Re: [Gnu-arch-users] Tutorial Question

[Pierce T . Wetter III]

> >  Ok, lets see if I understand branching.
> >
> >  Ignoring the step where we make yet another archive. (Why is this
> > here? Is this a "best practice" thing that people tend to make a
> new
> > archive when they branch? why?)
>
> No.  From what I remember, he's simulating two-developer workflow
> here.  It would be common and probably preferred to have an
> archive for each developer's branches.

   OK, so its because people tend to have personal archives then. He's 
really got three developer work flow, Alice and Bob are sharing an 
archive, then candice makes her own, and creates a branch. So we go 
from one archive to
two, for what seems like no reason.

> >    I named the branch "punctuated", not "candice". I think that
> makes
> > more sense, because "punctuating" is the purpose of the task, and
> > there's less confusion between the "branch" and the "archive". I
> > recommend changing the tutorial to reflect this.
>
> And when the next change on that branch has some other purpose?
> The idea is that the branch is for Candice's changes, not
> necessarily only this one.
>
> But, anyway, arbitrary.  It doesn't matter what it's called.

  It mattered to me, partially because it confused me with the
archive of the same name, but also because branches seem task focused 
to me, not user focused. I also think "detours" are more typical then 
branches, in that code forks, makes a few turns, then gets merged back 
to main.

 candice might have other tasks she would be doing in parallel, right?

  hello-world--punctuated--1.0
  hello-world--using_sockets--1.0

 So it seems like using a person's name as a branch label would be bad
practice.

 Probably the real reason though is that I tend to "branch" only when I 
need to collaborate with another developer, so they wouldn't be "my" 
changes.

 Pierce