[linux-elitists] Re: SCO Teleconference transcript (Friday, May 30, 2003)

Brian McGroarty brian@mcgroarty.net
Wed Jun 4 05:42:47 PDT 2003

On Mon, Jun 02, 2003 at 11:56:26AM -0700, Nick Moffitt wrote:
> begin  Wendell Cochran  quotation:
> > The SCO farce begins to remind me of the old Shakespearean-code hoax
> > (cf the extant Bible Code): any sizable text can be made to produce
> > any message desired.
> 	That's precisely what I was thinking of when I posted the
> above.  
> 	I don't think you'd find anyone in the lkml who thinks that
> "hundreds of lines" would be difficult to replace.  I think you'd be
> hard-pressed to find people who think that "hundreds of lines" are
> even *incriminating* if they're speckled about the place in chunks of
> a dozen or so lines.

I've been thinking about this.

One could preprocess C source, running the C preprocessor output into
a stream of C tokens, a generic symbol representing all specific
symbols, and discard everything else. With the resulting stream in
hand, it should be possible to compare large codebases and quickly
find expanses of 'common code.' This would work even when the source
formats and naming conventions are completely different.

If anyone's got their hands on SCO code, I bet we could "prove" it's
been derived from nethack, emacs -and- vi, or something similarly

Damning chunks from five to ten or fifteen lines, they say?

More information about the linux-elitists mailing list