Communicating requirements to kernel developers

As I understand it CC'ing lkml is accepted practice because it means kernel
developers get a heads-up/chance to review if a feature for a sub-system
touches other non sub-system code.

lkml can seem like a fire hose of information but I don't know how many
people actually read all of it and how many rely on filtering it for
interesting topics. Although I subscribe to lkml I only really follow
threads I've marked as interesting and/or involved with. To keep upto date
with the kernel in general I rely on the more usefully condensed information
on our very own LWN Kernel pages and more recently the very useful Kernel
Summary Podcast.

That's pretty unsurprising, given that SubmittingPatches says:

"Unless you have a reason NOT to do so, CC linux-kernel@vger.kernel.org."

>>The volume on linux-kernel is far too high for ordinary people to cope with.

>That I claim is a problem with developers always cc'ing linux-kernel. If
>there is a subsystem list on vger — linux-fsdevel, linux-netfilter, you
>name it — at best only post there.

Well i have read Ingo Molnar opposing this a lot of times in lkml... he
claims that everything has to be cc'd to lkml (his tipbot does exactly this
and the result is a lot of big duplicated mail/patches on subscriber's
mailboxes)