On Sun, Oct 30, 2011 at 7:03 PM, <[email protected]> wrote: > I know it's kinda OT, but was there a particular reason you wanted more > than btrfs could offer?
No, I like btrfs, but as far as I know Plan 9 doesn't support it. > There's a port of Plan 9 utilities to Linux userspace. It's called Plan > 9 Port (aka p9p). See http://swtch.com/plan9port/. Linux machines with > p9p should be able to talk to fossil/venti. Okay, thanks, I've been using p9p to play around with acme and such on linux, but I didn't know it could do this. > Of course, if your "external drive" contains a CPU (i.e., one of the NAS > boxen being sold as "external drives"), the "drive" itself may be able > to run 9p. :) Hm, I don't think so. Maybe external drive was the wrong term, what I have is actually just a little enclosure thing that I pop a normal 3.5" drive into and power on, which then has a USB port. > Would replica meet your needs? Under some circumstances, rsync > outperforms replica. For some purposes, a DVCS such as Mercurial (which > has also been ported to P9) is more appropriate. It depends on a mix of > factors. Thanks, I'll have a look at replica. On Sun, Oct 30, 2011 at 7:35 PM, andrey mirtchovski <[email protected]> wrote: > I've been using the ufs version written in Go by Latchesar Ionkov > (go9p.googlecode.com) recently with great success. It connects the > underlying OS with Plan 9 on a hosted virtual box image, linked via a > bridged network. It's not completely bug-free so you can't expect to, > for example, do a complete build of Go over it, but it is at worst > 1.5* slower and at best 2* faster than a native fossil. Interesting. Incidentally I've been wondering what the status of Go on Plan 9 is lately. cat-v.org tells me "Status: Being integrated with mainline", but they haven't updated in a couple weeks and I can't seem to find much other info.
