On Fri, 7 Apr 2000, Ed Carp wrote:
>>Scott A. Yoder (syoder@ims1.imagestream-is.com) writes:
>>
>>We've been having this *huge* debate internally over (1) do we go open source and let support/consulting drive revenue, or do we go binary-only and let the development side pay their own way? It's a huge problem, and being in the middle of the fray, I
can understad both sides. I'd like to be able to make a living off this, but I'm so afraid of getting ripped off it's not funny. And, of course, there's the Miller Effect to contend with - no matter how good your code is, there's always someone out ther
e who's going to make a comment about it's esthetics or style - and I like having my code laughed at as much as the next guy. People who know me for a while have taken to calling it the Carp Effect - the closer it ets to release time, the uglier the code
gets because of the hacks and bug fixes!
I know exactly what you mean. The code does get uglier as it progresses. All code does because you always end up having to
support some feature that doesn't quite fit right with your design, but of course that feature is needed yesterday because
someone had to have it to close a sale that must ship overnight tonight.
Oh and you're learning as you go too. There are so many details to learn about the kernel. Nobody, even Linus himself had it figured
out with version 1.0. As you learn something new, it gets used from that point forward. Of course you wait until version 2.0 (or if
a section really, really bugs you and you have time you rewrite it). You can imagine what the code looks like.
>>I understand the rock/hard place - everyone's got to eat. But Diamond finally gave in when they realized they were missing a huge market by making people sign idiotic NDAs. NDAs for public hardware products don't even make any sense - you *want* peopl
e to use your stuff, and the best way to do that is make it as easy as possible for people to write to your APIs and to write drivers, and you don't do that by going to NDA route. Penny wise and pound foolish.
It doesn't make sense. I think SDL wanted to get a jump on their competition and protect their hardware designs.
But slapping an NDA on their DDK doesn't really accomplish that objective.
>>> I can't release the WANic source code because of that damn NDA with SDL. I also can't
>>> release the Frame Relay source (as written) because of similar licensing problems.
>>>
>>> So I rewrote the crap they had with a design that allowed me to release as much of the
>>> source code as I legally could.
>>
>>You might want to consider having someone else who has never seen your code do a rewrite. That would avoid the licensing issues.
The Frame Relay source really isn't my code. I've had to provide at least 100 bug fixes to it. It's not going to be
hard to rewrite the code. I want to do it from the ground up anyway. If you guys think my core code is ugly, you'd
roll around on the floor laughing at the original Frame Relay source code!
Ok. Enough list bandwidth.
Good luck with 2.0,
Scott
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sat Apr 15 2000 - 21:00:11 EST