Ask Slashdot: Should An Open Source Hardware Project Support Clones?

Long-time Slashdot reader Ichijo has a question about "(not quite) open source hardware": One hardware project that calls itself "open source" doesn't want to make its hardware design source files publicly available because doing so would, in their words, "make it very trivial for e.g Chinese companies to start producing cheap clones... we'd be getting support requests for hardware we had no idea of the quality of." This answer was in response to a request by a user who wants to use the design in his own projects.

Have any other open source hardware projects run into support issues from people owning cheap "clones"? Have clones been produced even without the hardware design source files?
Leave your answers in the comments. Should an open source hardware project support clones?

Comments

Re:GPL vs BSD-etc.

[Z00L00K]

And the chinese clone everything, doesn't matter if it's open source or not.

This isn't a thing.

[DMJC]

Considering you can send chips to China now and they'll offer complete reverse engineering/duplication services. I don't see them keeping the schematic to themselves as being a real solution to stopping the Chinese from knocking off their hardware.

Re:This isn't a thing.

[lkcl]

I don't see them keeping the schematic to themselves as being a real solution

schematics are not protected by copyright law. as in: they are uncopyrightable, by definition of them being a "functional description". it is a common mistake (even amongst the open hardware community) to assume that schematics may be copyrighted. what *may* be copyrighted is for example an aesthetic layout of a PCB, because that is a creative process.

Re:This isn't a thing.

[msauve]

"schematics are not protected by copyright law."

Authoritative citation needed.

Anyone who's dealt with electronics knows the difference between a well drawn schematic and one which isn't. There is significant skill involved in creating good schematics. Else, why are there tools for creating and editing them which let you control the layout? Why not simply enter a netlist and then let some program create the schematic?

Re:This isn't a thing.

[KreAture]

http://www.tiplj.org/wp-conten... Ref section 99.

Re:This isn't a thing.

[unixisc]

Considering you can send chips to China now and they'll offer complete reverse engineering/duplication services. I don't see them keeping the schematic to themselves as being a real solution to stopping the Chinese from knocking off their hardware.

Actually, publishing the spec or even the HDL code (if it's chip level) or schematics (for board level) is something they should do. The Chinese could take it and copy it, but rarely do we see them actually improving on an implementation.

But all FOSS licenses have a disclaimer of no liabilities if the software doesn't work. For hardware, that might be trickier, since it is a tangible thing that a person is getting that is made up of actual matter, and is not gonna be given 'gratis'. Still, any license is bound to say that the hardware designer doesn't own any responsibility for any hardware, except for one made and sold by the hardware designers themselves. So nobody would have had to give any support to clonemakers or their customers.

Re:This isn't a thing.

[PurpleAlien]

"On 30 November 1999, the U.S. District Court in Seattle, Washington, dismissed Mackie claims that Behringer had infringed on Mackie copyrights with its MX 8000 mixer, noting that circuit schematics are not covered by copyright laws."

https://en.wikipedia.org/wiki/...

Re:This isn't a thing.

[Anonymous Coward]

Wikipedia says the thing, but when you follow the cite that claims Seattle, Washington the case turns out to actually be in the UK.

A different link gives a media article that says the also sued in Seattle, but nothing about the ruling.

What matters in a ruling is the details. You're not going to find a case where the judge just says, "Golly, you can't copyright them thar circuit diagrams." What you'll actually find is pages of detailed analysis about why they can't make the claim in a particular case. And obviously, since a circuit diagram does have copyright protection, you misunderstood. Perhaps Mackie made a silly claim such as, "their physical circuit is too similar to our drawing of a circuit," and in that case the judge explains that the a diagram of a circuit only gives you copyright protections over the exact depiction, just as writing a story only protects the exact story and not similar stories. That is a lot more likely.

Even the layout of a phone book is protected. A book containing nothing but uncopyrightable facts still has copyright protection, because there are different ways to display facts. To whatever extent your presentation is distinctive, it enjoys protection. Same as a circuit diagram.

Re:This isn't a thing.

[msauve]

LOL. No, that's not authoritative, it's just someone's personal opinion which happened to impress you. One can always find a lawyer to argue either side of an issue, that's how the system works. Authoritative would be case law, preferably at the appellate level.

Re:This isn't a thing.

[msauve]

No, a link to Wikipedia is not a proper citation. Find the actual case you say supports the claim.

Re:This isn't a thing.

[PurpleAlien]

No, you'll find texts arguing the difference between artistic and literary copyright. If you move the parts around on your copy, artistic copyright is no longer applicable. Literary copyright might still be relevant, but only when the design is not obvious (and those would be covered by patents if it really is novel). I found these three cases described in "Contemporary Intellectual Property: Law and Policy":

https://books.google.fi/books?...

In practice, this means that if you don't patent something non-obvious, your circuit diagram only gets protected under copyright as to it's identical reproduction (artistic) or literary protection in some cases if identical components, values etc. are used. However, since most integrated circuits come with reference designs and recommended values for surrounding components, even those can invalidate any protection under literary protection. Going back to the original topic, the board in question is essentially a couple of big ICs and some connectors. There is nothing special on that board, and if you were to take the design, shift components around, change some values, etc. it would look just like any other board that were to use those three chips to do what the original board would do - just with a different layout. Because the design is trivial, any expert will say that this is how a circuit based on those components would be structured. This is why, in the legal sense of the word, copyright on the schematics gives you very little if any protection.

Re:This isn't a thing.

[PurpleAlien]

In addition to a reply I gave to AC above, this provides some additional reading: https://cases.legal/en/act-uk2...

Look, I'm not saying that a circuit diagram has no copyright; the actual diagram and component layout are under copyright. However, a circuit derived from said circuit diagram doesn't necessarily. As I mentioned above, if you take a generic chip and use the reference design to build up a circuit that contains two or three other reference designs and so on (which is what the original board in the article really is), you can't really claim copyright.

Re:This isn't a thing.

[msauve]

No, you can't copyright the circuit itself, but a circuit is not a schematic. A schematic is a graphical representation of a circuit. The OP's claim was "schematics are not protected by copyright law," which is not true. Schematics are copyrighted. Any protection for a circuit would be via patent.

Not really open source if the source isn't open.

[iCEBaLM]

That said, no, open source hardware projects have no obligation to support anybody, let alone clone makers, but it's not open source if the source isn't open. Meaning they shouldn't actively block clone makers.

Freely Support the Design, NOT the Implementation

[Mageaere]

You can support the design of your open source hardware project. Actively doing this keeps the community cohesive and on point. When it comes to supporting the implementation, If you built it, and you sell it, then you support it. If someone else builds it and sells it then they can support it. If the potential customer buys from someone else and asks for your support because the low cost outfit they bought it from cannot help them, then, charge like wounded bulls, be very suspicious of what your are working with, point out all the problems you so conscientiously find and help out as much as you can. This will offset your costs, make you look good, encourage the customer to buy from you in future and make your low cost competitor look incompetent.

Let yes, support no

[Kjella]

It's not open source if it doesn't come with the complete source that someone else can "compile" into hardware. It's not open source if it comes with a "look but don't touch" license. They're under no obligation to support them. They can trademark their name/logo to protect the brand and their own sales. But they can't stop them. If that doesn't work for them, maybe open source is the wrong business model.

Re:Let yes, support no

[ShooterNeo]

What if it comes with a "look but no exact copies license". (where an exact copy is defined as a product that does not have at least 20% more or different parts)?

What if it comes with a "look but any clones have to be sold for a minimum price" license?

Re:Why would anyone copy it?

[Big Hairy Ian]

Surely if it's Open Source it will largely be supported by the community which means that if you want to buy cheap Chinese knockoffs you can expect to be supported by cheap Chinese knockoffs.

Common problem

[melting_clock]

I've used hardware from several open source, and not really open source, hardware projects that have attracted a lot of clones. The biggest problem is that the clones are often not exactly the same as the official versions. Clone can have higher version numbers to mislead consumers into believing they are better than others. Even something as simple as different connectors catch people out.

Over time it can become a confusing mess for everyone involved, making support of every clone impractical. If you want to offer some help to clone buyers, limiting that support to a list of approved clones, that meet minimum requirements, is a way of avoiding insanity.

Re: Common problem

[oobayly]

I had exactly this - I built some radio controlled pwm-dimmable led light bricks using an Arduino clone. I wanted a hard reset using the fob - long press would charge a capacitor to pull the reset pin low.

It worked fine on some, but not on all. As it turned out some of the clones has a different value pull-up resistor that changed the timing - very frustrating.

Transistor tester

[Anonymous Coward]

There is a project developed mostly by German classical electronics enthusiasts (they do not use the openHardware et al. buzzwords but in fact software as well as hardware files are freely available) developing a cheap transistor tester (actually does a lot more),

https://www.mikrocontroller.net/articles/AVR_Transistortester

There are 100s of different clones of these available in china. Go to e.g., aliexpress and search for "transistor tester". Most of the things that come up there are the mentioned clones. (One example here in case your search gives different results https://www.aliexpress.com/item/Mega328-Transistor-Tester-Diode-Triode-Capacitance-ESR-Meter-MOS-PNP-NPN-M328/32679703774.html ).

Once people find out the source of this they come to the huge discussion thread of the project and ask for help (why does it not work, software upgrades, ...) for their devices. The main guy is actually helpful towards them. In some cases the Chinese clones have been changed in software or hardware and help is not that easy. Looking at the discussion thread this takes significant resources.

Re:Transistor tester

[stooo]

That'S a sucessful open source project :)
A kind of crowdsourcing the manufacturing.

Several problems here

[janoc]

First - it is not really an open source project if it doesn't want to publish the design files/documentation. There is nothing wrong with wanting to keep that secret, but then, please, don't use the "open source" moniker.

Second - yes, the issue with clones is real - just look at Saleae (they produce USB logic analyzer). Their original hardware was widely cloned, because it was basically just a repurposed devboard for a common chip loaded with custom firmware that they made freely downloadable.

That said, the Saleae case also shows how not publishing the design files is ineffective - Saleae didn't publish anything, but all it took for the device to be cloned was someone buying a genuine one and reverse engineering it. It is not that difficult to do if someone really wants to do it.

So in the end who gets punished by the files not being available? Certainly not the cloners but more likely your own legitimate customers who will have more tricky time integrating the device into their own projects or repairing it.

Concerning support of the 3rdparty clones - nobody should be obligated to support unofficial hardware. Just don't be an ass about it, pulling another FTDI (company that tried to sabotage/brick the clones).

Re:Several problems here

[AchilleTalon]

First - it is not really an open source project if it doesn't want to publish the design files/documentation. There is nothing wrong with wanting to keep that secret, but then, please, don't use the "open source" moniker.

That's why I believe this question to /. community is not an honest one. It's like the question is asked because the OP doesn't know how to protect his design from being cloned by Chinese manufacturer when he will reach the actual sampling and production milestone in his project. In fact, everyone contracting fabrication of hardware to the Chinese or anyone else (they are not the only one to manufacture and copy hardware) is having this very same problem open or not open.

isn't that the whole purpose?

[Anonymous Coward]

1. The community creating an open source design is not obliged to support it. Sure not.
2. The reason to go open source is actually to let others make it. Not? And Chinese are included in other.
3. This has worked for a zillion projects. Having cheap Chinese clones is actually a benefit. Universally available (think free shipping) affordable hardware is boosting the project. Yes, there will be support requests. But also there will be talented new members to the project to provide that. Worst case there will be a fork, and a new community forms around the cheap knockoffs.

Vajk

Re:isn't that the whole purpose?

[AHuxley]

Yes most of that was covered with the US issues around the need to reverse-engineer the IBM BIOS.
If its open, everyone gets a look and can create, ship, market.
The only way around that is to ship closed hardware and give each person buying a device a full copy of all useful code and offer support.
The user buying into the product can then create anything they want, the hardware been a pay for dongle.
As for the clone production line, the moment its signed over another factory or even the same production line will be running 25 plus hours a day vs any reported much slower production run :)
So the actual production runs are underreported to match any contract but a lot of extra product is made and the trade dress is altered, renamed.
Any counters or management oversight are fooled, manipulated until a much cheaper product is selling under a new name 100% locally owned.
The investor or inventor is left with an expensive branded product that won't sell and full design and support costs :)
The next investor is invited in to start the cycle all over again with the lure of low costs and local gov investment support.

Support demarc/mpoe?

[grasshoppa]

Why not have a support demarc/mpoe, where regardless of the hardware involved, you choose to operate under the assumption that the asker is using "offical hardware" and proceed from there?

Of course, this assumes we're discussing free support. Paid support? No. You want paid support for an open source hardware project, you buy and authenticate yourself correctly ( using "official" hardware ).

Re:Why would anyone copy it?

[ShanghaiBill]

Surely if it's Open Source ...

Except in this case it is not open source because the source isn't open. Just because you call yourself "open source" doesn't make it so. You have to actually open the source, and they haven't.

Re:Freely Support the Design, NOT the Implementati

[Anonymous Coward]

(sorry for the AC but I used mod-points here)
Usually I agree. However, if price differences start to become, like, 1/10th of original retail for a clone with exactly-same specifications, components and good looking board, or 1:20 for an obvious knockoff, using cheaper components but still fully functional, the original manufacturer does start to get a problem. With those kind of ratios it's hard to compete, saying you deliver the original.
It's the difference between 'Let's protect this expensive little piece of hardware, making sure I don't make mistakes connecting other stuff to it' and 'Meh, if it fails I'll throw it away and buy a new one. It's free shipping anyways'.

Re:GPL vs BSD-etc.

[janoc]

No we can't. Circuit schematics is generally not copyrightable, because it only documents the workings of something else - a physical circuit. It is not considered a work in itself.

Even if it was, then the only thing that copyright license would do is to protect the schematics - not someone reproducing the actual circuit. For that you would have to patent it - which may not be possible (circuit is well known, for ex.) or not practical (patenting costing more than the widget itself). Not to mention that patents are not likely to stop an Asian fly-by-night cloner.

Not an open source issue

[thegarbz]

The Chinese will clone your stuff if there's value in it. Full stop. Being open source just adds potential for the clones to be more likely to work without issue.

Re:Not an open source issue

[Bob_Who]

I like Clones best when they wear big shoes and juggle. The Chinese are really good jugglers and they need not clone what they already prototype and manufacture for the West. Apple begins with Asia. Cupertino has reverse engineered the weak Chinese currency to benefit their labor costs while price gouging their countrymen and keeping that wealth out of circulation, in reserve for share holders. So they send in Clones and we provide the Comedy and Tragedy of the decline of US prosperity. Now send in the clones, and we'll provide the Clowns.

Re:Freely Support the Design, NOT the Implementati

[Mageaere]

And this is why we have patents.
To protect innovators ability to exploit their innovation so that they can be free to innovate some more.
Unfortunately this mechanism has experienced an enormous amount of abuse, and so we come to the counter mechanism of open source.
Thus the best solution maybe to use both..
How you use both is then the question to answer.
This will depend on the technology involved, what component you can choose to control, what components you can make open source, and also on other aspects of your potential business models.

Huh

[buck-yar]

Isn't the whole point of open source to allow anyone to make it themselves?

I see this debated in the Pixhawk circle quite often. The software developers get all upset that chinese vendors used the schematics to produce their own products (??? isn't this the whole point). The 3DR(American) pixhawk costs 2-3x as much as the Chinese version (many wouldn't even own a Pixhawk if they had to buy it at those inflated prices). While the developers make all these claims about how their product is QA tested, they still have all the problems the chinese ones do (or one could say the chinese version has all the problems inherent in the design).

For example, the 3DR version had IMU1 problems because the design had the chip too close to the edge of the PCB, and the vibrations from cutting ruined the chip. The American version has that issue, but the newest 2.4.8 chinese version moves that chip inwards. Wait- the chinese improved upon the design, shocking! The american developers that rip into the product non-stop never mention that.

Or the IMU2 problems with the chip stuck in a brownout state (apparently a very common problem with the LSM303d accelerometer). The original open source design doesn't provide proper discharging of the sensor rail. Chinese fixed that as well

Clones? That isn't very accurate terminology. More like "forks." The 3DR Pixhawk left many issues unresolved, years after they were discovered. Now 3DR has stopped making the original Pixhawk, so if it wasn't for the Chinese forks, there wouldn't be any more. That doesn't get mentioned as a plus either.

Re:Huh

[drinkypoo]

Any input on a Mini PX4? I've been putting off finishing a quadcopter based around one of those and a readytofly 4in1 ESC. I've never messed with Pixhawk before specifically because it is so very expensive...

Re:Huh

[c]

Clones? That isn't very accurate terminology. More like "forks."

IMHO, the threshold between "clone" and "fork" is whether the forks are available to the original/main branch.

A major beef is that these cloners take the designs, but contribute nothing back.

Re:RTL network chips

[ruir]

All as shittier than the original apparently. The network cards are nothing stellar, their wifi chips are pure garbage.

Re: Why would anyone copy it?

[oobayly]

It's quite impressive that these people have succeeded in getting people to describe their product as open source.

Simple solution : Keep the name

[stooo]

Usually, when you open the source and want to market it, you trademark the product name, so everybody can copy and clone it ( which is the purpose of open source), but not under the name you control. So they will not be able to taint your reputation.

Simple as that.

Re:Simple solution : Keep the name

[Jeff DeMaagd]

That works if you do have the means to defend it. But the means to defend it get pretty thin when you're talking about clones selling on Alibaba. Even if you have legal resources in China, they're going to use your brand. Even those only selling very different competing designs will use your brand.

Re:Simple solution : Keep the name

[stooo]

you don't really compete with them because you don't sell on alibaba.
And also, their customers will buy the same cheap stuff, regardless of the brand.

Re:Freely Support the Design, NOT the Implementati

[perpenso]

And this is why we have patents. To protect innovators ability to exploit their innovation so that they can be free to innovate some more.

That's one reason. Another reason is so the innovators can recoup the investment they made for their innovation, the R&D. A cloner only has to pay for the marginal cost of production, not the original R&D expenses. Without the ability to recoup R&D expenses it would be nearly impossible to find investors will to fund an endeavor.

Schematics not needed

[CODiNE]

Have clones been produced even without the hardware design source files?

Certainly. Clones can be made at any level of the hardware. The boards can be reverse engineered, entire chips can be cloned at various levels of the design. Portions of a chip that perform a certain feature can be clones, let's say a media decoder or an encryption function. Those blocks of logic are licensed out and pretty expensive, so they have watermarks and fingerprinting codes hidden within their circuits. A company can save millions by copying chunks of IP without paying the designers of it. The tricks is not getting caught.

It's a fascinating game of cat and mouse on the hardware side, there are ways to prevent people using parts or all of your hardware but it adds overhead and complexity. Both sides have a trade off on how much it's worth making or breaking protections based on projected sales and how much it costs to add or defeat a protection.

For an open source hardware project they'd probably just use a watermark and not even bother with fingerprinting individual chips, leaving it wide open to being copied. They don't have a budget for many security techniques.

Of course not

[drinkypoo]

Support them? No. You provide a forum and if the community wants to support them then it can. But equally you must not go out of your way to impede them, or you will destroy your community. Arduino would be less than half as popular today without cheap chinese clones.

Re:Why would anyone copy it?

[perpenso]

Surely if it's Open Source ...

Except in this case it is not open source because the source isn't open. Just because you call yourself "open source" doesn't make it so. You have to actually open the source, and they haven't.

Its possible for the software to be open source but not the hardware. For example the drivers for a device may be open source, the system software for an appliance, etc.

Re:Not really open source if the source isn't open

[RDW]

I wonder if some developers genuinely don't understand what Open Source means, and just use the tag as a convenient buzzphrase? Describing projects that aren't really Open as Open Source is a problem that goes back to the early days, and affects software as well as hardware. e.g. LinuxSampler still defiantly claims to be Open Source a decade or so after after being dropped from major Linux distributions because it clearly isn't (and has a contradictory licence that doesn't make it properly non-free either).

What about USB PID/VID numbers?

[Rufty]

I've got a software+hardware project I'd like to (when it works) release as BSD/GPL/open-source. I'd be flattered if it was popular enough get knock-offs and derivatives. But I don't want to have to deal with my software not quite working right for these third party hardwares. So can I restrict use of my software+firmware+hardware to, "yes, here's all the code and design, open source, but if you change anything don't use my VID+PID, get your own"? Is this still open source, and can I still use the GPL? (BTW, I got my VID+PID from http://www.mcselec.com/ for 15euro, and there's open hardware projects, OpenMoko, I think, that also provide them.)

Re:What about USB PID/VID numbers?

[guruevi]

Your VID/PID can/should be a non transferable part of your system, GPL covers code not design, look into a Creative Commons that fits (e.g. Non commercial use etc) . Clones don't really care about licensing anyway and you can't stop the clones from China in court.

Re:Not really open source if the source isn't open

[perpenso]

That said, no, open source hardware projects have no obligation to support anybody, let alone clone makers, but it's not open source if the source isn't open. Meaning they shouldn't actively block clone makers.

The software side of the project may be open source. The drivers, system software, etc. That would make it fair to describe a project as open source.

Also to expand on what you said, open source does *not* mean making your source code or hardware design universally available. You are only obligated to provide it to your customers, albeit with no restrictions on redistribution. Therefore it would be entirely consistent with open source to verify someone is a customer before providing any support. Furthermore it is also entirely consistent with open source to charge for any technical assistance beyond providing the source code or hardware design.

FWIW, the company could charge for technical support of non-cusomters and refer these paying non-customers to their hardware supplier if the hardware is in question. We sort of did that at a company I did technical support for long ago when I was starting out. Not many non-customer paid for support but very few were angry since they understood they had not purchased anything from us. It probably helped that these were more technical folks and not the public at large.

Re:Freely Support the Design, NOT the Implementati

[perpenso]

not the original R&D expenses.

Since patents are rewarded regardless of the R&D cost, even if the cost is zero, it's pretty disingenuous to claim that patents are needed to cover R&D cost.

No, as the preceding two posts demonstrate recouping R&D is not the sole motivation for patents. The fact remains that some innovation requires a substantial investment in R&D, and without an ability to recoup that R&D the innovation will likely remain unexplored. Recouping R&D is essential to supporting some innovations, patents help make this possible, therefore patents support innovation.

That said, yes there is abuse in the patent system, the awarding of unworthy patents. However such abuse does not change the fact that patents provide an important tool that supports innovation.

experience from another OSHW project

[Anonymous Coward]

I'm member of a tiny OSHW team with a well known product and tons of clones on eBay, Banggood, Alibaba and so on. Actually, we're happy to have the inexpensive clones. It's more expensive to make the device yourself (the project was intended as a DIY project for electronics hobbyists). Also we don't have to deal with any manufacturing and certifications (FCC, CE etc.). We don't sell anything, just provide the schematics, firmware source, documentation and some support. So we have more time for improving the firmware and adding new features. Meanwhile there's an active community supporting newcomers and discussing which clone is the best at the moment, for example. Some have great ideas for new features, some support us with small donations like component samples or LCD modules. The downside of the clones is that a lot of the vendors don't provide any documenation or a link the project's web page. Some modify the firmware, predening it's theirs, and don't submit their changes to the project. Several lock the MCU to prevent users from reading the firmware (WTF! It's open source!). But the clone quality is quite good in most cases. In the beginning we've seen a few small issues with badly choosen components which could be fixed easily with a soldering iron.

Re:Not really open source if the source isn't open

[jcdr]

Whenever it's for hardware or for software, the point to make a project open source is to create a community to support and improve the project, not to make money. Starting from that point I found cheap clone a rater big advantage, because this offload the production problem and make the project accessible to more and more peoples with almost no management. Instead of fighting against clones, it better to take advantage of them.

Re:Not really open source if the source isn't open

[GrumpySteen]

While you are correct that the word "open" has been used (some would say misused) that way, the phrase "open source" clearly has a different meaning and refers to making the source available, not just access to an API.

AUTHENTICATION (and "no")

[gavron]

If they are an open-source hardware project but haven't released the spec, then THEY'RE NOT OPEN SOURCE.

If they release the hardware specs but sign their own hardware and don't release the key that's still fine.

They can then request anyone who wants "support" (the question in this /. posting) to provide authentication of OEM purchase.
No auth - no support. It could even be automated. Just like Motorola's "Can my device be unlocked?" site, they could have a
"do I have a Genuine OpenSourceHardwareProject Board eligible for support?" page.

Everything else is dissembling (that's the politician's word for LYING and EVADING).

Either they REALLY ARE OPEN-SOURCE HARDWARE PROJECT (no, right now they're not), or they are.

E

That's not an authoritative citation.

[tlambert]

http://www.tiplj.org/wp-conten...
Ref section 99.

That's not an authoritative citation.

It's the opinion of David G. Luettgen in a journal article, which claims computer programs are also not copyrightable. It also claims in its conclusion that electrical circuits are a creative expression, while computer programs are not.

The guy is kind of talking out his arse.

Re: Why would anyone copy it?

[Anonymous Coward]

Agreed. Also, reference isn't implementation. If you're in the business of selling and supporting artifacts, identify and police your work. If you rely on the community to support it, it should also be able to accommodate clones.

If you're worried about clones flooding the market (a sign you're doing something right), use a trademark. Trademark law is rather impressive at what it can accomplish.

If a community is fearful that clones will kill innovation (meh), then shun the clones with the tools you're familiar with. Otherwise, suck it up and expand your insular world to include clones; you might find that they add to the culture.

I thought that was the "open" / "free" distinction

[Ungrounded Lightning]

It's not open source if it doesn't come with the complete source that someone else can "compile" into hardware. It's not open source if it comes with a "look but don't touch" license.

Huh?

I thought that was the distinction between "open source" and "free foo":

"open source" software MIGHT come with a license that ALSO makes it "free software", but "free software" includes an explicit license to use and modify (with the only restrictions being some variant on requiring derivative works to also be open - which may include not enforcing patents against license-conforming users ) while "open source" may only guarantee you can read and pass on the design, not that you can modify it, run it, "practice the invention" or whatever.

Of course open source software is USUALLY also free software, or close to it, depending on the license.

Has the language drifted? Did I misunderstand from the beginning? Or is the second part of your assertion mistaken?

Wouldnt' be in arduino without clones

[EmperorOfCanada]

I have used and loved arduino for a while. Without clones, I quite simply could not have afforded to buy them in the quantities that I needed to make them useful. And please don't give me any crap about supporting the originators, if that is something that you want to do, then go right ahead. My choices are not going to be affected by others' whining.

I find it curious that there aren't any real raspberry pi knockoffs. I really would love me some of those, not out of some dislike of the raspberry people, but because of the supply chain in between. I can't get a raspberry zero, and even if I could it would probably be in the $20 USD range delivered. It is close to $50 USD for a Raspberry Pi 3 from my local store. If I could get a Pi knockoff from China I am willing to bet that it would be $20 delivered. I feel jerked around by the suppliers of Arduinos and Pis, Ebay/alibaba is "You give us a minimal amount of money, and we reliably(and slowly) ship you a product with a very modest markup."

Re:fucking clueless

[Anonymous Coward]

open source does *not* mean making your source code or hardware design universally available.

You are only obligated to provide it to your customers,

what a LOAD OF BULLSHIT, maybe you should actually READ the software licenses you claim to be an expert on

Seriously, take your own advice. For example the GPL only requires me to provide source to someone I provide a binary to. I am under no obligation to make the source code universally available. I am free to do so for my own convenience, but the terms of the GPL do not require it.

supporting is not the same as making open source?

[drolli]

Is Linux "supporting" crappy compiled linux kernels or just "enabling" them?

My experience that there are crappy vendor clones of linux of (i.e. terribly hardware specific, marginally documented, buggy, not maintained), but i have never seen the crap they do being "supported" in the way that it would have made it's way into the mainline (yes, that is what "supporting" means).

That's the whole point, isn't it ?

[alvieboy]

First, let me try to shed some light on what "hardware project" is, comparing it to a "software project". But before that, let me introduce myself, and introduce what I do in regards to Open Source, and my still active projects.

I am the author of ZPUino, which is a SoC (System on a Chip) targeted at FPGAs although it can be built on an ASIC. ZPU (Zylin CPU [1], which is the "core" of ZPUino) was not designed by me, in terms of its ISA (Instruction Set Architecture). The ZPU core inside ZPUino is however much different from the original ZPU, featuring a fully pipelined design and yielding very very good performance, whilst maintaining the "small" footprint as originaly designed. It would not be possible to design, implement and "ship" this version of ZPU unless Zylin had a highly permissive license - BSD.

ZPUino merges this enhanced ZPU core (ZPU Extreme core, written by me) with a huge set of devices, as commonly seen in a SoC. So we have, as open-source hardware: UART SPI Timers Interrupt LED HDMI VGA I2C, Memory SRAM SDRAM DDR plus many other eccentric controllers you cannot find in regular SoCs like those in rPI. All those are Open-Source, and the HW design is released on BSD license - so anyone can benefit from them even without giving back. [sorry for lack of commas, the lame filter kicked in]

This is a hardware project. The designs are hardware designs, and despite being written in VHDL, does not make it software. You can not say that, since it's not a printed circuit board, and no wires to see, that it is not a hardware project. Hardware projects describe hardware primitives and interconnections.

Still, they are described using languages, much similar to how software is (for example, VHDL is very close to ADA, which is still widely used in the space industry). PCBs, schematics, can also be described in languages (think EDIF) - as well as their outputs (thing GERBER and DRILL). So there is no much difference between software and hardware here.

Now, back to the "cloning" topic: someone said "chinese clone them all, does not need to be open source" - and this is correct. Your design is not protected just because you did not open it. If you require protection, seek patents and trademarks. And if someone massively clones your HW and SW, you're a hell of a lucky guy you made something people want (cause it does sell, otherwise no one would clone it), you just seem to miss the target price point.

Plus, you can for sure give added value from buying the original product. Arduino (they seem to have reach an agreement today) sells their own HW at 20x price you can buy from china. Still they do sell, and they are not bothered by it - it is expected.

If you want to go open, go open and they clone. If you want to close it, they will open it and clone. How can you benefit from all those clones ? That's the big question.

Alvie

Re:Not really open source if the source isn't open

[iCEBaLM]

The software side of the project may be open source. The drivers, system software, etc. That would make it fair to describe a project as open source.

Sure, the software side, yeah, but this is called the Open Source Scan Converter, not the Open Source Scan Converter Software, or the Open Source Scan Converter Firmware. The question was specifically about hardware projects. If you're billing yourself as an open source hardware project, then you have to make the source hardware designs available or else it's just not true.

Now I'm not seeing anywhere where the OSSC guys have billed the hardware as open source, there's no announcement, no web site explaining their goals and vision, just some guy who made a board and is selling kits and giving away the firmware, so I'm not convinced it really applies here. That said, if you are touting yourself as an open source hardware project, and the information on how to build the hardware isn't open, then you're just lying to everyone.

Re:I thought that was the "open" / "free" distinct

[cheesybagel]

You are wrong. The OSI definition of open source software is basically the same as the free software definition:
https://opensource.org/osd

The license must allow modifications and derived works, and must allow them to be distributed under the same terms as the license of the original software.

The license must not discriminate against any person or group of persons.

The license must not restrict anyone from making use of the program in a specific field of endeavor. For example, it may not restrict the program from being used in a business, or from being used for genetic research.

etc.

Re:I thought that was the "open" / "free" distinct

[Ungrounded Lightning]

You are wrong. The OSI definition of open source software is basically the same as the free software definition:...

Thank you.

Any third party

[tepples]

The source code only has to be available to the people you ship binaries to.

Consider the following excerpts from the GNU General Public License, versions 2 and 3:

(From GPLv2)
b) Accompany it with a written offer, valid for at least three years, to give any third party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange

This offer must be available to "any third party".

(From GPLv3)
b) Convey the object code in, or embodied in, a physical product (including a physical distribution medium), accompanied by a written offer, valid for at least three years and valid for as long as you offer spare parts or customer support for that product model, to give anyone who possesses the object code either (1) a copy of the Corresponding Source for all the software in the product that is covered by this License, on a durable physical medium customarily used for software interchange, for a price no more than your reasonable cost of physically performing this conveying of source, or (2) access to copy the Corresponding Source from a network server at no charge.

This offer must be available to "anyone who possesses the object code", including someone who received a copy other than from you.

Re:Any third party

[Yaztromo]

The source code only has to be available to the people you ship binaries to.

Consider the following excerpts from the GNU General Public License, versions 2 and 3:

(From GPLv2) b) Accompany it with a written offer, valid for at least three years, to give any third party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange

This offer must be available to "any third party".

Good job at selectively quoting the GPL to fit your narrative, but you're still wrong.

The section of the GPL v2 you quoted above is section 3b. The main text for Section 3 just above it reads (with emphasis added):

3. You may copy and distribute the Program (or a work based on it, under Section 2) in object code or executable form under the terms of Sections 1 and 2 above provided that you also do one of the following:

...followed by three options, one of which is the one you quoted. There are two other options that be be selected instead, keeping you in compliance with the GPL, neither of which requires you give source to third parties.

Yaz

Re:Any third party

[tepples]

In GPLv2, the other options are a), to distribute every copy of the object code with a copy of the complete corresponding source code and not distribute the object code in any other manner, and c), to forward the offer you received. Option c) is not valid for commercial distribution. So for the avoidance of doubt, you are recommending option a), correct?

Re:Not really open source if the source isn't open

[RDW]

So, not open source, eh?

No, not Open Source, which rather makes my point. The Open Source Definition explicitly forbids restrictions on commercial use ('No Discrimination Against Fields of Endeavor'). The developers claim there are 'many definitions' of Open Source, which isn't correct for software, and the current version of the Wikipedia page they link to doesn't support this claim. This is not the main problem with the LinuxSampler licence, however. Instead of writing their own licence, they've added a restriction to the GPL, which isn't allowed if you still want to call it the GPL. This means that their licence contradicts itself, which is one reason why you won't find Linuxsampler in (say) Debian non-free. The developers are of course perfectly free to distribute their software for non-commercial use only, but they just haven't gone about it in the right way. They should just drop the 'Open Source' claim and come up with their own licence (which could be based on the GPL, but should not include the preamble, GPL branding, or anything else that conflicts with their non-commercial clause).

Re:GPL vs BSD-etc.

[lsatenstein]

No we can't. Circuit schematics is generally not copyrightable, because it only documents the workings of something else - a physical circuit. It is not considered a work in itself.

Even if it was, then the only thing that copyright license would do is to protect the schematics - not someone reproducing the actual circuit. For that you would have to patent it - which may not be possible (circuit is well known, for ex.) or not practical (patenting costing more than the widget itself). Not to mention that patents are not likely to stop an Asian fly-by-night cloner.

A question
If a patented software algorithm is redrawn as a circuit diagram, or as a source code listing, or distributed as source code text, then is it truly open source? Of course, if you compile the code, that code would be executable. But... The vendor of some fancy hardware would not be selling the algorithm, only giving away the description.
But you mentioned that the circuit diagram is not copyrightable. If so for circuits, then so for source code listings.

Was not PGP initially distributed that way?

Re:Why would anyone copy it?

[suutar]

This particular project is pretty much hardware, unless you're counting the FPGA config as the software. Check out http://retrogaming.hazard-city... for a description (it's a review, but it has pictures and some explanation of functionality).