Feedback from evaluation in a corporate environment

[Philippe Lhoste]

On 08/01/2010 11:37, Paul Moore wrote:
> If bzr init is *not* the correct command for beginners to use, it
> should probably be removed or renamed as it's an attractive nuisance.

I understand the irony (yes?) but I felt I should explain why a simple 'init' is still useful.

Perhaps I am just a dumb user, but for most of my repositories, I use only a simple "bzr 
init" command... That's nearly one year I use them, with 171 revisions in the biggest one 
(so they are still of modest size).

Now, I have a rather unusual workflow (perhaps), which as described here before, and was 
said as "correct"...

I remember when I first read the Bazaar documentation, I was quite confused why there was 
a init-repo command while we have a init one. Perhaps it was clarified now, and after 
reading again I understood better anyway.

For the record, here is my workflow:
Most of my coding is made of small experiments in various languages (no big project yet).
So I have a repository (made with 'init'!) per language... (Java, JavaFX, Processing, Lua, 
Nimrod, etc.). Also one with my SciTE (text editor) settings (lexer properties and so on).
These repositories have a number of sub-folders, holding several little programs, or one 
folder per small program, etc.
Basically, I don't feel the need to create derived branches, unlike the usual workflow of 
DVCS: when I want to change something, I just hack the program, then commit when I made 
significant changes or finished.
Of course, I am the only coder on these projects... The 'lone programmer' scenario isn't 
so unusual, I think.
So if anything goes wrong, I can just revert, and continue to hack.

I suppose that on a bigger project, I would do feature branches, and so on, but here, no 
such need.

I also push my branches to Launchpad (to +junk branches as they are not worth of full 
fledged projects) and pull/merge from there, on two computers.

-- 
Philippe Lhoste
--  (near) Paris -- France
--  http://Phi.Lho.free.fr
--  --  --  --  --  --  --  --  --  --  --  --  --  --