[Aptitude-devel] Inconsistent help message
dburrows at debian.org
Sun Mar 30 03:07:25 UTC 2008
On Sat, Mar 29, 2008 at 07:23:10PM +0100, Jens Seidel <jensseidel at users.sf.net> was heard to say:
> On Sat, Mar 29, 2008 at 08:57:08AM -0700, Daniel Burrows wrote:
> > Sure, but I can't find you in the list of alioth accounts. What's
> > your username?
> It's jseidel-guest. But do I need to be added to the project? I first
> thought "yes of course" as it is required for non distributed systems
> such as CVS or Subversion. But isn't Mercurial a distributed system?
> To be honest I do not have much experience with distributed systems
> (but will learn it quicklywithout doing any harm) since I work mostly
> with svn nevertheless I would prefer not to make a own repository
> available :-)
It depends. With a distributed system you *can* work in your own
repository and periodically either send me patchsets via email or tell
me to pull down your changes. This is very good if you're doing work
that shouldn't be published immediately, if you're doing work that
conflicts with the main-line all over the place, or if I don't trust
your changes without review. On the other hand, it's less work all
around if you can publish directly, and our experience has been that
this isn't a problem for translation changes: they generally don't go
through extensive periods of being totally broken, and they generally
don't conflict with mainline changes (since I almost exclusively work
The biggest benefit to me of a distributed system is that I can work
while I don't have Internet access. A secondary benefit is that I get
checkpointing and versioning for code that I'm not ready to publish (my
local branch occasionally diverges quite a bit from the head branch
before I push my changes up).
> PS: Thanks for CC:ing me. I'm not subscribed yet but will change this
I'm glad I did; TBH, I probably did it by accident (although I see now
that you specifically asked for it, so I'm Ccing you again).
More information about the Aptitude-devel