Subject:
RE: Stupid licensing thread (Was: Request for comments - Microwindows)
From:
"Vidar Hokstad" ####@####.####
Date:
5 Oct 1999 08:02:55 -0000
Message-Id: <19991005075905.8490.qmail@mail.relight.com>
On Tue, 5 Oct 1999 09:16:44 +0200 you wrote:
>
>Hi,
>I think this remark is stupid!
>Of course we don't want that the code we wrote is used by commercial
>companies,
For some, that holds true. For me the only reason I have the time to
work on NanoGUI and NanoWidgets is because I'm _paid to do so_,
because the company I work for see the benefits of supporting open
source where we can.
>but don't you think that those companies can make great steps in
>free software?
Of course I do. I'm paid to do development on free software. I wouldn't
even be in this job if I hadn't been promised from the start that we
would contribute code back.
>Example:
>Do you think Linux would be what it is today if RedHat, SuSe, Caldera,
>etc... don't make money on it. And don't forget that RedHat for example use
>ONLY free software!
>
>If we think like you think : "... those companies can write their own
>library...", the work of these companies is just loosed for open-source
>software.
I agree, and if you reread my message, you might see that I made that
remark to show people that choosing a restrictive license will have a
very real effect on what people are willing to contribute. We (Screen
Media) _can't_ spend lots of time contributing to a project if the
license is too restrictive for the project to be of any use for us.
But for us the real result of a too restrictive license would be that
we would _have_ to write our own library, or find a less restrictive open
source project to contribute to.
>If we think that those companies that have people working
>full-time on open-source software can add value to open-source software and
>help our work in this area, it will be a good thing.
>But that don't means that those companies must release all their work under
>open-source. Think of Oracle, Sybase, etc... porting their DB to Linux. They
>use the work of the glibc coders, the kernel coders, etc... but don't
>release their source-code. Nobody is complaining about that.
>
>That's the reason why the LGPL, MPL licenses were created. And we MUST use
>it where it's possible to use it. Don't forget too, that KDE and GNOME is
>developped by some people that are paid by commercial companies to work full
>time on it. It won't be possible if those companies don't make money on
>open-source software.
I mostly agree. I think the LGPL and GPL is fine for _most_ software.
Where I think they are too restrictive is for important parts of our
"infrastructure": system libraries, code that we _want_ everyone to use
because we want it to become standard. If NanoGUI shall have any chance
of becoming pervasive in the embedded/small systems market, then it need
a license that means that embedded and small systems developers _can_
use it.
And this market is a bit special, because it is a market so extremely
dependent on size. By going with a commercial browser instead of an
open source one (Mozilla), and paying both porting costs (to port it
to NanoGUI) and license costs, we save a lot of money because the
commercial browser is small enough that our hardware costs go down
enough to _more_ than cover the licensing costs.
For high volume, low margin markets, cost cutting like that is essential
to survival. If we want our company to survive, we _can't_ choose a
program that means we will have to add 8MB more flash, like we'd likely
have had to do with Mozilla (and for the record, the company we're
cooperating on with the browser has been very reasonable about licensing
issues compared to some of our other partners), just to go with an open
source product.
I'd rather use Mozilla, if it was economically feasible for us, because
I see tremendous benefits in having full access to the source, and being
able to do modifications (and contribute them back to the main tree).
But open source doesn't work for everyone and in all cases, and if the
goal is for NanoGUI to be "X for small systems", then we need be pragmatic
about making it possible for everyone to use it.
Regards,
Vidar Hokstad
