[OH Updates] licenses wrap up...
g.andrew.stone at gmail.com
Fri Sep 9 11:50:08 PDT 2011
Thanks for your replies. Some comments inline:
On Fri, Sep 9, 2011 at 1:11 PM, Bruce Perens <bruce at perens.com> wrote:
> On 09/09/2011 06:48 AM, Andrew Stone wrote:
> For many of us, the licensee bought a product (the embodied hardware) so it
> was in fact a financial transaction. So could 7.3 actually be *damaging*,
> implying the converse?
> Whether or not you purchased the hardware, the source code and the rights
> conveyed by the license very definitely not conveyed in exchange for
> payment. The point the license text is trying to make is that you are not
> purchasing the source code and do not have the rights that a purchaser would
> have, even if someone has sold you hardware.
> 2. The TAPR does not disclaim warranty on the Product in section 7.1
> which seemed "bad" to me as I was working on my Open Device License, and so
> I checked CERN and in fact they do disclaim on the Product.
> Actually it does, you're missing the word "PRODUCT" in 7.2.
> But you're also missing the point that the TAPR license isn't connected to
> the product, just the documentation.
I'm not missing these points, in my original email I wrote: "And no you
can't argue that TAPR only covers the documentation, because it does cover
products in 7.2."
Section 7.3 specifically refers to section 7.1 and 7.2. Section 7.2 covers
the product which ain't free. I'd be ok if TAPR did not cover the product
at all (requiring a completely different document when the product is sold),
or if it fully covered it. But my point is that it seems to be partly
covered, and CERN seems to have thought similarly, since they modified 7.1
to add "product" language and dumped 7.3 altogether (section 5 in CERN).
> If you sell or convey a product and wish to do so without a warranty, you
> should disclaim the warranty in a document that is connected with the
> product. But if you are selling the product, disclaiming *all *warranties
> would probably be unfair to the customer. We should not make our license
> enforce unfairness to the customer.
Hmm, I agree with you for a "standard" unmodifiable device. But our
situation is very different. The problem is that many of the devices are
intended to be used in ways which the creators did not anticipate. That's
what is great about them! Sure, we could provide a warranty if used to only
do stuff we've already tried; maybe limit the warranty to the the provided
demo sketches. But if the warranty is that limiting, why bother?
In my ODL, I'm asking for "open hardware", not just "open source hardware".
So the maker of some device, say a toaster oven, has to leave a USB port on
there so the oven's firmware could be updated (an oven is of course a
worst-case example). Clearly once new firmware is uploaded the original
manufacturer should not be liable if the oven suddenly turns itself on at
2AM. Perhaps in real life the sort of devices that can easily cause harm
won't ever be made into "open hardware" by the original manufacturer.
But consider that people use Arduinos in "mission critical" applications
without even thinking about it. Like having them trigger fireworks, model
rockets, drive tractors, or allow real time tweaking of engine parameters
(I'm no expert but I think you can "tweak" these to the point the engine
would lock up, say going 80 on the highway).
Anyway, there seems to be nothing expressly geared towards limiting
> product liability and in particular protecting against consumer
> Actually, there's quite a lot. These paragraphs do that:
> IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW WILL ANY LICENSOR BE
> LIABLE TO YOU OR ANY THIRD PARTY FOR ANY DIRECT, INDIRECT, INCIDENTAL,
> CONSEQUENTIAL, PUNITIVE, OR EXEMPLARY DAMAGES ARISING OUT OF THE USE OF,
> OR INABILITY TO USE, THE DOCUMENTATION OR PRODUCTS, INCLUDING BUT NOT
> LIMITED TO CLAIMS OF INTELLECTUAL PROPERTY INFRINGEMENT OR LOSS OF DATA,
> EVEN IF THAT PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.
> 7.4 You agree to defend, indemnify, and hold each Licensor harmless
> from any claim brought by a
> third party alleging any defect in the design, manufacture, or operation of
> any Product which you make,
> have made, or distribute pursuant to this Agreement.
> 4. Finally, I hate to throw some cold water on the forums (because they
> are a great strength of OSHW) but from my reading, sellers may be liable if
> we "know" about intended misuse. So if someone sends an email saying "will
> this work?" and you (as the seller) reply back saying "not sure, just try
> it!" then you could be liable for that (mis)use of the product.
> I think you are reading about the law without fully considering two things:
> 1. The supporting case law, that is cases that have gone before in the
> jurisdiction or its superior courts, all the way up to the supreme court,
> and have been ruled upon, and which the judge must consider when ruling. So,
> to understand the law you also have to read cases. 2. That it's judges who
> rule, not computers that mindlessly interpret the law.
> It's not tremendously likely that a producer of equipment that is meant to
> be tinkered with who says "I don't know" to a participant in a tinkering
> forum will end up being liable for the misuse. Were I hired to present
> expert testimony upon such a case, I don't think I'd have much trouble
> explaining the clearly experimental nature of such a device and where the
> responsibility would lie.
Well, I am well aware that I am not considering many aspects of the law due
to lack of knowledge. That is why I am posting. I'd be interested in any
supporting cases, perhaps from makers of "kits" that went bad.
I agree, its not likely. That was my "weak" assertion. But my "strong"
assertion is that I think that if you send a casual response email that is
incorrect or misleading (after all, nobody is perfect every time, especially
for "free" support) it may be much more likely to go against you. And the
prior sale of product is a financial transaction so perhaps judges and
juries would not see it as just firing off a casual post on some forum. I'm
guessing there's a good reason why lawyers and financial advisors always put
that "does not constitute a relationship" disclaimer on the bottom of every
> On the other hand, the opaque sort of equipment which is marketed for the
> naive user has a more severe liability load connected with it, especially if
> it harms the user with something like an electric shock, and the user can't
> be expected to understand electricity. So, if you are marketing a Chumby to
> end-users at Best Buy, you have a greater liability load in connection with
> the naive folks, even though it happens that tinkerers can download
> documentation and you are not so liable regarding them.
Is this supported by case law? In fact, I think that this idea --
basically since the "licensee" contains the full design files the licensee
must determine suitability for a particular application -- may be an
adequate defense for certain regions especially if certain language is used
in the license agreement.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the updates