Open Source Hardware
OSHW Definition: We take a crack at saying what it means to make a piece of hardware "open source".
Alicia Gibb and Ayah Bdeir put together and hosted the Open Hardware Summit a few days before Maker Faire New York. It was an amazing event! Almost all the usual suspects were there, too many to list here, but it was great to hear from everyone (over 300 people attended) on their thoughts on what Open Source Hardware should be. There is a very active group of folks (namely Windell Oskay, David Mellis, and Phillip Torrone) who are hashing out what OSHW should be, what it should stand for, and how we should define it as a group. It's fairly amazing to me to watch our community debate cordially and tirelessly towards this goal. I am proud to be a small part of a group that can get along so well and create such amazing art, ideas, and businesses.
The Arduino Fio - an OSHW collaboration co-designed by Shigeru Kobayashi.
We've been hammering on the definition a lot. Version 0.3 of the Open Source Hardware definition is here. The version 0.4 draft is here. We would love your help making the definition better.
The OSHW statement of principles:
Open source hardware is hardware whose design is made publicly available so that anyone can study, modify, distribute, make and sell the design or hardware based on that design. The hardware's source, the design from which it is made, is available in the preferred format for making modifications to it. Ideally, open source hardware uses readily-available components and materials, standard processes, open infrastructure, unrestricted content, and open-source design tools to maximize the ability of individuals to make and use hardware. Open source hardware gives people the freedom to control their technology while sharing knowledge and encouraging commerce through the open exchange of designs.
What is the point of having an OSHW definition? So that you and I can have an expectation of communication. Said another way, when you create an open source hardware design, I have a level of expectation about what is included: Editable design (PCB, mechanical, etc) files, firmware, etc. What we (SparkFun) need most is a way to mark our boards (products/electronic devices) as meeting a level of openness. v0.3 of OSHW is pretty reasonable in our eyes.
Another piece of OSHW - the Xbee Explorer USB.
Open Hardware = some device that you can talk to because it has a well-documented interface.
Open Source Hardware = some device that you can replicate and develop freely because you have access to the source. I like and want Open Source Hardware.
One of the major points is with an OSHW project you are allowed to re-mix and re-sell (make money) on another person's design as long as you maintain attribution. All our OSHW products will have this stipulation. Non-commercial license are not fully open in my mind. Read more about the Case for Free Use: Reasons Not to Use a Creative Commons - NC license.
A handful of items from companies supporting the OSHW movement.
We realize we are far from done and this is just the definition (no SparkFun-friendly license has yet been written). I welcome comments and be sure to read up and discuss the latest on the Open Hardware Summit forum.
Am I missing something?
yes :) - you wouldn't make open-source hardware unless that's what you wanted. sparkfun, arduino, adafruit and a long long list of others want others to improve their hardware, publish open designs and work together to create and share hardware.
keep in mind OSHW does not mean you will not work with a company and have a business relationship. many makers who create OSHW get royalties from their kits sold at sparkfun, adafruit and others.
business and open-source hardware are different things - make sure you figure out what your goals are for each.
as diego said - "Creative Commons Attribution-NonCommercial-ShareAlike" might be a better fit for you.
please post up in the OSHW forums, your question / statement is one that comes up often, it would be helpful to others to see the replies from everyone, etc.
I support the movement on a OSHW license, but i have to ask. Where did you get that bag (And Goodies).
I would accept and use the 0.4 OSHW license, i but i feel that it's missing something, although not sure quite what it is.
this is not related to what i think the license is missing but, Placing limits on encryption might be a good thing to do in the license.
Just my $0.02
--MattTheGeek
As an electrnics newcomer, I support and believe that Open Source is the way to go. Without it learners and newcomers such as myself will suffer lack of transparency and encounter additional hurdles to surmount in an already difficult learning environment. OS offers transparency, adaptability and explanation while employing off the shelf components and technologies in new and inovative ways, If OS is not embraced and exploited, inovation may be stifled or reduced. Just my .02 from a newbies point of view.
Bob
I trust the OSHW definition developers will touch upon manufacturing rights. It might be sufficient just say "The OSHW def does NOT address manufacturing rights.." This will be more reconciled with the reality that those OSHW manufacturers such as SFE and Adafruit do indeed make fair arrangements with the designers.
SFE, Adafruit, and others, have established growing businesses where OSHW is just one small piece; Their businesses are about delivering little unfinished bits of electronics in chunks that are affordable and digestible for the educators, hobbyists, experimenters students, etc; There's a huge amount of work beyond just designing a one-off widget; The SFEs and Adafruits of the world add a great deal of value.
Nate: _
...
We do post the editable Eagle design files which in my opinion are upstream of the gerbers and more valuable.
...
In summary, by defining open source hardware we can point and say "here's what I expect"... _
I'm sure you're aware that Eagle is not free software - It's not open source, has only limited capabilities in the freeware edition, and has an undocumented file format.
This is why the OSHW definition includes the "Necessary Software" clause, which mandates that designs use software which is either open source or has a documented file format.
You are a signatory of Version 0.3. Does this mean that you plan to move to a documented/open source software in the future? I realize that the afterword allows you to use other forms of communication. However, a more meaningful endorsement of the OSHW initiative would be compliance with it in the designs you sell.
If you don't want to do this, I hope you're aware that you use the words 'Eagle' and 'open' together with disturbing frequency. Your statement concerning "here's what I expect" says that Eagle is good enough for you, and your statements carry a lot of weight. Can you explain your rationale?
@reemrevnivek - b00m! in less than 5 days of your comment there is an exciting statement from EAGLE:
**I'm sure you're aware that Eagle is not free software - It's not open source, has only limited capabilities in the freeware edition, and has an undocumented file format. **
CadSoft EAGLE development team developing XML format for schematic, layout and parts libraries!
http://www.element-14.com/community/thread/5589
"Does this mean that you plan to move to a documented/open source software in the future?"
We have no immediate plans to move to another PCB software, but every day I have a want for a better libre software: a want for accessible, teachable, usable PCB software. I grant you that Eagle is not the completely libre software that we all want, but I do argue that Eagle is slightly more accessible than say Altium/Protel (there is no gratis version of Altium that I am aware of). So today, at this moment, we offer the design files as 'open' or libre to use. Yes, it does require that you use a piece of software to edit those files (Eagle PCB is gratis to download, not libre for format interoperability), but I contend that by posting the files in an accessible, teachable, usable format we at least open the door for worthwhile sharing.
If/once Limor and Phillip can get Eagle to output a visible text format, then it will solve a lot of ills. I can't wait to be able to do real diffs! That piece is very crucial and will take us one more step towards much better libre software options.
@reemrevnivek - excellent question / comment, please consider posting these in the OSHW forums.
i have great news - very soon there will be an open file format. there is a group of us talking with eagle this week and we're also prepared to great the format on our own if needed.
keep in mind the OSHW def and statement says "ideally" we'll have open format and open tools - we're not there yet, but we will be soon and that's what matters, the goal.
OSHW license may and will require open tools and formats, please cut us a little slack we are working very hard on this - please consider joining, posting in the OSHW forums and helping out.
Ok, I spend months of my time designing and building a widget and provide all the details, the source code, the gerber files, etc. You guys take my hard work and make an improvement, say 4-40 screws instead of harder to find metric versions or vice versa, start cranking them out on your production line, where as I have just myself and my work bench at home, you pump them out at high volume, turn a profit and I simply get a thank you and my name silkscreened on it? pfftt... Am I missing something?
I think "open source hardware" is a bit overrated: the core functionality of open-source hardware is closed-source components, and open-source hardware is really nothing more than breakout boards for proprietary chips. The service that companies like Sparkfun and Adafruit provide--and I am thankful they do!--is making it easier for people to interface with proprietary chips, because they have the means to pick-and-place and solder reflow chips onto PCBs. Releasing the PCB designs doesn't hurt anything because hardly anybody will bother to use them on any commercial scale. It builds good-will without risk.
How do you make money as a small-fry? The software (firmware)! Nobody would bother at all with Arduino if it were a pain to load designs onto the chips with its easy-to-use bootloader and IDE. Arduino's software sells its hardware. So rather than spend months on a hardware design, look around for some ready-made solutions and concentrate on the brains of your project.
@Ryan, yes, you probably are missing something.
First off, you are posting at the Sparkfun site. I will make the assumption that you are either a customer or are looking possibly to be a customer.
You asked about how someone can make money on their design that they also give away the "technical readouts of their battlestation" (schematics, firmware, pcb art, etc). But the surprise is that this is EXACTLY what Sparkfun does. Look around, many of the products that Sparkfun developed are provided with all the details that would allow anyone else to recreate them for sale and totally cut Sparkfun out of the loop. They would eat Sparkfuns lunch.
Or maybe not. Sparkfun just hit something like ten million in annual sales and has hired something near 100 employees.
So giving the details away seems to bring people back for repeat business. I have purchased quite a bit of stuff from them and plan to continue. Why? Because I often need those tech docs freely provided to fix/modify/expand the toys.
Bottom line: if you create a cool widget, you are more likely in todays market to be successful with the open approach rather than the "try to hide" everything approach.
I think what you've missed is this:
Whether you decide to release the fruit of your labour as an open design is your own choice. If you want to be the one who makes the money, then probably you don't release your design as "open source hardware"... Or at the very least you'd want to somehow position yourself in the food chain, and just accept that opportunists will do their thing.
It's a personal choice. I wouldn't fault anyone for deciding differently for me... But personally, I feel that I owe a lot to the free projects I rely on for a myriad of things. I like the idea of giving something back. I like the idea that other people will have the opportunity to take what I made, and even change it to suit their needs if necessary. Of course, if I developed something that I thought was really worth a lot of money, I might feel differently about it. Who knows?
"You guys take my hard work and make an improvement"
My question is whether there is a requirement to then leave this improvement as open and under the same license (in the GPL style), or if this improvement can now fall under a different license by the improver (in the BSD style)?
As a developer of free software (and more recently hardware), my personal preference is for licenses that ensure that derivative works remain free, but I also contribute to (and derive benefit from) works under more permissive licenses. Where does OSHW stand on this? Instead of reviving a 20 year old flame-war, I think there should be two types of licenses differing in this important aspect.
Profit isn't the only motive for people to develop and share ideas. That's a big part of what makes open source software and hardware and many of the ideas behind web 2.0 (pardon the clich?) so viable. People work on open source software and write wikipedia articles and share arduino projects not for profit but to satisfy their need to create and contribute to something larger than themselves.
People are motivated by more than just money because money alone can only help you ascend the base levels of Maslow's hierarchy of needs.
"make an improvement, say 4-40 screws instead of harder to find metric version"
Actually, what you describe is a very common way that projects die. People design a thing, but use hard to find or retired parts. By opening up the design, you insure that the thing you create will live on past technological improvements.
Let's say you design something really cool, but over time they don't make that screw any more. If it's a really cool design or widget, SparkFun or any other company now has a reason to help improve it, find a good replacement part, and allow the widget to morph and improve over time.
Yes, you need to find your part of the revenue stream. There are many current options including royalties (Arduino), to becoming a kitter (MakerBot), nurturing community (DIYdrones), to becoming the best support company out there. My point is that we are quickly moving past intellectual property as a defensive measure, and moving towards adding value and making better products along the way.
If you think like that then you should probably choose a different license for your design =]. Maybe something like "Creative Commons Attribution-NonCommercial-ShareAlike". Nobody will think you're evil ;)
"This competition keeps us on our toes, keeps us sharp, and keeps us innovating as fast as we humanly can."
This point is exactlly why OH is the greatest thing ever.
Closed source keeps you from having to innovate to still turn out a product. If i freely give the competition the means to compete with me, yet am still the best, what does that say about me? It says I am one ridiculous engineer, thats what.
Nate, I wanted to talk to you at the OHS (I attended) but was to shy and had to catch the 7 train back to Penn then to Albany right when the cocktail hour started. I am a Mechanical Engineer so I totally love the idea of Open Sourcing stuff like generators, wind turbines, off road vehicles and the like. I can just think about the cool stuff a garage nut could build in if I open sourced a Corvette Z06 or something of that nature.
Just wanted to add...
The reality is big buisness will always be there. But if you can create an effective brand that is essentially the "Red Bull" (think extreme sports) of engineering then your marketing and viral support will blow the walls off of your competition. People will buy from you because you are the coolest and the best. Sure your product may cost a bit more, but when they see you flying to the moon in custom neon pink rocket ship shaped like an xwing fighter; people will think that is the sickest thing ever and want to get on board. Be the coolest engineer not the most profitable. The kids remember the iPhone, not DOS.
Ok? I didn?t exactly intend on causing such a stir with my comment. Share it so someone at home can make their own, sure, why not? But; giving it up to a company that is in the business of making money for itself and its employees while I get nothing in return for my idea and hard work, no forking way. I can see Ada and Sparkfun doing idea swaps, but they are in the business and they both end up making a profit. The rest of you can give your ideas away for that purpose if it gives you warm and fuzzies. Just not for me. Yeah, I have bought stuff.Premade products that are not Sparkfun red that they get in volume and simply sell as component parts, and I didn?t use them in any products for sell, simply my hobby stuff. And as for insulting someone?s intelligence for having a differing idea? I?m sure you are an absolute joy to be around both online and in person. It?s a great position to take when you?re trying to defend open sharing. Makes me want to share more of my thoughts?
Seems like OSHW should maybe use some of the concepts from OSS (open source software) like the dual license pioneered by MySQL. this is an OSS structure that allows anyone to get & use the source, until they embed and resell... then a separate agreement is required that includes additional license terms, royalties, etc. Could have both OSHW licenses; one completely open with attribution, one dual license structure. Am I missing something?
MarcL
http://www.aras.com
I have been a loyal SparkFun customer from the beginning. It has been a lot of fun to watch this business grow from a small group of college grads in the basement offering up cool break-out boards and source-code snippets, to a full-fledged 10M/Annual company supporting a large group of employees with lots of cool gadgets!
I would love to find a way to become a part of the OSW community and pay my bills. To me, it seems that the companies that are successful with this business model aren't really producing consumer-ready products, but supplying pieces of a complete product to electronic enthusiasts. Obviously, this can be a great business model if that is your target audience. (I know I have purchased my fair share from SF).
I think the thing that would persuade me more than anything to actually start releasing "Open-Hardware", would be if I saw a bit more than "required name recognition" in the license. I really have no idea how this would work, but that is the problem. How does someone pay the bills when they give away all the designs for a product that has taken 10k, 200k to develop? Could an open license be modified to require hardware re-producers to pay some form of royalty to the originator? Or maybe some form of "moderation" system could be developed to "vote" on how the distribution of royalties are dispersed among various contributers? I know I'm way out there right now, but just some thoughts.
Maybe a company like SparkFun could form a project submission / acceptance system where projects that were accepted for production / manufacturing / marketing through SF would pay something back to the various contributers based on a moderation system?
I don't think anybody has ever made much money from shareware software, and what you're describing is along the same lines.
In my first lot of boards, I've managed to mostly pay for having the lot made and kept a few boards for my own use. Selling my boards lets me make a large lot to keep the costs of my projects reasonable.
If I can justify making a trip to the post office to ship an order and maybe pay for lunch on the way, I figure I'm accomplishing something.
Jon
oooh a grab bag!! can i get one like that?? that looks sick!!
Was a reply to Nate's comment:
"OMG I completely agree with you. We're beginning to post our Eagle Lib and our Eagle designs to our GitHub page, but it's less than ideal. I really do wish there was a better way to collaborate on a design/project. We don't want to internally develop it, but would be happy to sponsor a public project. Perhaps an open/visible PCB design format will be enough to get the community going, but I think we have a lot more work to do."
Lets go from board hosting, to project hosting. Imagine a site where:
Source(Boards, Software, parts lists)
Documentation(Wiki style, mostly.)
Collaboration(Forums, Design Whiteboards, online board/schematic creator/editor)
Version Control
exist in one place, with thousands of more projects, for free.
thats my vision. Almost like a combination of github and instructables.
--MattTheGeek(Matthew)
Now would be an excellent opportunity to look at the mechanisms by which we share the board designs. Right now there is no good (dedicated) solution.
SparkFun and DIYDrones Store have some products with links to the board files in the bottom, but they aren't in any kind of source control, and any license notes are put in unseen layers in the board design.
I (personally) use GitHub for any open source board designs I have (i really should go open source more of them when i have time). BUT! GitHub is less than ideal. It'd be great if we had a site (like) GitHub with version control with something like SVN/HG/GIT/choose-your-cvs-poison that was open-source hardware board-design aware.
Integrate things like gerbv to generate images of gerberse, allow sharing (cloning/forking to use git terms) of CAM files, and let some CAMs be run server-side.
GitHub provides wonderful easy version management, and forking/cloning and pull requests (social coding at its best!)
BatchPCB allows us to upload designs and view images (on a webpage) of the layers and the whole.
Can we create something in between?
~Alex
How great it would be to have a github tuned for hardware --with circuit layout images and version control and instant pricing to order boards, sets of components, printed enclosures, etc from an array of vendors in a range of countries. From the builder's side, this thing should communicate in a psychic manner with layout software so we are not forever zipping gerbers and maintaining uploads to yet another site.
BatchPCB is aligned in this general direction, and it's great to see the comments here showing other people recognize this need. What will it take to do in hardware what the software people have done for years: rapidly download, tweak, and compile one instance of a project from a creator who may have only five customers?
Lets go from board hosting, to project hosting. Imagine a site where:
Source(Boards, Software, parts lists)
Documentation(Wiki style, mostly.)
Collaboration(Forums, Design Whiteboards, online board/schematic creator/editor)
Version Control
exist in one place, with thousands of more projects, for free.
that's my vision. Almost like a combination of github and instructables.
--MattTheGeek(Matthew)
OMG I completely agree with you. We're beginning to post our Eagle Lib and our Eagle designs to our GitHub page, but it's less than ideal. I really do wish there was a better way to collaborate on a design/project. We don't want to internally develop it, but would be happy to sponsor a public project. Perhaps an open/visible PCB design format will be enough to get the community going, but I think we have a lot more work to do.
I realize it's not specific to hardware designs, but SourceForge seems to offer all the options that people want. Anyone can create a project, it has great visibility within the community, there's documentation-ability (a new word I just made up), and all the other things everyone seems to want.
OTOH, I'd be willing to do the leg work to set up some kind of portal to list the projects, but I'm not sure how big the demand is, or if there's even enough people willing to contribute to such a project. With the cost of server space and datacenter hosting, is there enough community support to fund such a project?
SourceForge strikes me as just short of a totally dead letter at this point in history. It continues to limp along on the inertia of older projects, but there are probably a dozen better options right now for starting anything from scratch.
(I don't mean to sound hostile in any way here. It's just that I spend a lot of time sorting through various open source projects, and every time I come across something on SF, my first thought isn't "oh, this might be interesting", it's "awwww man is this even worth the hassle?" Contrast with the experience of using GitHub, or even Google Code, and it's pretty telling.)
I suggested SF just because I think more people have experience with it. Google code or GitHub are equally suited for the purpose. (That is-- good, but not quite a perfect fit.) We need somewhere to do it. As I said earlier, I'd personally be more than willing to donate my time, experience and money towards a better solution, but I'd need to make sure that there's enough of a community backing it to fill in the blanks, and to help carry the load.
(I don't disagree that SF has gone down hill as of late. They really need a way to prune all those dead projects. Yes, there really are some very actively maintained projects there. There are just far more abandoned projects than live ones.)
Fair enough.
For whatever it's worth, I think after reading other comments here (and after spending 3 years at SparkFun) that a lot of the technical and social issues really revolve around formats and versioning/collaboration tools. A project hosting service with enough momentum for widespread community use could be great (whether that's an existing site or something new). We've talked about this problem a lot here, and it's something I'd personally like to put resources towards.
That said, I'm a software guy, and the free/open software community has had pretty decent tools for a while now. I'm convinced this matters a lot. Hardware folks at the SparkFun end of the scale are basically living in the stone age. That's not surprising - source code is plain text, and plain text is easy to edit & diff meaningfully. Complex binaries, not so much.
Set up an email address if anyone's interested. If there's enough demand, I'll start working on it.
Send Ideas/comments to:
oshw at thebecwar.com
On my TAP-28 application board for 28-pin PIC18F-series micros, I've used a Creative Commons by-sa-nc license, with a slight difference. An immediate waiver is granted to use the board commercially as a component of a larger system. What this does is allow somebody to build a widget with this board as the heart and sell it, while not allowing people to sell the bare boards without permission.
This seems like a reasonable approach. Somebody can use the board as the basis to make a fortune but I won't feel screwed by somebody using my intellectual property without adding anything to make money selling bare boards.
Jon
Here's a link to the license: http://www.clever4hire.com/throwawaypic/tap-28-license
TAP: Very cool little board. That's a decent way for you to sell your product/design. But I worry that by attaching the -NC requirement, you will hinder outside collaboration.
Let's do a -NC vs non-NC example: Let's say I (not SparkFun "I", but Nathan "I") see a way to make your board better. For example, I buy 3 of your boards, put them together, and notice that the headers keep falling out when I how to solder them in. To make your board better, I re-design the header footprint to make the holes slightly offset, also known as 'locking foorprints', so that the headers stays in place when the PCB is flipped over.
For the readers, you can see this technique comes from the SparkFun Eagle Library and has already been implemented on the TAP-28 board: http://tiny.cc/TAP-28
Back to the example: As a person who made the board better, I may want to make money for the change/improvement I've made. If the design has -NC attached to the license, I may never make the improvement because I know I cannot legally resell the improvement. Now if you publish the design without -NC, I can then release the new 'locking footprint' version of TAP-28 under the same CC-SA license giving you attribution for the original design, then begin to sell my version. A week later, you can see the improvements I made, incorporate the improvements back into your original TAP-28 design, and sell the new and improved TAP-28 board using the work I did. We build on each others work, creating a better product.
In the -NC example, improvements are made when you are able to develop them. In the non-NC example, improvements are made any time someone feels the improvement is potentially fiscally viable.
Btw JonC - I couldn't find your original Eagle files, but I live and breathe PCB layout. I'd love to give you layout feedback/improvements if you want it. 1st would be to increase the annular rings on your PTHs - it will make soldering for noobs a lot easier, and decrease the possibility for the PCB fab house to mess up a via.
How about you just give him your idea for the offset footprint and let him make a profit? Thats what you are asking of him to start with. He can make three boards at a cost to himself as he had to deisgn the thing, get the boards made and hope that someone, you, would buy them to start with. Then you take his idea, make a minor tweek, set it forth on your production line with 100 employees and then send it back to him to make another change for you to make even more profit from when he can even begin to match your production volume... and he has signed away all rights to his idea. You are getting all the rewards of having an outside consultant for whom you don't even have to pay! muliply that times an entire comunity of hobbiests? sounds great for Sparkfun!
Thanks for the comments Nate. I haven't posted the Eagle files yet since I had a couple errors in the first layout that I want to correct. The revision is being made now, and once I get the boards back and tested, I'll post the files. I figured it was best not to get a copy of a file that has errors out to the public.
As far as the -NC part of the license, there's a blanket waiver for commercial use of the board as part of a system. What I wanted to prevent was somebody taking the design and selling it "as is" without any changes. I guess I've feel ripped off if I saw somebody selling my work, possibly at a much greater price without adding any value to the design.
My goal was to support the community with a unique design, at a cost lower than an individual might make a few copies.
(continued)
(continued)
I'd certainly be willing to entertain a waiver for somebody to make and sell the boards, particularly if improvements are made to the design.
The redesign can be seen here: http://digital-diy.com/forum/viewtopic.php?f=5&t=652. A few refinements, including the ability to use right angle LEDs and switches if desired so that the board could be used behind a panel. I've also changed to a mini-USB connector. I debated about this as it's a surface mount connector, but it's located by pins and a pretty large pattern, so I don't think people will have much problem soldering it.
By the way, I obviously like the lock-in idea for the headers, but I refined the patterns somewhat. The SF layout has the part centered under pin 1 instead of the centroid, so rotating it really makes a mess. The other change is in the pin numbering of the keyed connectors. I don't know if there is a "right way" but the numbering was reversed compared to every reference I have.
Jon
Sparkfun, any chance you could to a teardown of all the different licenses in laymans terms. I am getting more and more confused with all the different licenses and never know which is best for myself or my project! Is there anyone else here who would like a simple table-like breakdown of the features and benefits of each type of "open" license? I try to read them, then get all confused!
Maybe I am just too simple in the head :S
Can't wrap your head around all the options? Me too! I wish there was a better 'tutorial', but I don't have the technical legal expertise to really write one. Also, I don't believe there's enough case law yet to water test many of the licenses currently available.
For lack of anything better, I currently use and promote the CC-SA v3.0 license:
http://creativecommons.org/licenses/by-sa/3.0/
Why do I choose this license? Creative Commons makes it REALLY easy to understand. The SA 3.0 allows for commercial use, requiring only attribution. Currently, this is the best option, but it's not as definitive as I'd like to see for hardware. I want open source hardware to include the requirement that design files be released in an editable form (not gerbers, editable like a CSV file). That's why we're working on the OSHW definition and then some day a license that we can release our designs under. Until that day, CC-SA is good enough for me.
There is some good information here: http://creativecommons.org/about/licenses/
But if you are looking for breakdowns between public domain, CCs, and OSHW, i don't think anything exists just yet. There has been a lot of cross-pollination from OSS/FOSS/FLOSS into OSHW and the Open Source definition (in reference to software) was pointed to often during the early drafts of OSHW.
The OSHW draft is in plain english though, unlike the GPL/LGPL. I urge you to check it out so you can voice your opinion of it, before the thing becomes set in stone.
This has been interesting to read through. I didn't understand open sourcing myself until recently. I helped with the CHIPINO.cc project and it made me see the difference. If someone adds a screw its an improvement. But then someone finds a blue screw and then someone else a red screw. The someone finds a way to put an led in the screw. Before long the original design has been improved to something far beyond what the originator could have created because they didn't have the time equal to what 1000's of contributors could do. Look how many great shields and boxes and projects and accessories you can find for the Arduino.
Creative common licensing just tries to keep some kind of guidelines on it. In reality it doesn't need it. People will improve what they like and make it better. And many will make a few bucks in the process. Good for them.
It's the $200 million dollar bonus CEO's business model types that don't like the idea because they can't control it.
One point missing in the advantages of open source hardware (well, open hardware is already good for that aspect) is when viewed from the end users side of things.
I know quite a lot of people who would rather buy some hardware that is sure to work flawlessly with open source OSes ; I have the feeling that it very well may be the same in the near future for OSHW itself: those things would be easier to upgrade, repair, interoperate, ... and these are huge advantages IMO.
Two other things I really liked in Lady Ada talk (a must see) on the subject :
- OSHW helps building a community
- OSHW helps people making better designs and learn
Well, this comment doesn't really help on the topic of the current draft, but I guess it is important to think about how and why one would even get into OSHW.
Anyway, I'm really eager to see what OSHW will evolve to, and I want to be a part of it :)
However, I think some of the claimed benefits of OSHW are exaggerated (e.g. OldFar-SeeingArt's "if you create a cool widget, you are more likely in todays market to be successful with the open approach rather than the 'try to hide' everything approach").
OSHW certainly isn't for everyone - some products "may" do better with an open approach, but doing OSHW doesn't mean your product will be a success, just like doing closed hardware doesn't mean it will be a success either.
DIY drones is a good example, they are open and their competitors and closed, i would say a case could be made that DIY drones is doing better and will ultimately move more units than their closed competitors. DIY drones has a loyal community, people are part of the products - they support them and support DIY drones.
again, please post these great thoughts and comments in the OSHW forums and mailing lists, thanks :)
Nate: I would be interested to hear more of your thoughts on just how open hardware can be while still maintaining a viable business model. I've always appreciated SparkFun's level of openness--e.g., offering schematics for most parts. However, my intuition tells me there's a big difference between that and offering complete Gerber files, and perhaps that's why SparkFun has not yet taken this step.
This article does a great job of discussing the OSHW side of things. As someone who's trying to build a business around a new product, I would be very interested to hear a more balanced discussion of SparkFun's business model.
(Posted on OHS as well: http://www.openhardwaresummit.org/forum/viewtopic.php?f=5&t=78)
I am happy to be as transparent as possible about the SparkFun business model. I want to point out that we don't publish gerbers as they are not editable and not really that valuable. We do post the editable Eagle design files which in my opinion are upstream of the gerbers and more valuable.
What we design here, what we build here, is nothing magical. Anyone with a few months of free time can learn how to read a datasheet, layout a PCB, and solder it together. If they felt so inclined, they could easily setup a webshop and sell extremely similar products to SparkFun.
Take for example an accelerometer breakout. Many companies now offer these breakout type products. Whether they based their product on our eagle files or not, doesn't matter. The design is straight forward enough they probably did designed it on their own. So in this example, OSHW doesn't make a difference to the commercial competitors. We open the breakout board designs because we figure people can learn from it, modify the design for their specific application, etc.
Now take for example the Arduino Fio. Slightly more complex, more time consuming to replicate. The design files are CC-SA licensed, and competitors can come in and replicate all they want. We could potentially lose sales to a competitor that can build the board for cheaper, be out of stock less, or provide more support. Is this bad for the end consumer? Nope. Bad for SparkFun? I would argue emphatically and definitively no. If someone can do a better job of soldering caps to a board, then they win the sale. This competition keeps us on our toes, keeps us sharp, and keeps us innovating as fast as we humanly can. By sharing the design files via OSHW SparkFun is stronger and more agile.
Why doesn't SparkFun sell more types of components like capacitors and resistors? We would then be competing with Digikey. Can we "out Digikey" Digikey? No. And I would simply exhaust resources trying. If someone can "out SparkFun" us, then we must change. Presently, I believe we are doing "SparkFun" better than anyone else can.
In summary, by defining open source hardware we can point and say "here's what I expect". Defining OSHW is important to me so that we can share ideas more easily. Only good things can come from more knowledge, more competition, and more innovation.
-Nathan
Thanks for the well thought out response.
You're absolutely right that Eagle files are better than Gerbers. When I had a quick look through SparkFun's site to check for source files, I looked only only at a couple of products I'm familiar with--the WiTilt and KinetaMap--which don't seem to offer Eagle files. Looking at a few more products, like the 6 DOF IMU and innumerable breakout boards, I am genuinely impressed to see that they do include Eagle files.
I believe I am offering my own product for a price at which it would simply not be worth someone building their own. In particular, the heart of the device--a u-blox NEO-6Q GPS receiver--is prohibitively expensive in quantities less than 250. This makes me feel like, perhaps, I could open not just the firmware and schematics, but perhaps even the Eagle files.
My main concern, in this case, is that a larger, better established company--one who might be able to invest more and purchase parts in larger quantities--would have everything they need to produce the exact same product at a lower price. Given that it's taken me years of prototyping to develop it, I can't quite see my way to opening things up completely.
I have a lot of respect for what you've done with SparkFun, and have modeled many of my business practices after what I see. My question to you is this: Do you think there is an argument to be made for products with (partially) closed source? If not, as a small business owner, how would you protect your intellectual and financial investment in a design, while still pursuing OSHW ideals?
@crwper - you wrote:
**how would you protect your intellectual and financial investment in a design, while still pursuing OSHW ideals? **
OSHW is not a business model - it's a cause and something people do for reasons besides profit.
business and open source hardware are two different things, you can be good at one at bad at another, or bad a both, or good at both!
how do you protect your intellectual and financial investment in a design usually? copyright, patents and trademarks.
i think many people want some type of promise that doing OSHW will guarantee success (in fact, some has asked for exactly that) - that will never happen, just like there are no guarantees for any business.
doing OSHW doesn't mean you're going to be the next sparkfun - but one thing i know for sure, doing OSHW will definitely make your part of their community.
we do open source hardware to become better engineers, to become better people, to be part of a community, to make the world a better place through engineering, we do open source hardware because it's fun, some of us even do open source hardware because "business" can be a rewarding side effect from all of these things too.
Multiple people have expressed concerns that succumbing to some of the OSHW hype is not a viable business strategy, and you are offering this ?success is not guaranteed? banality. By posting such a straw man argument (implying crwper is asking for a guarantee of success) you are strengthening what I suspect is many developers? gut suspicion that ?[being] ? good at both!? is actually extremely unlikely for professional product designers.
On a slightly related note, I think OSHW supporters do a disservice to at least some of their stated ideals if they turn this into too much of an ?us vs. them? argument. It?s good for someone to create something valuable and give it to the rest of us, but let?s not forget how great it is for someone to create (and allow us to buy) something so valuable we want to pay for it.
- Jan
@jan - exactly, OSHW is not a "business strategy"? there are dozens of companies and thousands of individuals doing great right in and around the OSHW ecosystem it certainly possible a "business strategy" using OSHW works, at least for them. that doesn't mean it will work for everyone, and for professional product designers - they'll need to see if it fits 1) their personal goals and 2) their business goals. OSHW isn't for everyone, it might not be for the professional product designer - but here's one quick example that did work. i introduced mitch altman, tv-b-gone inventor, to limor fried. about 5 years ago - he never heard of OSHW (no one had) i convinced him to do a kit with adafruit that was OSHW - now mitch sells kits, more kits, OSHW, founded a hacker space, travels the world teaching workshops. mitch is a professional product developer, OSHW worked for him as a life choice and a business. please post in the OSHW forums for more examples :)
i'm pretty sure sparkfun offers up eagle files for many of their products, that's a good thing - they're editable. gerbers aren't. one day there will be an open file format so (besides eagle) people will be able to publish their files in an open exchange format to encourage editing and improvements. i'd rather have eagle files than gerbers, but keep in mind file formats are not open yet (working on that now).
please post these great thoughts and comments in the OSHW forums and mailing lists, thanks :)
You're absolutely right--Eagle files are certainly better than Gebers. The few products I checked didn't seem to have Eagle files available, but looking at a few more, it seems that many of the products SparkFun offers--even ones with complex designs--do.
I'll be sure to check out the forum.
I was recently talking about OSHW to someone relatively active in the electronics community. They brought up a point I thought was a bit cynical, but maybe should be discussed early on. Some of the more vocal proponents of OSHW run businesses where a large proportion of the products sold are actually open source designs from third parties. The point being made was that if more people start releasing OSHW with explicit commercial permissions, these businesses will have more products to build and sell. That doesn't make much sense to me because all the businesses mentioned also have a large amount of original designs that they in turn open source, but it's a point that might be raised again later. What can be done to better expose the altruistic motives of the OSHW creators, and remove the appearance of a conflict of interest?
I think your points are good but you made one mistake. Don't argue with PT, it is useless. He will refuse to see your points because you haven't cited specific sources. If you cite a source he will say "Interesting, you should bring this discussion to this other board."
What you wrote is very easy to understand. You're saying that members writing this draft benefit from open designs because they profit from them. This is obvious. I like your approach of asking if there is a way OSHW could shed that appearance without implying anything sinister on the people writing the draft.
The only thing i would add is that once this draft is completed, nobody can say their device is OSHW unless it meets the definition. Which means if you made something and want to share it via OSHW but don't want anyone to sell it.. you cannot use the term OSHW (well, you can, but people will hound you for it). This topic is on the OSHW forums. The general response from people writing the draft is "Not everyone can do OSHW, we understand it is difficult". This sounds like an answer from half of the crowd, not a consensus/compromise.
@Maha - i don't think it's fair to pick on me for asking for 1) a specific example 2)asking his friend with their opinion to post and 3) to do what nate asked and post in the OSHW forum.
i do not think anyone who has worked on the OSHW draft(s) has anything "sinister" planned.
as you said they could call it OSHW, but it might not be OSHW - just like you can attempt to call windows "open-source software" but it's not and there are definitions provided by the OSI and the community that help define what it is.
why would someone want to call something OSHW and not want it to actually be OSHW? at the summit, the 150+ people who signed the draft and for the people doing OSHW non-commercial for years they collectively have not wanted non-commercial in OSHW, just like OSS does not want non-commercial in OSS.
you're right, many people besides me will say OSHW doesn't fit everyone's needs - if they want to make something and not allow anyone to sell it, OSHW isn't for them but there are plenty of other licenses to choose from.
for OSHW there isn't going to be a compromise for NC, it's not wanted or needed - just like it's not wanted or needed in OSS.
i'd encourage you to post in the OSHW forums and mailing lists so i am not the only one saying this and you can dislike other people too :)
@macetech - the OSHW community is trying to get the comments on this (as nate said) in the forums and the mailing lists. can you post these up there? great stuff, i am sure many people want to talk about this.
here are some quick comments:
Some of the more vocal proponents of OSHW run businesses where a large proportion of the products sold are actually open source designs from third parties.
who? please be specific and for which products.
The point being made was that if more people start releasing OSHW with explicit commercial permissions, these businesses will have more products to build and sell
for many, that's the goal of releasing their designs as OSHW.
That doesn't make much sense to me because all the businesses mentioned also have a large amount of original designs that they in turn open source.
need specifics, which businesses, which designs, etc?
What can be done to better expose the altruistic motives of the OSHW creators, and remove the appearance of a conflict of interest?
what do you mean by appearance of conflict of interests?
Phillip, as I mentioned this isn't my own opinion, it was one that was expressed to me recently by another kit designer and manufacturer. I really can't list dozens of companies and the hundreds of third party open source products being manufactured, but you know that adafruit and SparkFun are examples. That person's worry about conflict of interest is that OSHW appears to be pushed by people who may have something to gain from wider acceptance, i. e. more products to sell that they didn't have to design.
Of course my own opinion is not in line with the above, based on the fact that the same businesses are also releasing their own designs. I just thought it was an interesting concern that may come up again, and I didn't really have an answer at the time.
@macetech - can you have this person post their opinion in the OSHW forums and/or mailing list? that seems like a fair request right?
That person's worry about conflict of interest is that OSHW appears to be pushed by people who may have something to gain from wider acceptance, i. e. more products to sell that they didn't have to design.
i think it's unfair to sparkfun and the OSHW community to relay this opinion for someone else, more so since you say it's not your own - we can't ask questions back, we don't know who they are talking about or what specific examples they can cite. it's impossible to know what a conflict of interest means in this context.
**Of course my own opinion is not in line with the above, based on the fact that the same businesses are also releasing their own designs. I just thought it was an interesting concern that may come up again, and I didn't really have an answer at the time. **
that's good to hear, what can we do to convince your friend to post up in the OSHW forums or mailing list, if they're super-shy they can email me directly or even nathan, i'm speaking for him but i bet he'd be more than willing to help get this person involved in this important discussion.
It'll be up to the "mystery person" to get involved with OSHW if they choose...that seems unlikely due to the current opinion. I think the "conflict of interest" concept is expressed in plain enough terms. I also think it's more than fair to relay this opinion, since it's unlikely the opinion would be articulated by people within the OSHW community. I believe progress is made by overcoming obstacles, not ignoring them. I actively hunt for obstacles. Sometimes if a person or group is unaware of an obstacle I will bring it to their attention.
It seems way better to deal with this "conflict of interest" opinion now, at definition 0.4, then to have it sprung on you at a more critical time. Imagine adafruit is on national TV talking about OSHW someday. The anchor points out that you're one of the major proponents of OSHW and you also make a lot of your income from third party designs. Instead of having to come up with a vague answer on the spot, you have already seen this and will likely have a thought-out response.
Yes this would be great on the OSHW forums but discussion happens everywhere. The more places it's discussed, the better recognition it will gain.
@macetech - please invite the "mystery person" to get involved with OSHW.
i still do not understand the "conflict of interest" issue, but thank you for the example, i'll gladly answer that.
Imagine adafruit is on national TV talking about OSHW someday. The anchor points out that you're one of the major proponents of OSHW and you also make a lot of your income from third party designs.
but we do we make a lot of income from our own designs. but let's say we didn't - i think i would say "we work closely with these third party designers in the way they wish to be worked with, from compensation to credit".
that's what we do now, and that's what we'll always do.
please post this up in the OSHW forums and mailing lists, thanks.
I have not spoken to macetech about this, but I'm pretty skeptical about some OSHW claims, and high up on the list are claims like "altruistic motives" or "large amount of original designs". As with open-source software, a lot of the corporate support for anything free will come from those who sell the complementary products or services; if the majority of a company's revenue comes from manufacturing and distribution, it will want the cost of product development to be as low as possible. I think it's great for individuals who are spending time on personal projects anyway to let others benefit from their work, and I think it's great for companies like Sparkfun to help coordinate some of that or deal with the logistics and distribution issues that a hobbyist would not want to bother with. However, I think some of the claimed benefits of OSHW are exaggerated (e.g. OldFar-SeeingArt's "if you create a cool widget, you are more likely in todays market to be successful with the open approach rather than the 'try to hide' everything approach").
- Jan