1970 was a year that marked a significant event for rock and roll music lovers around world - Jimi Hendrix, one of the greatest guitarists the world has ever seen, died. While many technically accomplished guitarists have since come and gone, Jimi redefined what the guitar could be. A couple of years before his death, Jimi recorded the classic song “Voodoo Child (Slight Return).” The intro has a classic Hendrix riff that used a wah-wah (or waa-waa) pedal.
I would be remiss not to mention one of the best covers of this song performed by another guitar legend who died before his time - Stevie Ray Vaughn. His Austin City Limits performance is legendary:
You can hear the wah-wah effect used by SRV as well. I am of the opinion that Hendrix was born to create songs that Stevie Ray could make better. What does this have to do with open hardware designs? Stay with me and I will link it up.
For almost a half a century, Popular Electronics was the magazine for electronics enthusiasts. It had a plethora of articles, how-to guides, schematics, and building instructions for many cool projects.
Psst… I’ve got news for you software guys. We hardware guys have been doing “open source” for a long time. We know how it’s done!
The “American Radio History” site has PDF versions of all the issues of Popular Electronics available for download.
The January 1970 edition featured on its cover a do it yourself wah-wah pedal.
The article in Popular Electronics has everything you need to know about the wah-wah pedal. For electronics hobbyists, front and center is a schematic with a parts list.
The article includes an awesome description on the theory of operation of the circuity. The functions of the different subcircuits are clearly explained.
The explanation of the key circuit elements is what made Popular Electronics magazine so… well... popular. It provided more than just a dump of the schematic — the authors actually took the time to explain what the circuit was all about. And another cool part… PCB artwork was included! A 1:1 foil pattern for making your own PCB, as well as a picture of the stuffed board, was provided for reference.
“But a raw circuit board is not a complete project!” you say. Well, a design for the housing is also given:
This is a fine specimen of “open hardware.” Take note as to what was included:
- A well-drawn schematic
- A parts list
- A description of the theory of operation
- Artwork to recreate a circuit board
- An enclosure design
Let us return to the year 2015. There is much discussion on the nature of what open hardware is. Many try to use the GPL or BSD definition of “open.” I feel that this is a mistake. Hardware is not software. Hardware, by its very nature, is not free (as in beer) to reproduce. Hardware is physical. Applying a software philosophy to a hardware design is the wrong approach. Many people incorrectly try to put onerous requirements on the open design process, including the tool(s) one uses to create the design. They feel that the tools used to make the hardware should be 100% open and free as well. This seems ridiculous and is not pragmatic. Pragmatism should be a core value of all good engineers. Also, in almost every case, one does not recreate open hardware verbatim. It is often used as a guide, with which you hack your own version, using the tools you have.
Let’s take the software purist viewpoint to its logical endpoint. If a design tool source should be 100% GPL, what about the hardware running it? Do I also have to have the PCB layout of the computing system 100% GPL? What about the CPU design? Should the IC layout tools be 100% GPL? The mask set? What about the process for extrinsic silicon manufacturing?
See how applying the rules of software developers to hardware quickly becomes unmanageable? There certainly is much to be said in favor of open source software, but is a completely different beast altogether. Bits on a storage medium are completely different than physical objects. Hardware folk have to be careful of applying the philosophy of one domain to the other.
Our forefathers in the hacking movement had it figured out. They were not concerned with the nature of open source legal contracts and the implications of how to share what, where, when, and how much. They provided everything needed to recreate the design. You could build and tweak and have fun with the hardware. They did not insist that you use the same pen and paper they used to draw the schematic, nor that the design for the pen they used be fully open and documented. They did not insist that you used the same drafting table, soldering iron, or wire cutters.
Engineering is about using tools build stuff! Making is not about the tool used. If you spend too much time getting caught up in philosophical arguments, you are wasting valuable time in which you could bebuilding cool stuff.
All that being said, I posit that an open design is one akin to the Popular Electronics model:
With that information, you have everything you need to create, modify, tweak, and hack the design. Certainly having the raw design files and toolset is useful, but not necessary to get work done. In most cases, users are not trying to use those files directly. It is more important to see the design so that you can hack it or derive new work using your tool of choice. While it may give a software guy a warm, fuzzy feeling to have source for the tool available, it is not requisite to get work done. A hardware guy is going to take that design and build it his way.
Someone may argue that the tool used to create the schematic and artwork needs to be open so that it can be rebuilt 40-years down the road. Would you honestly use the same tools that the author did 45-years ago when the wah-wah pedal article was written? Outside of a quaint experiment in antiquities, absolutely not! Everything from the soldering iron, drafting table, and pen used simply does not matter. The author provided all that I need to recreate this design using the tools I have now! That is the essence of open hardware design. As long as I have the vital information about the design, I can problem solve to make it work with the tools I have.
It should be noted that the overwhelming amount of “open” designs to date have been done using EAGLE. This has nothing to do with anything other than it is free (as in beer, not freedom) - EAGLE is a proprietary tool!
So, the next time you are caught up in a philosophical debate over open hardware (it happens to me all the time…), think about Popular Electronics. Have you provided a theory of operation document for the user? Or have you just created a file dump with open tools? Are you trying to check box something to appease a GPL philosophy, or are you really trying to share the design?
There are currently efforts on defining what Open Hardware is and what can use the “gear” logo: http://www.oshwa.org/2015/09/19/open-source-hardware-certification-version-1/
While I appreciate the effort, the reality is that most folks of an engineering or maker ilk are not going to abide by this to avoid a fine. The maker will simply release what he will release. Hardware is a completely different animal. My observations have been that much of the discussion is coming from software folks. Maybe it is time to tap some old grey beards who have been doing hardware since well before the current crowd was even born. But then again, they have already done it! This is old news! Just readPopular Electronics. They did not care about the religious aspect, they focused on making cool hardwareand describing how it worked. It seems like that aspect - the primary objective! - has fallen to the wayside.
For myself, It doesn’t matter whether EAGLE, CircuitMaker, or KiCad exists in five or ten years (though I hope they all do). I will be making and sharing with what I have available under the assumption that future tinkerers of the world will be willing to work with the tools they have available. The design isnot the tools. The design is an abstract concept. That is what I really want to share, and what I want to access.
That is the essence of open hardware.
Originally published at CircuitMaker.