[3309] in linux-net channel archive

home help back first fref pref prev next nref lref last post

Re: Binary Driver Issues

daemon@ATHENA.MIT.EDU (Doug Ledford)
Tue Jun 18 19:49:26 1996

In-Reply-To: <Pine.LNX.3.91.960618200221.326A-100000@devel.conexio.co.za>
Reply-To: dledford@dialnet.net
Date: 	Tue, 18 Jun 1996 13:31:55 CDT
From: Doug Ledford <dledford@dialnet.net>
To: Mike Kilburn <mike@lserv.conexio.co.za>
Cc: Dennis <dennis@etinc.com>, linux-net@vger.rutgers.edu


On 18-Jun-96 Mike Kilburn wrote:
>>
>
>
>But what I am saying is the opposite. The low level board driver can
>be binary only since that is an extension of the firmware (or lack of)
>for the card. While the "value-added" kernel enhancements should require
>source. I know this will never happen with ET. But *new* companies can 
>plan from the start to make a profit and be free.

I guess this is where we differ.  The way I see it is this.  ET has written
what you call a kernel extension.  You feel that the source for kernel
extensions should be distributed.  I believe you mentioned doing this to
avoid raping the linux kernel (since it is free) and to contribute something
back to the community that contributed Linux.  The ET protected info is
considered to be a kernel extension because it loads into the kernel module
space and interacts at the hardware control level as a part of the kernel.

Here's what I know about these things, and how I feel about them from my
own experience.  You can extrapolate these things back to ET, Inc. as you
see fit.  Unlike the vast majority of people working with linux and writing
code for linux, I don't have any other job besides my linux work.  If I
write some usefull code that is not truely unique and valuable, then I give
it back to the community.  On the other hand, if I set down and write
something that is valuable, I have a responsibility to myself, my wife, my
creditors, etc. to NOT give that piece of information away, but instead, use
the sell of it to finance all of the other stuff I am doing as well.  As
such, the kernel extension you speak of is ET's "valuable" work.  Furthermore
it's not even Linux specific.  I don't think I could tell someone they have to
give their code away when I know what circumstances I work under.  Those of us
that work 60+ hours a week on the linux operating system and have no other
income, have to get what we can where we can in order to survive, and maybe
even, God forbid, make enough to take a break sometimes :)

Also, I sent through a message to the list, but I think it got killed off
before it got out :)  I know that one of the things people have been talking
about is the idea that ET should release the driver and make their money from
the board.  In that message I mentioned that I understood ET's position because
the board market is very competitive and the deciding factor on my purchase of
their board was their software.  Take that away, and ET very well could have
lost the sell.  Apply this to the statement you make above about new companies
planning from the start to make a profit and be free.  I think you are right,
as long as what makes this company better than the others, what gives this
company the competitive edge, is NOT the software, but the hardware.  If this
isn't the case, then the company can't be free with its software and stay in
business.

*****************************************************************************
* Doug Ledford                      *   Unix, Novell, Dos, Windows 3.x,     *
* dledford@dialnet.net    873-DIAL  *     WfW, Windows 95 & NT Technician   *
*   PPP access $14.95/month         *****************************************
*   Springfield, MO and surrounding * Usenet news, e-mail and shell account.*
*   communities.  Sign-up online at * Web page creation and hosting, other  *
*   873-9000 V.34                   * services available, call for info.    *
*****************************************************************************


home help back first fref pref prev next nref lref last post