[En-Nut-Discussion] RFC: Moving to github

[Thiago A. Corrêa]

On Tue, Jul 21, 2015 at 10:43 AM, Harald Kipp <harald.kipp at egnite.de> wrote:

> Hi Thiago,
>
> On 21.07.2015 15:06, Thiago A. Corrêa wrote:
> > IMHO it would be best that the "official" repository to be git repository
> > then. Hosted anywhere, either SF, GitHub or GitLab.
> > Release files could be kept in SF.
>
> I can read this statement, but I'm missing an argument that may convince
> others that this would be the best solution.
>
>
> > I understand the argument that devs break other platforms unknowingly,
> but
> > still I don't think making "official" fragments would be beneficial, it
>
> Aha! So... we like decentralization and anarchy, but somebody has to
> tell us, where is the focus and whom we have to follow. Anarchy for
> bookkeepers. ;-)
>

I never said I like it. I tolerate git. If stackoverflow ever goes offline,
I would be unable to use git at all :)
IMHO decentralization in git is just an optimization that makes bisect
tolerable in large trees, doesn't imply in anarchy :)


>
> What's your definition of "an official repository"? The one that tells
> you: "Yeah well, I'm buggy, break old apps, don't care about licenses
> and patents, but this label on my back clearly states, that I'm the
> official guy." Probably not. What you need is a repository, that fits
> your needs. If one exists, support it. If not, select the next best fit
> and create a fork.
>
>
Forks will always exist, but it doesn't mean we have to follow that as a
development model.
The official tree is where ethernut.de points to. The place where we want
to upstream our changes that we feel we should share, or really, as Nathan
pointed, wherever Harald is commiting to :)



>
> > would be just another place to patch bugs in the arch independent
> portions
> > (or worst, forget to patch them).
>
> Philipp's TI-CM3 branch is _the_ proof, that this can work. It is one of
> the (sub-)repositories I'd trust. If you think that an important patch
> is missing, you can offer it to him. But it's his decision to accept or
> reject it and your decision to follow him or create your own fork.
>

We would have n people tracking m repositories for each change and evaluate
if it is worth cherry picking. I don't see it working on the long run. At
some point the changes are unmergeable.


>
> Is freedom really so scary?
>
>
Freedom is in the license, not the tools.

The linux kernel has a huge number of different repositories for different
archs or whatever new stuff someone is working on, yet there is an official
tree.