On 2006-11-24, Lourens Veen wrote:
> On Friday 24 November 2006 19:31, Petter Urkedal wrote:
> >
> > I think the issue of closed firmware value add boils down to how the
> > product is marketed.  We might require that when the OHF mark is
> > used, then the company must distinguish the features provided by the
> > hardware from those added by any proprietary firmware and drivers. 
> > That is, when OHF mark is used in marketing, the marketing must be
> > fair with respect to the OHF mark.  Otherwise would be misleading
> > advertising.  Once the FOSS community has created a firmware/software
> > stack, the implied features can be advertised along with the OHF
> > mark.  In practise this could be done by requiring that a statement
> > is placed next to the logo with information where to find the
> > alternative feature-set.
> 
> That's a good point. But what if the hardware didn't have firmware, 
> would open hardware for which proprietary drivers existed be any less 
> open? I'd say that in that case, the manufacturer would be entitled to 
> calling it open hardware, since the HDL is available and drivers aren't 
> hardware.

The point I'm trying to make is that the manufacturer must identify
the subset of the hardware, firmware, and drivers which they claim are
open.  Given that subset, it's either open or not, there is no middle
way.  The practical implication is that when marketed as OHF compliant,
the ad can only list the subset of features which are implemented by
FOSS software/firmware (or it could have a disclaimer next to the OHF
logo which tells where to get more info).  IANAL, but I guess the
trademark would serve as means to enforce fair advertising, and enable
us to use our definition of openness in doing so.

> So maybe the question is whether the firmware is part of the hardware. 
> If it is, then the openness of the firmware affects the openness of the 
> hardware. If it is part of the driver, then it doesn't.

But if the hardware is fully documented and it's possibly to upload new
firmware, then I can't think of a clear way to decide if the firmware is
part of the hardware or the driver.  From a philosophical point of view,
I'd never consider firmware part of the hardware, but I may consider a
"product" to comprise hardware, firmware, drivers, and even software and
services.

> I think it's hard to make a clear definition here. How is firmware 
> different from software?

I don't think there is a principal difference.  Two different
architectures, one of which might be pretty exotic.
_______________________________________________
Open-graphics mailing list
[email protected]
http://lists.duskglow.com/mailman/listinfo/open-graphics
List service provided by Duskglow Consulting, LLC (www.duskglow.com)

Reply via email to