Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 29 Sep 2006 23:59:13 +0100 (BST)
From:      Robert Watson <rwatson@FreeBSD.org>
To:        Julian Elischer <julian@elischer.org>
Cc:        freebsd-current@freebsd.org
Subject:   Re: isofs/cd9660 -> relocate to fs/isofs/cd9660?
Message-ID:  <20060929235659.J73166@fledge.watson.org>
In-Reply-To: <451D93EE.9040309@elischer.org>
References:  <451ADC21.50206@centtech.com> <200609271727.29775.jhb@freebsd.org> <451D4787.4050309@samsco.org> <200609291618.09492.jhb@freebsd.org> <451D93EE.9040309@elischer.org>

next in thread | previous in thread | raw e-mail | index | archive | help

On Fri, 29 Sep 2006, Julian Elischer wrote:

>>> What about moving all of the net* directories into /sys/net?.  And don't 
>>> forget putting i386 and friends into /sys/arch!  Ah, I love the smell of 
>>> fresh paint in the morning.  Smells like.... napalm.
>> 
>> Baby steps aren't hard. :)  Back when I first made rumblings about this 
>> sort of thing we didn't have a sys/fs at all, but now we do and over time 
>> we've actually moved most of our filesystems into it. :)
>
> there was a sys/miscfs which could have been used..
>
> Matt Dillon took the oportunity to redo the tree in DF. I wonder how that's 
> working out?

I think the best model for handling utterly gratuitous source tree 
rearrangements is to do them only in the context of the code finding a 
maintainer, and in the context of a larger piece of work.  I.e., if a 
developer finds their location irksome, they can first agree to take 
responsibility for the code before randomly moving it around, rather than 
having it be a one-pass drive-by.  For example, I recently had uipc_socket2.c 
renamed to uipc_sockbuf.c (mostly) as part of overall cleanup and work on the 
socket layer.  I would not have dreamed about doing this if I hadn't already 
been working on the socket layer heavily.  This is something along the lines 
of "you break it, you buy it", only it's a case of "You fix it, you avoid the 
bikeshed".  :-)

Robert N M Watson
Computer Laboratory
University of Cambridge



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060929235659.J73166>