[Ksummit-2013-discuss] [HOBBYIST ATTEND] Typical hobbyist problem: I have some spare time, kernel experience and want to do some kernel hacking

Andy Lutomirski luto at amacapital.net
Sun Aug 25 17:19:03 UTC 2013

On Sun, Aug 25, 2013 at 5:48 AM, Peter Hüwe <PeterHuewe at gmx.de> wrote:
> Am Freitag, 23. August 2013, 03:08:20 schrieb Andy Lutomirski:
> However almost every time you read an article about LKML it contains a
> statement like "it's a very harsh environment, but don't take it personally if
> you get flamed".
> I know the intention of this statement, and while it might be true, I'm not
> sure whether this actually creates the opposite effect for some people.
> Especially for people of other cultures where the risk of losing one's face is
> taken pretty seriously.

Definitely true.  It can be hard to realize "this piece of your code
is utter *expletive*" actually means "I think the general concept is
okay, but you need to rethink this piece of your code, and I think you
can make it work.)

On the other hand, after grad school, this is second nature. :)

> Imagine a house with thousands of "beware of the dog(s)" warning signs (in
> this case maybe dragons because of flaming ;), the dogs standing at the fence
> baring their teeth already and the owner yells to you "come in, and nevermind
> the dogs, they won't bite they only want to play".
> If I don't know the dogs yet, I would probably wait at the fence.

>> If you want to find something to do, read a random lkml post by Linus,
>> Ingo, etc where they suggest how something could be better and do it.
>> The community could probably do a better job of getting the message
>> out, though.
> Yes - but how ? ;)
> Maybe tag it somehow so it is at least searchable?
> I doubt the chance of spotting a valuable suggestion/idea in message 245 out
> of 342 in a heated discussion is close to zero by hobbyists which aren't
> involved in the discussion in the first place.

True.  And a lot of people probably get bored by message 150 or so and
never get there.

Something like a searchable kernel wishlist (maybe indexed by who made
the wish) could be helpful.  Maintaining it might be hard, though.  It
would need the buy-in of at least a couple core people for it to be
any good.

>> > - "assign"/"connect" them to companies who are in need of Linux Kernel
>> > Skills (but maybe cannot pay for it) (there are still a lot of crappy
>> > inhouse kernel- stuff floating around in a lot of companies, done by
>> > somebody who knows how to spell Linux)
>> I'm interested in the other side of this: how do companies find kernel
>> people?  My company has a decent collection of kernel changes I'd like
>> to see, I have limited expertise, and I don't have nearly enough time
>> to do them all.  Sticking "system programmer" on a career page doesn't
>> actually work very well.
> Interesting aspect. We do have a somewhat similar problem.

I wonder what it looks like from the hobbyist side.  Are there a bunch
of people who want a job where they can still hack part-time (paid!)
on the kernel?  Where do they look?  Do they find these jobs?


More information about the Ksummit-2013-discuss mailing list