curious: why use own hosting rather than github?

J Rt jean.rblt at gmail.com
Fri May 22 11:14:22 CEST 2020


Yes, I agree with you. For me it looks like the 'no github' people
have the following main arguments:

- github is company owned -> true, and your point about self-hosted
gitlab may be a very good answer. The other thing is, even if it is a
company, everybody still have their local copies of the code, and
anyways, the value of github is probably far too tightly related to
its relative openness / open -source friendly attitude for it to be
thrown away at least in the foreseeable future.

- we do not want to change the workflow to make it easy to a category
of users, as there will always be another category asking for
something else -> this is quite true. But at the same time, you want
to choose the solution that makes most people (or a
'value-for-the-project-weighted' sum of the people) most happy. Here I
think for the meteorite / casual users github gives most value, but it
is a question to know how valuable these people actually are.

On Fri, May 22, 2020 at 11:06 AM Ondřej Synáček
<ondrejsynacek at fastmail.com> wrote:
>
> Yes it is interesting conversation indeed. The point made about
> non-technical users not able to report bugs is a good one, although I
> still think that can be solved by email as well (just a regular form and
> the contents could get posted to this mailing list or different one).
>
> If people here are against proprietary Github, why not just use
> self-hosted Gitlab solution? Personally I prefer Gitlab over GH but both
> are good.
>
> On 22 May 2020, at 11:03, J Rt wrote:
>
> > I agree with you, and I suppose most of this discussion is becoming an
> > interesting pros and cons weighting of different approaches :) I
> > definitely think that everything is a question of tradeoffs, and the
> > point made by 'pro github' participants here is that it is quite
> > likely that github is de facto the dominant platform / way /
> > methodology used in the open source world, with most users familiar to
> > it, and that, therefore, having a github workflow may be the best way
> > to engage a large(r) community. But I agree that it also comes with
> > its downsides, to be weighted against the benefits it could bring.
> >
> > On Fri, May 22, 2020 at 10:57 AM Erazem Kokot <contact at erazem.eu>
> > wrote:
> >>
> >>> This is true, but at the same time, this is 'yet another method to
> >>> learn', while nowadays a vast group of users are quite proficient
> >>> and
> >>> used to github and similar. Using something a la github would take
> >>> away some entry barrier for most user IMO.
> >>
> >> Although I understand what you mean, I don't think just because new
> >> users want the contributing process to be similar to what they
> >> already
> >> know, that projects would have to change their workflow to suit such
> >> users. This is not a great way to learn for the user and pretty
> >> pointless for the project, since if for example using Sourcehut or
> >> Gitlab, users could still use the same argument of comfort with
> >> Github
> >> to try and move the project to Github.
> >>
> >> Projects shouldn't be forced to change their workflow to suit a small
> >> minority of the contributors or possible future contributors.
> >> If you were maintaining a project on Github, you wouldn't want users
> >> submitting pull requests over email, so why would it be any better
> >> the
> >> other way around.


More information about the Password-Store mailing list