Inverted OSHW Project Funding

[Nominal Animal]

I am personally much more in the software side and hobbyist electronics projects, but here's two that I came up with one morning this week:

• Arduino-programmable 32-bit ARM programmable graphics processor board for ribbon-connected ILI9341 320x240 display modules
• Arduino-programmable 32-bit ARM programmable audio processor for headphones, line out, or 8Ω or 16Ω mini speakers (configurable in hardware)

Both are designed to operate as a I²C or SPI slave, or via UART.  They are multi-purpose development modules.  The closest equivalent to 1 are the Digole display modules, but they use a PIC18F.  The closest equivalents to 2 that I know of are the various audio carrier boards, like the Teensy Audio board, but they usually don't have an on-board microcontroller.

The key to commercial success is in the firmware.  Or, libraries rather, as they are intended to be user-programmable.

The first one has two example firmware/library implementations.  One is a simple serial/VT200/xterm console for debugging microcontroller projects and hacking Linux SBCs and embedded systems like routers and switches.  The other is an SDL-like graphics library with tile/sprite support, color LUTs, and a simple blitter.  This can be used to create old style handheld games, or nice UIs.  The ILI9341 is used in 16-bit color mode (R₅G₆B₅), and a 32-bit microcontroller with hardware multiplier is needed for alpha support and blending.  The main issue is the amount of RAM a good implementation would prefer.  (I have a lot of the firmware/library details worked out already, and am familiar with efficient routines and algorithms to do all this.)

The second one has two example firmware/library implementations: a MOD/S3M/XM/IT tracker, and a MP3/AAC player.  Again, having a 32-bit microcontroller with fast hardware multiply is a key.  It may even be possible to merge the two into a single library/firmware.  The tracker is basically the audio module for old-style games, and for sound effects.  The MP3/AAC player would best be illustrated with a SPI Ethernet module (although the audio processor would have to be SPI master in that case), streaming music from free internet radio sites.

To explain why this idea makes commercial sense, we'd have to delve into existing use cases and implementations and their downsides, requirements for the above hardware modules (for example, you definitely want the MCU in 2 to have two hardware SPI channels, for obvious reasons; and at least one must be slave-capable), and so on.  If you are interested, I guess you have to come to Finland and buy me lunch, and I'll answer any of your questions on it, as I am not interested in developing it any further.  Not because it does not have potential, but because I've already burned out twice, and keep away from the business side completely, to retain whatever sanity and capabilities I have left; I am just not a business driver type of a person.

[Nominal Animal]

2) I don't believe that open hardware projects are (or even should be) economically viable. The same holds true for open source software. Yes, there are companies that make great money using or supporting OSS project but the open source paradigm is ideologically driven and not by market (except those shitty companies abusing OSS for pure marketing reasons with NO value to the community). When Linus Torvalds created Linux there was initially no market value in it. At the very beginning it was practically unusable in any commercial setting. Yet, people collaborated around the globe to create what Linux is today.

I disagree.  Even your description is in contradiction with your initial sentence there.

Open source is economically viable because of, and through, market competition.  Let that idea sink in a bit, and I'll elaborate.

If your commercial plan is to mass-produce a widget, and sell it for less price but more value for customers than your competitors can, open sourcing it makes absolutely no sense.  However, that is not the only way to do business.  It is just one of the ways of doing business that is not compatible with open source.  And, as we have seen from the Chinese manufacturing abilities, it is not even a long-term viable way of doing business: someone will eventually find a way to mass-produce an even better version of the widget for smaller price.

The "ideology" in Open Source is market competition: various ideas and implementations are tried and tested, and the "best" ones are incorporated into the common pool everyone can use without draining it.  The currency is not about the end widget price, but about the usability and worth of the common pool.

(To go back to the town well analogy, keeping the water pure and flowing is good for everyone.  If you think the only way to make market sense of such a well is to sit on it and charge money for access to that water, you are an idiot, and will find out the town will eventually just simply move somewhere else, making your investment worth nothing in the long term.  Instead, you can for example bottle that water, and sell it outside the town.  Or, you can provide water quality monitoring services for the town.  It is very hard to think of new business models when you've only sold widgets or charged rent, but that does not mean doing so is the only, or even the best, business strategy.

If some of your fellow townspeople are happy to piss in the well, you have a problem.  If someone installs a pump and pumps away all the water, you have a problem.  If someone spreads rumours that this well is poisoned, you have a problem.  If nobody cares about the water table level, you have a problem.  This means, you need careful management of the well to keep it viable in the long term.  Unlike wells, FOSS and OSHW projects are easily forked; they key difference is that the underlying resource cannot be depleted, and reproduction costs are essentially zero.)

A proof of this is the Apache HTTP Server.  Its license allows proprietary closed-source derivatives.  There used to be those around, but they could not compete with the open source version, or even keep up with the development of the open source version and its features.  License cost (for the proprietary closed-source versions) is not an issue, because those who really need a good web server and advanced features, often make a LOT of money from the service; and license costs are neglible compared to the related hardware and network traffic costs when implementing such a service.  In fact, as far as I can tell, Microsoft IIS and LiteSpeed Web Server are the only proprietary web servers still actively developed (Apache and nginx covering over 80% of the market).  IIS is strictly tied to other Microsoft products, and LiteSpeed seems to be preferred by QUIC users; they are both effectively niche products.

If your theory of FOSS/OSHW not being economically viable was correct, Apache and nginx -- or Arduino! -- would not exist.

[bloguetronica]

I'm amazed that this thread is still alive and kicking. But anyway, judging by the opinion of all members except the OP, it is clear that this idea doesn't have any practicality. To put it simply, people don't want to work for freeloaders, with OSH or without it. It is human to expect something in return, that being money, recognition or whatever. What the OP is proposing is to someone work for free while the team leader provides the idea and takes the credit for it (this kind of communism frequently happens inside big tech corporations, BTW - just read the "About" box of any known paid software and see if you get an extensive list of names).

If you don't have the means to realize your idea, then, either:
1- Your idea is not practically feasible (contradicts laws of physics, or whatever);
2- You don't have the knowledge and/or the means to realize it, which means that you'll need to acquire those (eg: by going to a library, buying equipment).

The second hypothesis is the doable one, but requires effort. Success comes with a lot of effort, sacrifices, and a truck load of failures. If God intended the world to be free, there would be food falling from the sky and the climate control would be provided by an air conditioner. Life is certainly not a picnic and we are here to toil. Whoever is against this philosophy probably lives like a parasite.

...
If your theory of FOSS/OSHW not being economically viable was correct, Apache and nginx -- or Arduino! -- would not exist.

Second that!

Kind regards, Samuel Lourenço

[floobydust]

This entire thread makes no sense at all.
OP's notion is the chef and kitchen staff cook up and serve for free whatever the customers (idea people) come up with. It seems a little unrealistic.

Apache and nginx are not hardware. I think OSHW is a flop, it rarely churns something out and proof is the threads here with tons of energy about building X and then the one-man  carrying the work collapses and the project dies.

Crowd-sourcing and design by community in theory is a great way to go but it doesn't seem to work for hardware.
"A camel is a horse designed by a committee" should sound familiar.

[NorthGuy]

The "ideology" in Open Source is market competition: various ideas and implementations are tried and tested, and the "best" ones are incorporated into the common pool everyone can use without draining it.  The currency is not about the end widget price, but about the usability and worth of the common pool.

There's no competition here. The "worst" ones are incorporated into the common pool just the same.

Competition implies a limited resource, and the only useful consequence of competition is the death of the unfit which frees up the resources for survivors.

[homebrew]

I think OSHW is a flop, it rarely churns something out and proof is the threads here with tons of energy about building X and then the one-man  carrying the work collapses and the project dies.

This is a strong motivation for a better system for collaboration. Exactly as you say - it is often too much for one person to chew on.

Crowd-sourcing and design by community in theory is a great way to go but it doesn't seem to work for hardware.
"A camel is a horse designed by a committee" should sound familiar.

Seems to be a common theme here.

OP's notion is the chef and kitchen staff cook up and serve for free whatever the customers (idea people) come up with. It seems a little unrealistic.

Now why would that be the case? I've not described it that way. No, of course not! If you won't believe in an idea you would OBVIOUSLY NOT participate and NOT invest ANY resources. Why would anybody seek out for collaboration if he/she is not interested in some aspects of the design?

What I've read several times now is the opinion that only the person supplying the idea would earn the credit. For example:

What the OP is proposing is to someone work for free while the team leader provides the idea and takes the credit for it (this kind of communism frequently happens inside big tech corporations, BTW - just read the "About" box of any known paid software and see if you get an extensive list of names).

But who wants that? Just because companies often declare contributions like that, it doesn't mean that we can't do better. Of course credit must be liberated throughout the entire team.

Whoever is against this philosophy probably lives like a parasite.

What a great way of discourse! Cool - whoever is against your worldview lives like a parasite! Just wow ...

Sorry, but many people clearly don't think that way. They do all sorts of things as engaging in foundations, associations, donate money, doing charity work and whatnot JUST because "the market" does not or cannot fix some problems in the world these people care about. It's about intrinsic motivation. Money can only provide extrinsic motivation ...

The "ideology" in Open Source is market competition: various ideas and implementations are tried and tested, and the "best" ones are incorporated into the common pool everyone can use without draining it.  The currency is not about the end widget price, but about the usability and worth of the common pool.

There's no competition here. The "worst" ones are incorporated into the common pool just the same.

Competition implies a limited resource, and the only useful consequence of competition is the death of the unfit which frees up the resources for survivors.

Second that! Just look at GitHub and the metric ton of half-baked, abandoned and unfinished stuff. It is there, side by side, the same way as multi million dollar projects. The only limited resource seems to be people willing to engage into a project to keep it alive. Which brings us back to the topic that people collaborate and donate their resources if and only if they are convinced by the idea.

[Nominal Animal]

There's no competition here. The "worst" ones are incorporated into the common pool just the same.

No.  The "common pool" does not refer to all software that has ever existed, but to the set of actively developed software (with compatible licenses) useful to its developers.  Just like the town well does not refer to any old puddle of dirty water.

Competition implies a limited resource

Quite.  With FOSS/OSHW, the limited resource is time and effort needed to maintain and develop a project.

Fundamentally, the number of users a FOSS/OSHW project has is irrelevant.  Only the developers matter, because they are spending the limited resources.  The only sane reason to spend that time and effort, is usefulness -- useful to its developer(s), that is.  These form the market: the limited resource, the value sought after, and the competitors.

This is post-scarcity stuff, with basically zero reproduction cost.  (The number of users might matter to some of the developers for various reasons, but unlike in commercial proprietary projects where an user corresponds to a sale, an user by itself in FOSS/OSHW corresponds to nothing.)  Not understanding this has tripped untold numbers of people and businesses.

[SiliconWizard]

Second that! Just look at GitHub and the metric ton of half-baked, abandoned and unfinished stuff. It is there, side by side, the same way as multi million dollar projects. The only limited resource seems to be people willing to engage into a project to keep it alive. Which brings us back to the topic that people collaborate and donate their resources if and only if they are convinced by the idea.

Short of a very few exceptions, this is the destiny of most open source projects, even those that were once popular, unless some big company/foundation comes in and keeps them alive thanks to some form of management, marketing and significant direct funding. Alright, you may say that it's still open source and doesn't change the philosophy of it - but that's definitely not projects that just live thanks to the good will of users and developers.

Ideas are nice, but are not enough to keep developers interested in the very long run, and projects that are not maintained regularly end up abandoned by their users.

[NorthGuy]

Competition implies a limited resource

Quite.  With FOSS/OSHW, the limited resource is time and effort needed to maintain and develop a project.

Of course, but open source projects, even taken all together, are incredibly small player on the workforce market. The majority of juicy developers get hired by companies paying real salaries. Others start their own businesses (if they believe their idea is worth the try). Whatever left for open source projects is just peanuts.

Now and then, a group of people takes over an open source project with a commercial intent, starts paying money, so the project gets more (paid) developers. At the same time, the people who contributed to the project earlier are left behind - they don't get a penny and they cannot contribute to the project any more.

... unlike in commercial proprietary projects where an user corresponds to a sale, an user by itself in FOSS/OSHW corresponds to nothing.

You bet. Otherwise, things like Systemd (which is bloated, useless, weakens security, but must be fun to develop for neophyte programmers) would never take off.

[bloguetronica]

What the OP is proposing is to someone work for free while the team leader provides the idea and takes the credit for it (this kind of communism frequently happens inside big tech corporations, BTW - just read the "About" box of any known paid software and see if you get an extensive list of names).

But who wants that? Just because companies often declare contributions like that, it doesn't mean that we can't do better. Of course credit must be liberated throughout the entire team.

Whoever is against this philosophy probably lives like a parasite.

What a great way of discourse! Cool - whoever is against your worldview lives like a parasite! Just wow ...

You are a funny guy. Obviously, either you didn't read the post, or you didn't had the cognition to interpret it correctly, despite the fact that I was crystal clear and left no room for other interpretations. Well, I don't give a damn about your interpretation, or by the fact your ego got triggered. I'm not PC, anyway. And I had my share of freeloaders, so I stopped contributing to those a long time ago.

Anyway, everyone in this thread disagrees with your idea. That should be evidence enough for you to start thinking that your idea is not feasible, let alone sustainable.

Kind regards, Samuel Lourenço

[djacobow]

Interesting thread. Also interesting that so many folks find (good) ideas easy to come by.

I've got a decent lab, some free cash, some free time, and pretty good skills (digital and software, at least) and I have to say, I have a really hard time coming up with commercially interesting ideas that I can swing with the resources above.

I've designed and sold a few simple devices. I found the effort to be pretty intense, honestly, and the returns meager. It pisses off my wife when I'm sweating the details of how I'll manage the BOM or assembly or testing or channel cost of X when she knows the chances of X materially changing our lives is practically zero! It's not good to piss off your spouse.

Bringing something to market -soup to nuts- all by yourself is hard, and for it to have even a chance of being worth it (financially), the idea needs to not be good, but forking brilliant.

Honestly, I've had more fun designing quirky things that entertain me, and just giving them away to friends and family. That can also be cheaper than building 5000 of something you think the world needs and finding out how wrong you were.