales model is built around this. It's why I order stuff from them, or used to.
The drug companies with patented medication would like to disagree. While 20 years seems a bit too long, intellectual property protection should and does exist so you can get a couple of years out of a product, and that seems okay. That some choose to go the Open Source route is their perogative. For those that don't, and make a closed source proprietary product, they're still going to get cloned if the product is popular, even with copyright and patents. (Trademark is a different story.) Look at FTDI and their USB-serial chips. Copying an IC isn't easy, can't just git clone that shit, and they still got ripped off. The story of an inventor who made a device, say, the clapper, and lived off that for the rest of his life may seem quaint, but why should it?
Making and patenting a new, novel, molecule (which is what the entire pharma, materials and chemicals industries basically do at the most basic level) is very different than making a hardware product, which is yet different than making a semiconductor product, which is yet different than a software product, yet different than a genetic product.
The context of this conversation is specifically a hardware product.
Like so many things, it's not about what "should be" but what pragmatically falls out of the chaos of the world. These things aren't master planned by some single person in control. In reality for a hardware product, and any product in general, they say your patent portfolio is only as good as your ability to protect it. That is why these big classic chemical and pharma companies like Dow, 3M, J&J, etc. all have insanely robust patent infringement protection infrastructures in addition to their patent portfolio. In many cases those companies didn't even develop the molecule originally, they bought it. They're basically manufacturing companies with patent enforcement teams.
People like me (I have many hardware patents) don't want to run manufacturing and patent lawyer companies. We want to develop new products and then sell them to the former.
You say why shouldn't someone invent the clapper and then run a company that manufactures only the clapper for their whole career. I say what great inventions did society miss out on because a creative inventor was playing lawyer and manufacturing manager instead.
Ultimately the answer is because different people are good at different things, at different parts of the process. Not recognizing this is to fall victim to "founders syndrome"
The design cycle for hardware products is typically 6 months to 2 years maximum. Moat consumer products get refreshed pr redesigned every year, typically to coincide with CES and other things like the holiday rush. Same with the toy industry, there is a big annual show that you need to release your toy at or you wait to next year. In medical devices it's a bit longer, 2-3 years is common.
If you take much longer than that to bring the product to market you have to start redesigning the stuff you designed first. Components go obsolete, computers and displays change every year, software needs to be patched. It's why programs like the f35 and f22 are such messes, they were designed over a decade.
Most hardware products we interact with daily took less than 3 years to design and bring to market, they are not F22s. Some things take longer like an MRI machine but they typically are done in a iterative way so that proved versions of a product are brought out every year or two.
If society really wanted to wring every possible invention out of inventors, we'd lock them in the basement and give them a workshop but instead we let them loose on YouTube so they can try to become Internet famous.
What great works of art are we missing out on because someone needed to work a job for money instead? What great works are we missing out on because someone was successful, made a bunch of money, and didn't have to work again?
You're totally right about specialization and people getting into traps as a founder.
No way you can compete with Chinese giving out your source code that took years to create so they can copy your product in weeks.
I think open-source is a laudable goal, but your competitors have to be willing to play by the same rules, otherwise you're hobbling yourself. I worked at an agriculture startup some years ago, and while we all wanted the gizmos to be hackable for our customers, we all knew that if we opened things too much, a real heavy like John Deere, Monsanto, or Simplot would swoop in, leverage their existing logistics and customer base, and put us out of business the instant we had a product valuable enough to steal.
I don't like that e.g. Sparkfun is putting out a product that's worth more on its own than as a learning tool, so I agree with you. This signals a shift in Sparkfun overall that I don't like.
Agree wholeheartedly. The clones are here to stay. I push people, beginners especially, in the direction of Adafruit because their documentation and build quality are excellent. I also use a lot of Adafruit hardware in my own freelancing work. Their products are well worth the price premium.
With the exception of M5Stack, I haven't found a product line that I think is as well thought out.
That said, clones have their own place in the ecosystem. Often the differences between a cheap clone and the more expensive original are nonexistent across all axes: quality, support, documentation, etc.
Most people are not going to pay more for an identical product.
In the case of the original Arduino, I have a Chinese clone purchased for $3 that has substantially better quality, and has many of the features you would want from a dev kit, like ESD protection on the I/O pins.
Maybe, but pre-arduino dev kits were often hundreds of dollars because doing low-volume PCB manufacturing was expensive. Now that it's cheap, Arduino is kind of obsolete.
Limor's response was: not really!
You see, the Chinese are literally the world experts at reverse-engineering electronics. It is nothing for them to take literally any piece of electronic kit on adafruit.com, crack it open, list out a BOM, scan and trace out the circuit board, and have a prototype ready before lunch time. If they decide to clone your widget, making it closed source isn't even going to slow them down.
YES, they will make money off your design. And you have to be okay with that. Because what they can't (or at least don't) do is build a thriving and supportive community (and ideally, repeat customers) around themselves.
Sparkfun never fit as well in the maker space, and their move towards closed source isn’t surprising in light of their slow decline over the past few years. Their marketing is much more engineer focused, including their product selection.
Adafruit and Pimoroni have the bravery to make new product categories knowing the clones are coming but that they will capture enough of the market to succeed in them anyways. Neopixels are the prime example of this to me.
I love shopping at Adafruit. They make it easy to want to buy from them. It will just work, I know it will just work, and I know they have a library and an example project waiting for me.
I can get the “same” thing on AliExpress often for half or less (depending on the part). But not with the same fit and finish, not with a no solder hookup, and not this week. And if I do buy it on AliExpress I will use an Adafruit library with it. Open source means everyone using parts they introduced to the maker market is their customer, some just didn’t buy the part from them.
Open source helps their customers reverse engineer their products more than their competitors. Paying someone with actual electrical engineering background to read data sheets closer/better than me is well worth it. I’ll buy their parts, prototype with a few of them, and then make the combined form factor I want with JLCPCB. Adafruit even published many of the tutorials teaching how me to do this! They win by trusting their customer enough to let them go free, knowing they will come back.
Eagle, Spark Fun, Arduino, Prusa, Red Hat, SourceGraph, VSCode Plugins (was it OmniSharp), etc, etc...
MIT & BSD licenses are used as a weapon against GPL more and more...
Rust's "Rewrite In Rust" movement is used to replace GPL tools with MIT versions which can be closed on a whim...
"{VSCode,Chrom}ium" projects give the illusion open source while being effectively used to harvest community effort, too.
I don't think we're on a good track.
Disturbing times.
If and only if publisher shares the source.
Permissive licenses are not “viral”. Sharing the source is not mandatory.
I can take your work, evolve/improve, publish a tool, and tell that the tool contains some code of you, if I don’t forget.
You can fork this version as much as you like. If you can find the source, of course.
What we have now is not a fair deal, to the point that people are trying to re-invent the notions that the laws were originally supposed to embody.
what there was: trademark law, copyright law, and patent law. but 3 turn to 1?
when I stared coming online, national talk about IP laws was just starting out. I don't keep up to date with law reforms, but in the late 90s mainstream awareness was shifting into this mindset of IP law. and back then, the law still spoke of those 3 concepts on their own. I suppose now it's really just one blob of ownership over ideas for the sake of some misguided market-ideology
There are plenty of countries that didnt enforce it for a time though and experienced a kind of mini renaissance.
I dont think it was ever really intended to "protect" the rights of inventors nor was it ever good at doing so.
See: America, England, and the Industrial Revolution
Patents don't take into account things like you can "invent" something naturally by exploring given idea. I have few times developed something and then through more research learned it has been patented, so I had to find a different way of doing the same thing and wasting time. Even if I invented something first I wouldn't have means to patent it.
Now we have a situation where things like VC funds are forcing companies to patent anything they can as a prerequisite for receiving money - in case the idea won't get executed correctly, they could chase any other company that comes up with the same idea, for money.
and yeah, you invented something, but for any reason you didn't or couldn't patent it and then some toff's engineer figures the same idea? They get the patent and you have to abandon it.
The whole patent thing should be scrapped is not fit for purpose.
Yes filing patents can be expensive but for these companies it shouldn't be an issue and if they truly are a tiny org without the capabilities to file as a standard org, they can file under the small entity or micro entity fee schedules which are far cheaper.
Open-source was just marketing for Arduino, and it worked for them when they were selling an undifferentiated dev kit for 10x the price it should actually have had (and at least a 30x markup on their actual manufacturing cost). They then load down those dev kits with software that is so inefficient that it upsells people on huge chips for problems that could otherwise be solved with a 10-cent chip. On top of that, the initial Arduino software was pretty much stolen from a grad student, who got no credit, and using open source also gave them free contributions from motivated users.
Fast forward to now and they have a "community" and are trying to start selling more complicated dev kits with the same ridiculous markup, and have found themselves unable to compete with Chinese companies that charge a fair price. The end result, killing the openness, is inevitable.
OTOH, I would easily agree that the market moved under them, because today others in the space can provide the same easy on-ramp at a lower cost with better hardware, which is leaving them flailing a bit.
Additionally, Linux has available a multitude of very rich in-band debugging and logging capabilities that help you understand and mitigate common problems— whereas a bad interrupt or peripheral configuration on some microcontroller could have you banging your head again a GDB frontend or shelling out thousands for a tool like Tracealyzer to tell you what on earth is going on with it.
Granted, I write microcontroller firmware for a living these days, but I still use RPis and Arduinos for proof-of-concept work because of the simplicity of the toolchains.
Not to say that your original comment didn't have merit; the beauty of hobby projects is that you can completely skip the last 20%.
A full OS brings a whole lot of software complexity and places greater demands on the processor, which means greater power draw, greater hardware expense (because you need a beefier system just to run the OS) and reduced reliability.
There is certainly a role for systems like a R-Pi that have a complete OS on them, but there is also a real place for lighter systems.
Unlikely. The marginal cost is so low that a "beefier" but standard processor that's widely used is likely to be cheaper than a lower-powered but less-popular processor. And using a standard system that everyone else does is more likely to have the bugs ironed out.
I don't like it, but realistically unless you really need low power it's cheaper and simpler to get an extremely standard off-the-shelf RPi like system and slap whatever on it. The savings you can shave off by going smaller just aren't worth the effort.
And having a full OS running means that more code is running. More code means more potential points of failure.
Of course, if your project is something that needs a larger processor or more advanced software, then what you say can be very true.
Also, if your skillset is more in line with programming larger systems than smaller ones, the development time will be shorter for you by going with a full OS.
I wouldn't even know where to start with a pile of undifferentiated chips. Arduino lets you spend a bit of money (relatively cheap when it comes to hobbies) to learn the ropes from vetted and curated parts that are made to work together, including the software.
At some point, yes, maybe you know enough to be able to evaluate the Chinese knockoffs on your own and avoid pitfalls and counterfeits, and find the correct vendors who offer an awesome product at a good price. But it takes a while to get there. I can hardly find reliable power stations and USB PD chargers these days. I wouldn't even know how to start to evaluate an entire dev kit.
If anything it seems like this is the fate of intellectual property in the age of global capitalism. Whatever we design, whether it's software or chips or fighter jets or solar panels or cars will be copied and produced much more cheaply there because their costs of everything is much lower. And actually, relative to most of the world population, it's probably the USA that is overpriced. We have our insane quality of life to keep up with.
But that's hardly the fault of any one company. As we move more and more into services, domestic manufacturing just can't keep pace. All those reshoring efforts don't really seem to be making an impact. Most things I see are still Chinese, especially at the price points I can afford.
I don't know that "killing the openness" is the inevitable result. Closed designs get stolen and copied too. Getting bought out and eaten alive by Chinese companies the same way Hollywood and video gaming have been going seems the more likely route?
The Arduino was a copy itself, and no credit was ever given to the original grad student who came up with it and did almost all of the work to make it into the system that was released.
Not enough credit perhaps, but this is untrue as Arduino does credit Wiring, the grad student's project. Probably worth noting that one of his thesis supervisors was one of the founders of Arduino.
That means it's a money grab? How is it their fault no one competed in this space?
Pivoting from a nonprofit to a for-profit should be illegal. Looking at you, OpenAI.
* Plugged into USB (The Decimila, Duelmilnova, and the Uno)
* Had a software stack that worked on Mac, Windows, and Linux
* Had a software library that allowed people to drop in features to accomplish goals.
Arduino was a bad fit for an electrical engineer making a product, but it was an amazing fit for hobbyists and artists trying make a one-off project or presentation. This is the thing that the EE evangelists and gatekeepers could never seem to grok: No one* cares if it's taking 4 cycles or 120 cycles to blink the LED, or turn the servo. Sure it starts to matter if you're doing a dozen other things simultaneously, but most people just wanted the LED to blink.
Arduino was a miracle when it showed up. No toolchain to figure out because it was all just in the package. No bitmask decoding. No being rejected because you're not running Windows. The C Superset was surprisingly readable code for people who didn't have a programming background.
It is shameful the way that Arduino was pulled out of Wiring, without real credit and acknowledgement.
*Yes, there's the subset of people who were chasing efficiency, or who were interested in doing things "The right way™", but this was never the target audience for Arduino.
The team was a group of academics who weren’t necessarily ready to build a business, but instead of letting it go when the market, they’ve made some desperate decisions. The industrial/commercial markets that a lot of people here use microcontroller boards for are well catered by much cheaper boards, and Arduino should’ve stuck to developing their original creative market with better tools.
and do what? That market isn't used to pay for the tools (hell, they made "good enough" one that's free), and their boards are too pricy even for some one-offs
You know, the thing they were saying that they were?
But there are of course plenty of examples of open source projects which sustain themselves with sponsors, etc.
That's debatable. I don't think China has the same labour laws or costs of running a business. Maybe it's fair from their perspective, but we really should have tariffs on such products so that they would cost as much as if manufactured in the west and perhaps use that money to help domestic businesses grow.
It's not the case of just chinese labour being cheaper, they earn massive profits on one and when you can have 5 or even 10 boards made for cost of one arduino it just becomes silly.
Also take into account that someone with a skill is not going to look at making a living wage. It's a poor return of investment in one's education.
They sell for $30.
China IP address show up within weeks of starting any new small open project, and 2 months later one often sees project cloned alpha PCBs available on Ali-express/ebay/Amazon/tindie/sparkfun. The defective legacy RAMPs 1.4 with potential fire risks are still being sold a decade later.
It has become such an issue, that even finding the original authors to support their projects becomes increasingly difficult as google starts to overflow with pages of SEO ad links.
I wouldn't say anything has changed, but open hardware doesn't seem sustainable unless you are an active small factory in China.
Good luck =)
Ultimately if your revenue depends upon you selling the hardware you open source it will be very hard going as other companies can easily churn out high quality (or indeed less high quality but super cheap) versions without paying any of the significant development cost and hence undercut you by a wide margin.
Indeed referring to these other versions as 'clones' seems to miss the point of open source hardware, isn't the entire point making the design open so others can build and iterate upon it for whatever uses they wish (including commercial exploitation).
It makes them fragile, and likely also a frustrating experience to deal with, since you wouldn't get support.
The advice to investors applies here as well, “the market can remain irrational longer than you can remain solvent”
Our company committed to open sourcing all of our code (it's in the web3/blockchain space), and we had, and continue to have, spirited discussions about which parts we should maybe license differently, as they contain novel IP.
But my main question is: if your code is open-sourced, and the community contributed: fixes, features, actual new products - what gives you the right to close it? Are you going to go back and compensate every contributor? How can you justify revenue made on the backs of contributors.
Side note: if what Prusa is alleging about Chinese patents given for open-source code produced in the west, and then having international priority, is true, I think the UN (or whoever handles international patents) should look into that. We can't control what goes on in China, but we can damn well make sure no Chines company makes money outside of China, with co-opted IP.
Typically you're not able to close source existing code, once it's open it's open. What you can do is make the changes going forward proprietary.
Depending on if you got a contributor-license-agreement you may not be able to close source the community contributions, but if the code was licensed under something non-viral like MIT or BSD you have as much right to close source it as literally anyone else does.
I guess I really don't understand the question. You have the rights as outlined in the license, people who contribute agree to those license terms.
I think that software developers are probably the kind of people who can know what deal their actually getting if anyone can.
So the question is really whether you are fine contributing to a copyleft/permissive project.
On my end, as long as I keep my copyright (i.e. I don't have to sign a CLA), then that's fine for me. If anything, any contribution I make makes it harder for them to re-license their project :-).
Seems fair to me :-).
And the "closed-source" part will be only the new code added after your contribution, but the project itself will keep the open source license (so that you can fork it at the state it was when the authors decided to go closed-source).
...especially the option of making everything proprietary.
The version of Foo that you released as open source will always be open source. Now I can make a copy of Foo, use it in my proprietary software Bar, and ship Bar to customers without sharing the sources of Foo (but I need to share the license; all permissive licenses that I know require attribution). Foo is still open source on project-foo.com, I just don't distribute it with Bar.
I can modify Foo, and ship it to customers as a proprietary library (with attribution). My changes to Foo will be proprietary, but project-foo.com will still exist and will still be open source.
Now you own the copyright of Foo, so you can decide to start shipping it (and all new versions) as proprietary. But you can't ask me to delete the fork I made from your open source version, foo-fork.com. So that one will still be open source.
0: https://www.youtube.com/watch?v=Vj04MKykmnQ
Edit: NSFW (but still compliant with YouTube obscenity standards)
Man, this discrimination is just disgusting.
This isn't the first time she's had issues either. Her openness about her personal situation, her relationship with Kaidi, etc, certainly skirts what's allowed to be spoken about in public.
Hopefully she's able to continue her core work without unbearable compromise.
This is completely false. You can bake your cake with your secret recipe and eat it too.
If you give someone else your improved cake though, you have to give them the matching recipe.
I do think there's a place for permissive licenses, particularly for academic and government projects. However, it seems like private entities can't be trusted to play nice, so copyleft licenses should probably be used by more open source projects to protect the public knowledge base.
But the unspoken good faith statement is "You're going to take this, and make it better, and we're all going to benefit from your efforts to make it better".
The printer clones were not made with an eye towards making things better, but made with an eye toward making things cheaper, and more specifically, more profitable for the manufacturer. It could be argued that cheaper is a form of better, but I think generally the consensus is that the cheaper clones did the job less well, and the sellers already had our money.
I guess you mean that the intent could be different in different cases or differently understood in the same case. Here, with 3D printers, some intended to encourage to make better printers, and some intended to allow the same quality for less, or at least to allow presenting a variety of price-quality offerings to choose from.
Sellers already had our money in case we paid those sellers the same money for inferior product. The product of similar quality - even if seller managed to make it cheaper - seems fair game, and with time we'd assume the price to come down.
So, we probably have a disagreement on what the intent was or is for the OSHW. For some intents the examples you give are expected.
I mean, I understand the enshittification point. Perhaps this is yet another example of that. Chalk up yet another victim to the financialization of literally everything.
Bambu seems more like a consumer product, even if it's a pretty impressive one. For people who just want no-hassle printed parts, I think the Bambu looks very compelling, but it just doesn't have that reprap spirit.
I've been very happy with my Prusa mk3s for the past few years, and excited to get the new one going. I'm worried about the company, though. Seems like Prusa might be on the path of becoming one of my favorite company and then running into trouble:
Pebble - no explanation required. Sparkfun - New CEO took over, company stopped doing ALL the things I loved about it. Printrbot - not sure what happened there? Prusa - hopefully different!
The one thing about this whole conversation that is frustrating to me is how people blame these companies for not being cheaper than China, essentially. “Why aren’t you better than Bambu?? Why is business taking so long??” Like, this business is freaking hard. Hardware tech is nearly impossible to succeed in as a small company with low cost, consumer machines! Now competing with China, who got to free ride on accomplishments from OSHW companies and community, is just something they’re not doing for fun or whatever??
These folks are in a nearly impossible situation. China has nearly every advantage. It’s a miracle that any of them have stayed in business at all!
So while I do think it sucks they’ve pulled back on open source, it’s completely understandable. These companies are barely surviving as it is. It drives me nuts how people take this all for granted.
That said, Limor Fried is such a bawse. All hail Lady Ada!
When it comes down to it, "Open Source" doesn't carry much of a price premium for a lot of people. "You can print the parts yourself" doesn't carry a price premium for a lot of people.
I think Prusa enjoyed the "It Just Works" and being the logical upper-level consumer printer recommendation, and a little bit conflated that with the ethos of the company. Which worked for a long time.
But I think he'd be better off pushing fair labor practices and superior support as the main things, rather than grousing about open source.
And there's a very good chance my lab will be replacing its current printer with a Mk4.
But I think the key is I don't feel that same excitement about building a kit. The open source nature of the thing really doesn't matter to me beyond "That's nice". And that's why I bought my Mk3. Because it was a no-muss, no-fuss printer that I ordered assembled from the factory, put on my desk, and got to work with.
The problem is that space is now a little more crowded, and it's hard to compete on a feature-by-feature comparison, especially at the price point. Some of that is stuff Chinese firms are getting away with, and some of it is genuinely that there's such a thing as economies of scale. But I also think Prusa has a lot of goodwill - I can't imagine another company releasing the Mk4 with a major advertised feature (input shaping) missing, and the plan for it being a one-size-fits-all approach and not getting eviscerated for it, rather than most people going "I'm pretty sure they'll work it out."
I bought a Prusa kit, had it sit around for a bit, finally put it together with one of my kids and since then we keep finding really good uses for it that I would never have thought of before I had the thing. The idea that you can fabricate small scale plastic components in a tiny corner of your desk has been a game changer in many ways. Just the other day a part on my car broke, which the manufacturer wants an absolutely outrageous amount of money for (it's a part of the door mechanism). An hour later or so I had a near perfect replacement in my hands (10 minutes to design it, 50 minutes to print).
But yeah, my interest in 3d printers is "I need an X" - either a bespoke, custom plastic part that is made in small batches for a research project, or for my home printer, wargaming terrain, and I really don't care about modifying the printer, etc. That's also what's been standing in the way of me building a VORON - I just...don't want to.
The man pressing the button has his own reservoir of clean water.
If I was ever interested in these companies, it was because I prefer OSH. Should they stop doing OSH, I'll simply look elsewhere.
- In the last years, the Arduino team has been releasing MORE open source stuff than it had been doing in the past. This is documented transparently in the yearly open source reports: https://blog.arduino.cc/2022/01/13/the-2021-arduino-open-sou... https://blog.arduino.cc/2023/01/20/the-2022-arduino-open-sou... Even recently, the flagship Arduino product (UNO) was renewed and with no surprise it's still completely open source.
- In parallel, a vertical line of industrial products (managed by a subsidiary called "Arduino Pro") was added that for documented reasons has all the requirements of open source hardware (including full schematics and fully open source software stack) except for the CAD files which are available on request. This level of openness is the same as a Raspberry Pi, if not higher because all components are available on the market. The existence of this additional product line does not imply at all that Arduino reduced its commitment to open source because, as said above, such commitment is even higher than in the past.
Based on these facts, in no way it is reasonable to state that "Arduino is going closed source" or that its business is "opaque". All opinions are legit, but if they don't take facts into account it's just misinformation or marketing...
- How does an Open source hardware company make profit?
- If they do want to make profit by going closed source, what is the issue in that ?
- Does it matter if they go closed source? People still figure a way out to tinker with most stuff.
> How does an Open source hardware company make profit?
The same way an open source software company does, which is to say, there's no definitive answer. If you're building an open source platform, you're just as likely to make a profit from something in the ecosystem around the platform as you are the platform itself. Sure, you can sell the hardware, but you're likely to make as much or more selling your expertise around the hardware, whether that's in the form of add-ons (think: open core), books and courses and training, certifications for certain compliance needs, or services customizing the platform to meet a specific need for a specific client or industry.
It's worth noting that pretty much every successful open source software company targets enterprise clients. It's (sadly) very difficult to make money in the long term from consumers and hobbyists, because price is so often the deciding factor in their decision making. If you crack that nugget, you'll be among the few.
> If they do want to make profit by going closed source, what is the issue in that ?
Mostly, it's the backlash of a perceived bait and switch, or "openwashing" something that isn't. It's harder to track contributions for open source hardware like it is software, but it's worth noting that if a copyleft license is used, outside contributors would need to agree to the license change (if there were any outside contributors).
It's also, frankly, a value add that goes away. I bought a Prusa 3D printer because of the openness. If that goes away, so too would go away my willingness to pay a premium for their product.
> Does it matter if they go closed source? People still figure a way out to tinker with most stuff.
It depends? One of the most important parts of open source is the ability to freely redistribute my changes. If I improve an open source product, I have a right to share my version with anyone I want, including by selling it. I may not have that right if it's closed.
It's also a slippery slope. The changes hardware manufacturers make to keep people from copying their products also tend to make them harder to tinker with. Or to be sure I could get parts from a different supplier if the original maker went out of business.
I'm sure others can add much more to my comments.
> [quote from the link] “So today, we dial up our vision for universal innovation with a clear strategy to expand our portfolio for professionals, supported by a Series B funding round of $32 million led by the global deep tech investor Robert Bosch Venture Capital (RBVC), joined by Renesas, Anzu Partners, and Arm.”
Remember folks, "VC's rhyme with feces". They will enshittify your business faster than every toilet being used during the Super Bowl.
Arduino is the latest casualty.
Why did they need VC funding?
Imagine the actual innovation we'd see with actual lawless ip laws. Everyone iterating from everyone else. It's how china caught up and in some areas surpassed.
Maybe our IP model is just weights on a runner's ankles.
it's unsurprising though because in general even in the happy case, you have locked an invention to a single company for 28 years and that's an eternity in the modern era. And future slight improvements effectively refresh/evergreen the patent, because there's usually only a couple viable ways of doing something. I think of this as the "e-ink" scenario, where there's this fantastic tech that's locked to a company that wants to work out recurring-revenue licensing models for hundreds of dollars per device per year and such, and slow-walks innovation secure in the knowledge that if a competitor does appear that they not only have years of head-start but could sqush them in litigation.
Then in the unhappy case you've got the places that get a patent for "updates delivered over the internet" or "e-commerce on a website" like the one that tried to shake down newegg. Most companies will just pay up rather than fight it on principle, regardless of how egregious the patent is. And such patents obviously should never be issued but the patent office don't care and the legal system allows jurisdiction-shopping that ensure that once issued, the patents will find a sympathetic venue in a red state that's happy to rake in the court fees. Imagine if it's your small LLC that gets targeted instead of a company like Newegg with enough resources to fight, this is a huge instance of that "big companies exploit the system to squash smaller upstarts" failure mode that reduces innovation.
and really the existence of the system at all essentially ensures at least some friction loss of innovation due to the mere possibility of problem B occurring. Let alone in systems (like ours) where it actually does occur frequently and blatantly.
even when it is not flagrantly obvious like the above-mentioned (real, issued) patents, the standard of "not obvious to a skilled practioner in the art" is not really being enforced, or an incredibly low bar of "not obvious" is being utilized (de facto it only seems to mean prior art). A shit-ton of these things are things that would be obvious to any engineer that sat down and worked whatever problem through. If it's going to exist, it ideally would be very narrowly targeted, and perhaps even the patent duration and terms should be customized to the significance and innovativeness of the invention. If you invent a cancer drug and take it through trials that should be handled differently than "3d printer but with a different kind of head inspired by a pastry machine that builds a print 10% better" or whatever.
unfortunately if you remove the patent system that actually disadvantages you internationally, because you didn't patent your thing! And of course it's against the international treaties/etc (which we wrote and could change but still). Again, this actually prevents you from even toning down the abuse of the system because if a US patent is harder to get than any other, you are just disadvantaging US companies. And if you create an uneven playing field for patent lifetimes/etc then that will ultimately benefit companies who can point to the billion $ they spent in R&D and against the person who invented something in their garage (how important could it really be?) and doesn't know how to sell it in the patent application/etc.
It's a race to the bottom at every level. Which is just an inherent problem with globalization (and the US federal system) in general. If you don't have a "minimum standard" then some places are gonna race to the bottom, and a rational actor is highly incentivized to find loopholes that let them eliminate those "minimum standards" and race to the bottom while everyone else is held to higher, more expensive standards.
The court-shopping problem is really just a microcosm/toy problem of these globalization problems. Same problem, different scale.
Bullshit patents that should have never been granted are. That's the problem.
I have no choice but to live in the world of patents. We have to file patents and protect ourselves from them. One of the reasons for which we have to endure such bullshit is the massive numbers of patents that should have been rejected and never issued.
The US Patent Office has granted so many patents that are, as they say in the trade, obvious to those skilled in the art, that almost every domain is an absolute minefield littered with these bullshit patents.
What do you do if you have to play in these domains? Well, go and file bullshit patents! It's an arms race. And the ammunition takes the form of PDF files approved by the USPTO.
Defending against a claim of patent infringement is very expensive, even if the patent should have never been granted.
In some ways, the bottom line is that you have attorneys doing what they do best: Make a mess out of something that could have been far simpler. You have attorneys at the USPTO working with attorneys being hired to put patents through. Everybody is happy.
And, if litigation happens, they are all even happier, because that's when the big bucks roll in.
I am all for true-invention patents, worthy patents. These require investment, time, effort and true discovery of new things. I am just fine rewarding companies and individuals with 20 years of protection for such work. Anyone who has developed difficult technology understands that it could easily take ten to twenty years to see results. My problem is with the 75% to 95% of patents that should have never been granted.
This is where I buy arduino stuff. https://www.aliexpress.com/item/1005004305055818.html
And yes, the price is $.57 and $1.29 shipping. Its a total of $12.98 for 10 of 'em. And that includes USB cables and headers. And, they're also individually sealed.
And on https://store-usa.arduino.cc/products/arduino-nano?selectedS... , they're only $24.90 EACH. That's $249 not including shipping for the same.
Or perhaps you need an Arduino Mega for 3d printing? Only $12.88 here. https://www.aliexpress.com/item/32909503032.html
You want ESP32 with usb-c ? $34.35/quantity 10 . https://www.aliexpress.com/item/1005005565990528.html
Or perhaps STM32 is more up your alley with arm? https://www.aliexpress.com/item/32792513237.html and that includes the ST-Link hardware programmer and the board for $3.53
And, as long as you ask the dealer for the manual on Ali, they'll almost always provide that. Might be partially or all in Chinese. But again, us DIY makers can afford this. We can't afford the jokes of prices in the US or European markets.
There’s no way it would get me to pay 5-8x over the clones/respins.
>And yes, the price is $.57 and $1.29 shipping. Its a total of $12.98 for 10 of 'em. And that includes USB cables and headers. And, they're also individually sealed.
Yes, I've bought Arduino's from AliExpress. But that link is for a $0.57 "micro usb 30cm", i.e. a cable. The actual boards start at $6.36. Isn't shopping on AliExpress fun?
Even when they work, there are often longevity issues, which make them unsuitable for projects that I am giving to others or that I expect to be using for years to come.
But if I'm doing some experimental project where I want a lot of boards, or where there's a high risk that I'll fry them, the cheapies make sense.
A counterfeit is a product that tries to deceive customers making them believe that it's manufactured by someone else (regardless of the open source nature of its design). Open source hardware design only makes for a part of a commercial product and does not cover manufacturing, supply chain, quality of components, warranty, customer support etc. etc. for which it's important to distinguish who is the actual entity responsible for that product.
I think I've only ever had one genuine Arduino brand device and that was sent to me by a client.
My experience is a bit different, obviously, but I'm happy that yours is better.
I'm not talking about clones in general, though, I'm talking about cut-rate boards from China and such.
My favorite clone has ESD protection on the pins, and I have never seen it get fried, which I have seen with a few genuine arduinos.
If the Western companies stop doing open source hardware, some other companies in China will go on with their own designs. Some have already copied your designs and some others will also make their own and some will copy those as well and also conterfeit other succesful Chinese products. That how it works in China.
The problem with most of these Chinese products is no documentation or useless documentation.
Of course you could buy something from aliexpress for pennies on the dollar and a month of shipping time, but a lot of people don't.
- general higher costs to manufacture in Europe (including labor rights)
- certifications about hazardous substances, radio compliance etc.
- R&D (clones just download the CAD file and load it into the machines with zero costs for product design, development, testing etc.)
- software development of the cores/board support packages (clones benefit from the upstream development with zero cost on their side)
- software development of the tooling (IDE, CLI...) which is a recurring cost as things need to be maintained (clones benefit from the work done by Arduino with zero cost on their side)
- marketing/adv/evangelism/documentation (clones have almost zero cost as they benefit from the long tail of the promotion made by Arduino)
The VC destroyed the Maker Faire and magazine because it did not make enough profit fast enough. Therefore it was more profitable for the VC to destroy the business and cash in the chunks left over.
Venture capitalists are shit no matter how you look at it. And they will take sustainable businesses and dismantle them wholesale to squeeze a few extra pennies.
I hope Make/Maker Faire will never again touch a cent of VC's money.
wait till you hear about private equity
With an open source product, it's like putting lunch on a table at a busy high street and leaving it unattended.
Nobody is going to risk their money only to find out someone took the product, remixed it and started selling at lower price using their access to e.g. large scale manufacturing and so on.
This is especially a huge danger for small business, where they don't have money for lawyers and can't use the economy of scale for their product due to limited funding.
Basically, open source hardware is only viable for rich manufacturers who can use it as a PR tool. Some even cynically try to get young and inexperienced engineers to open up their inventions, just so that they can pick the best ideas and use in their own closed source products.
In short it's a pipe dream.
You can take their improvements and use them in your product too. Here's the case of license abiding, and the whole topic is about it, but in a "good enough case" you may hope for that.
That somebody else will expand the product awareness for your product. They'll try to go forward and find a good way, or get some burns trying something market doesn't approve, and many of that you can use for yourself. You may have a profitable strategy by selling to them, by selling your expertise elsewhere, by finding a niche etc. - or even, having enough resources, by going more aggressively to them.
This forum traditionally thinks ideas are dime a dozen, and (lots and lots of) IP protections in hardware are weird, looking from the software point of view.
And I think it can work if you are trying to make money on services rather than devices used to provide them.
Say you're IoT company, selling open devices that are cloned easily doesn't affect you if you make money on providing best interface for them out there.
I've come to the same conclusion over time. It's really difficult to make money just selling hardware. But making money customizing hardware or basing a software product/service around readily available hardware is a much simpler business model.
You sort of proving my point. If you open sourced firmware, apps, any service software, there is no reason why someone else couldn't run it without having to incur R&D and other costs and take your lunch.
Hardware is rather easy to clone, that's why many manufacturers don't care to submit their designs to Chinese companies. Often it more or less follows the datasheets. Firmware, however, is a completely different pair of shoes.
And that's the problem, asking investors for their opinion. Look, I'm a capitalist as much as the next person but you don't start a company based on highly liberal ideals and then expect nothing to change after bringing on outside investors. If you want your non-traditional business model to succeed, don't hand over your vision to someone else. They will demand that you switch your business model to something that feels "safer" to them, and you lose the main thing that differentiates you from your competitors.
> With an open source product, it's like putting lunch on a table at a busy high street and leaving it unattended.
No, that's not a good analogy. It's like putting lunch on a table at a busy high street and _inviting people to help themselves_. With the hope (not necessarily expectation) of forming a relationship that will benefit everyone in the future.
Even that was a little awkward but more concretely, here's an example of Adafruit's business model. They sell hardware, at prices generally higher than what you'd find for the Chinese clones of their products on Amazon and eBay (not to mention AliExp).
But what sets them apart from the clones is they write great documentation, tutorials, and articles. They produce educational videos and show-and-tells. They highlight customer projects showing all of the neat things people are doing with their hardware. They pay people to write open-source libraries for a variety of microcontrollers and devices that they sell, so that people can use them easily. They host forums, they are active on social media, they speak at conferences.
An open source-business only succeeds when you build a vibrant community around it. If you can't do that, then yes, the Chinese cloners are going to eat your lunch.