Interested in Participating

[Lachlan Patrick]

Matthieu Moy wrote:
> Lachlan Patrick <loki at research.canon.com.au> writes:
> 
>> Yes, I think the only thing I need at the ISP is a SFTP daemon so that I
>> can push things up to it? Bzr on the ISP would be just so that I could
>> do a quick fix while logged in there and push it into the ISP-local
>> repository,
> 
> Whell, any time you can push on one side, you could have pulled from
> the other side.

I'm not so sure. If I have a work PC within a private network /
firewall, I might be able to use it to push to and pull from an ISP, but
be unable to push to or pull from the PC (because nothing on the WWW can
address it). Maybe I'm misunderstanding bzr here...

> Advantages of having bzr on the server are mostly the ability to do a
> checkout there, and ability to run a smart server, but this will
> become interesting when the smart server is actually there and fast.

I don't understand what is meant by smart server, but that's all right.
One attraction to bzr is it requires no server setup on the ISP. The
fewer installations, specialised configurations or extra points of
failure I need to deal with, the better.

> BTW, "b z TAB" = 3 keys. "b z r" = 3 keys too  ;-) .

Ah, good point, I was just thinking of newbies like myself trying to
find tools, more than using tab completion to type 'bzr'. E.g. "b z TAB"
on my Cygwin installation reports:

bzcat.exe         bzegrep           bzip2.exe         bzmore
bzcmp             bzfgrep           bzip2recover.exe
bzdiff            bzgrep            bzless

all of which seem to be related to bzip2. 'Bzr' just looks like "yet
another bzip2 tool" from that perspective. It's like the 'bz' prefix is
owned by bzip2, just as a 'z' prefix is owned (to a lesser extent) by
gzip (how inconsistent is that?). It's not a functional difference at
all, as you rightly point out. Meh.

Loki