linux.conf.au
The Trivial Patch Monkey Trial

From: torvalds@transmeta.com (Linus Torvalds)
To: linux-kernel@vger.kernel.org
Subject: Re: graphical cset stats
Date: Wed, 24 Apr 2002 04:18:45 +0000 (UTC)

In article <20020423203211.7a6d7078.dang@fprintf.net>,
Daniel Gryniewicz  <dang@fprintf.net> wrote:
>
>But it is interesting how it's an exponential progression.

I think it's a bit dangerous, because it is so misleading. Both the "top
performers" are clearly integrators rather than big coders, and I
suspect a lot of my "cset-points" are actually from the early BK tree
creation where every single cset got attributed to me simply because
they got merged from the historic non-SCM patch info.

Certainly looking at the SCM statistics from the last 500 ChangeSets, 63
of them were attributed to me ("Hey, Linus does 12% of all kernel coding
himself! Studly, man!"), but if you actually look at the details of the
changesets, you'll notice that I'm a total loser, and I end up doing
little coding and most of my changesets are merges, cset excludes,
kernel version updates etc ("Hey, Linus is a complete moron!").. 

So I personally get a bit nervous about pretty graphs - they _seem_ to
say so much, yet they clearly don't tell enough.  Which can be a bit
dangerous if somebody takes them too seriously. They're just simple
enough that you think you get the RealTruth(tm).

 				Linus