[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [GNUnet-developers] reuse of someone else's code

From: Igor Wronsky
Subject: Re: [GNUnet-developers] reuse of someone else's code
Date: Sun, 27 Apr 2003 20:13:03 +0300 (EEST)

On Sun, 27 Apr 2003, Krista Bennett wrote:

> I never said there was a problem with simply decompiling code. There IS a
> problem with including someone else's code in our source without their
> permission, patched or not. That is *both* a legal and an ethical issue.
> ...
> own); our argument with you has always been about the inclusion of someone
> else's code, however modified, in our source. The position on that has not
> changed.

Sorry for bumping in, but the following might yet again be stressed.

And this is entirely for Jan. ;)

Lets make it pretty clear.

  there were some kind of highly sophisticated proprietary codecs or
  other algorithms that simply couldn't be rewritten from scratch
  used through free software libraries, perhaps because of missing
  specifications etc.,
  some kind of desperate patch business might be in order for
  'the greater good', e.g. allowing some basic and common
  functionality on free platforms,
  in the case of GNUnet most of the problems or lacking functionality
  is best handled by just sitting down, thinking, and coding the
  thing -- and then the result won't be a perverse and fragile mess
  of this and that.

What I'd personally like to see is not reuse of incompatible code
but reuse or adaptation of good ideas. For example, I've been lately
impressed by the speed of the bittorrent protocol that I mentioned
earlier. It raises questions like a) what is it doing b) how is it
doing it c) could it be anonymized or d) could some of its ideas
be adapted to speed-up gnunet without compromising anonymity etc.
Anyone with some free time and a little background knowledge could
find answers to such questions, and imho something like that might
be much more fruitful than messing around with code reusing. And
bittorrent was just an example - unless Christian has read every
past, present and future paper on routing, graph theory and
cryptography, any technical library might hide a ton of secrets
with a tremendous practical effect if brought to light.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]