I backed out the sensors framework

[Poul-Henning Kamp]

In message <019901c80fe8$3cb62b20$6401a8c0 at Mewtwo>, "YoctoGram" writes:
>> If it is biased, it is biased against the code and not the person(s).
>
>I'm sorry, but who are you to decide?

A fair question I guess.

I am undisputedly the person who has been most active in FreeBSD
kernel development for the longest time.

I have severely cleaned up the VFS system and the attached filesystems,
written DEVFS, GEOM, a widely acclaimed implementation of malloc(3),
the timekeeping code, a disk encryption facility, large parts of
the FreeBSD release building and installation code, from scratch.
Add to that a bunch of device drivers and various other stuff.

In terms of raw numbers, the FreeBSD CVS repo statistics say that
I have made 4707 commits to the kernel, 25% more than number two
on that list.  If you add in the 2367 commits I have made to the
rest of the source code, you will find that on average, I have, on
average, committed something to FreeBSD every 17h20m, over the last
14 years.

I have also been the FreeBSD release engineer and a core team member,
for far longer time than is good for any persons sanity, and have,
through insight, reasoning, argument, manipulation, deceit and
downright trickery been instrumental in getting FreeBSD to where
it is now.

That of course does not mean that I get to make decisions in a
vacuum, or "just because", that's not how FreeBSD works.  But it
does have a tendency to give my opinion some weight.

I am also coincidentally, the perpetrator behind the "bikeshed.org"
meme.


Now, who are you, and what is your role in this ?


>Who did accept and approved this project in the first place?

Google and the FreeBSD SoC mentors I pressume ?

>Why should Google be spending any more money on "approved" FreeBSD SoC... 
>if, at the end, it get "backed out" ?

Ahh, but you seem to labour under the misunderstanding that SoC is
a fast path to the FreeBSD source tree.  That is not the case.

Being approved for SoC means that the project is a good project for
SoC, it does not imply that the code will end up in a FreeBSD release.

As far as I have kept score, only around half of the SoC projects
end up being imported in CVS, and usually only after significant
more work.

That is still an impressive score by all means, and I fully support
the SoC programme: how I wish such an opportunity had existed when
I was younger.

>So, if there was a problem with "the code"... why didn't you (or any of the 
>FreeBSD leaders) took action(s) sooner?

I did, and people didn't pay sufficient attention.

Poul-Henning

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.