Why did you choose DragonFly?

Samuel J. Greear sjg at evilcode.net
Wed Sep 22 10:57:21 PDT 2010


Samuel J. Greear <sjg at evilcode.net> wrote:
 > What has drawn you to use the DragonFly BSD operating system and/or
 > participate in its development by following this list? Technical
 > features, methodologies, something about the community?

Thanks all for your responses, this has been very enlightening to me
and I hope to others as well. I was somewhat surprised, but perhaps I
shouldn't have been, that the community seems to play such a role for
so many people.

That is largely the reason I am here as well. I have been a FreeBSD
user since 1998 and oversaw and maintained several rather large
deployments over the years. I was also a developer, but usually a
couple of layers above the OS (most often for the web). When I did
contribute patches to FreeBSD it almost seemed like a fight to get
anything to happen with them, oftentimes they were rejected on the
basis of white-space or etc., rather than merit. Whether they were any
good or not, well ...

I have been following DragonFly BSD since the fork and I have greatly
appreciated many of the design decisions that have been made in that
time. I have poked around in all of the BSD's kernels and I would
consider DragonFly by far the cleanest thanks in large part to the
years of cleanup and source documentation work done by Matt. While
finding someone to test patches can be a bear sometimes, the community
is usually very willing to accept and provide input on odd concepts
and alternative ways of doing things. For instance, sfbuf's, although
novel have kind of rubbed me the wrong way since the early 2000's.
When I finally decided to do something about them quite a few people
were eager to offer their own ideas and approaches, explaining
DragonFly's unique locking / kernel methodology along the way. lwbuf's
have since been committed, I'm not sure I would have had the ambition
to follow through and get them committed faced with some of the
"obstacles" I have hit in other projects. So a big thanks to all of
the recently active committers and contributors on that front, you
guys are all low pressure and make things pretty easy.

I also get the feeling that DragonFly really is what the developers
and users make of it, more than anything. It's not a "toy" project but
it's also not necessarily a 100% general-purpose project like FreeBSD
aims to be. My personal aim is for it to be a seriously reliable and
scalable piece of infrastructure supporting things like clustered
web/cloud deployments and clustered/redundant NAS. I don't feel, as a
develope/user, like there are any major hurdles to my making changes
to DragonFly toward this end apart from my own ambition and the amount
of time I am able to contribute. That might seem basic, but it is a
very liberating feeling having been involved in various other
projects.

Thanks again,
Sam






More information about the Users mailing list