On Wed, May 11, 2005 at 09:41:46AM +0100, Hamie wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Eric Smith wrote:
>
> > Timothy Miller wrote:
> >
> >> We want to give something to individual users. We do NOT want to
> >> give anything to big corporations that want to benefit from our
> >> work without giving anything in return.
> >
> >
> > I assume that you're not worried so much about FPGA-based products,
> > but rather about a company spinning their own ASIC based on your
> > RTL. If your RTL is GPL'd, they can do that. A solution is to use
> > a license that does not permit commercial redistribution. The
> > license could be otherwise similar to the GPL. Technically it
> > wouldn't qualify as "open source", but it would be close enough for
> > anybody not trying to use your own RTL to compete with you.
>
>
> Ah... Yes it would qualify as open-source. This is exactly what the
> LGPL does for TrollTech and Qt. You can download the code, use it as
> you wish in any GPL product, and give you apps away, BUT as soon as
> you want to use it commercially (i.e. to develop a program that is
> closed-source and you're selling) then you are required to PURCHASE a
> license from Trolltech to do so.
It's starting to sound like a choice between dual licensing and
trade secrets. Of course, the details are key.
Here's another wrinkle.
Of course we understand that all outside contributions are under the
LGPL. However, an important motivating factor for contributors is that
their contributions are part of a fully functioning system. The best
example is the way open source development suddenly took off when the Linux
kernel was added to the GNU utilities and libraries, forming a usable
open-source OS for the first time.
This means that if a non-zero amount of proprietary or closed-source
code is necessary to complete the system and make the board work, the
motivation will be a lot stronger if there is some form of firm committment
to keeping that logic permanently available, even if the company goes out of
business or is taken over by new owners with different policies.
There are probably a number of ways this could be satisfied, but it
should not depend on good will alone.
If it's necessary to keep some of the code closed for a time, a copy
might be placed with a trustee, instructed to publish it on a certain date.
A bank might do as a trustee; safe deposit boxes are pretty economical. The
date of full open-source release might be set some years off; that should be
OK with contributing developers and embedded customers as long as it can't
be unilaterally extended. If we need to subscribe money to compensate the
company for an irrevocal commitment to open-source it at a certain time,
possibly the price would decrease for a longer time frame until the release
date. Personally, I wouldn't care if the release date comes after the
design is commercially obsolete, as long as it's certain.
Possibly embedded customers might be offered a slightly different
deal, tied to the first-refusal idea: for a specific license fee, they could
buy the right to obtain a copy of the documentation directly from the
trustee and manufacture units for their own needs, if the company failed to
accept an order and deliver, regardless of reason.
But I still don't see how trade secret protection is compatible with
FPGA development in the field. Is it possible to add compiled logic to an
FPGA load image without revealing the HDL source code?
_______________________________________________
Open-graphics mailing list
[email protected]
http://lists.duskglow.com/mailman/listinfo/open-graphics
List service provided by Duskglow Consulting, LLC (www.duskglow.com)
