[Discuss] OSHW Best Practices / Layers of Openness

Thu Mar 7 16:26:33 UTC 2013

Well put David. It was precisely this issue that started this whole
conversation in the first place. The OSHW logo and label ("open source
hardware") are, as many have stated, a brand which is only valuable and
useable as long as it means something and points to a well defined set of
standards and rights - in this case, the OSHW Definition.

So how can we address products that have both open and proprietary
components? When I say proprietary I'm referring to designs the creator
controls, not something they purchased from another producer and that
everyone else can also acquire. Placing the logo on each part of a product
presents serious difficulties in terms of manufacturing. And I'm not sure
where a textual description would go - is it okay to bury it at the bottom
of a minor page on a website while calling the product "open source" on the
home page? I'd say no.

This is why I was so attracted to Tom's idea of a label that, no matter
where it's placed on the product, tells you right away what parts are open.

On Thu, Feb 28, 2013 at 11:09 AM, David A. Mellis <dam at mellis.org> wrote:

> These are interesting examples, Tom, but again I think they mostly point
> out the need to be specific about what part of a product you're calling
> open-source. For example, here I think it's clear that the PCBs for the
> SparkFun and Arduino products (I couldn't find files for the Adafruit one)
> are themselves open-source even if the components they use are not. I think
> we're much better off being specific about what's open and what's not
> rather than trying to set standards for, say, the relative complexity of
> the board vs. the components required for the overall product to be
> considered open-source -- especially given that basically every product
> uses proprietary components (whether ICs or radio modules or just screws).
> The latter approach seems like a potentially endless conversation. Again,
> look at Linux distributions, where people are still arguing about what
> level of proprietary software and binary blobs are appropriate to include.
>
> Also, my point about using the logo on an enclosure vs. on the PCB inside
> it wasn't meant as a comment on the relative importance of those two parts
> but of the semantic interpretation of placing the logo in those places. If
> I opened up a product and saw the OSHW logo on some part inside it, I
> wouldn't interpret it to apply to things around it. But if I saw the logo
> on the outside of a product, it's not clear whether or not it's intended to
> apply to the insides as well, making its use there confusing if the
> outsides are open-source but the insides aren't.
>
> As an example in the other direction (of electrical vs. mechanical) would
> be someone that uses a standard servo or DC motor and builds a complex
> mechanical assemblage around it. If they open-sourced, say, the CAD files
> they used to design the laser-cut parts that make up the assembly, I'd
> consider it reasonable for them to use the open-source hardware logo on
> their packaging or website, even if the motor were proprietary (since it
> would be just one component of the larger, open-source design). But again,
> these situations can be confusing and its important to be explicit about
> what's open-source and what's not.
>
>
> On Thu, Feb 28, 2013 at 7:16 AM, Tom Igoe <tom.igoe at gmail.com> wrote:
>
>>
>> On Feb 27, 2013, at 12:41 PM, Michael Shiloh wrote:
>>
>> 1) The overall guideline might be "can someone reproduce this project to
>> a reasonable degree (e.g. functionally the same, if perhaps the case is not
>> identical) with the information provided?
>>
>>
>>
>> So, let's pick a few specific examples, all of which think highly of, and
>> use myself (admitted bias on the third). But I struggle with defining them
>> as entirely open:
>>
>> https://www.sparkfun.com/products/11378
>>
>> The major piece of hardware on this board is a proprietary module from
>> Roving Networks.  Though SparkFun's support schematic is clearly open, the
>> module that makes this functional is not, nor is it reprogrammable. The API
>> for it is open, though. Is this OSHW?  What's the replacement part that
>> could drop into this board and make it work, with minor modifications?
>>
>> http://adafruit.com/products/746
>>
>> Similarly, the major piece of hardware (the GPS radio) is proprietary,
>> even though Adafruit's support schematic is clearly open. What's the drop
>> in part (note: Adafruit hasn't put the OSHWA logo on here, so it's possible
>> they don't claim this is open)
>>
>> http://arduino.cc/en/Main/ArduinoWiFiShield
>>
>> The WiFi radio on this board is proprietary, even though the support
>> processor and its firmware and board schematics are open. This is perhaps a
>> more complex board than the other two, but I'm not sure that complexity
>> changes things much. Or does it?
>>
>>
>> Contrast those three with this:
>>
>> http://logos-electro.com/zigduino/
>>
>> This is perhaps closer to the definition than the others, in that the
>> firmware for the radio module *is* open.
>>
>> My question is: do we need to differentiate between these in terms of
>> their openness,or not?  There are plenty of other examples I could pull. I
>> know my work would suffer if I decided not to use these parts, they're all
>> staples in my work. And I'm not an open source hardware absolutist, I use
>> plenty of proprietary hardware.  But I'm genuinely not sure where the line
>> is with some of the products we make and use every day.
>>
>> t.
>>
>>
>>
>>
>> _______________________________________________
>> discuss mailing list
>> discuss at lists.oshwa.org
>> http://lists.oshwa.org/listinfo/discuss
>>
>>
>
> _______________________________________________
> discuss mailing list
> discuss at lists.oshwa.org
> http://lists.oshwa.org/listinfo/discuss
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.oshwa.org/pipermail/discuss/attachments/20130307/3c880aa5/attachment-0001.html>