On 24 Nov 2014, at 9:35, Mickael Istria wrote:
On 11/24/2014 09:16 AM, Max Rydahl Andersen wrote:
> If it is not available in Mars we can't really depend on it without
> we start bundling it and then we own the maintaence of It risking
> others will bundle other versions and we'll be incompatible. It is
> bad enough there now are multiple terminals, but multiple ways to
> define and connect to remote hosts for file transferring. That would
> be a mess.
Why not simply contributing to the project in case we have things to
I have not said we should not contribute to it. That is probably what we
need to do - the current situation is not stable.
I don't understand why being in Mars/not being in Mars makes a
difference about how to contribute to this project and how much we can
rely on it.
If it is not in Mars we can't rely on it being in users base eclipse
It's just that current contibutors are resigning from this
doesn't require anyone to fork it.
If noone is putting it into Mars someone will either rely on what is in
Eclipse Kepler OR fork/use something else to define remote FTP/SCP
connections in their mars release.
meaning duplicate setup and whatnot.
> So sure, there are hacks for bundling this in. None of them are
> sustainable if the project is looking for termination.
If we don't want project to terminate, and if multiple people need it,
then it's up to its adopters (including us) to affirm that we are
ready to make the necessary efforts to provide and review patches to
keep the project somehow alive. Given the maturity of the project, it
shouldn't represent too much work to keep it alive.
Things will be a mess only if we let them be a mess, but just not
changing anything to current state of the project would be enough to
avoid that mess.
However, if there is a better replacement, it may be a good time to
use the better one; but if there is not, then we should IMO step up
and say that we don't want project to terminate.
Yes, that is why i'm on tm-dev and cross-project-issues now to figure
out what is going on.
In short: this is a big deal since this was *not* something we planned
to take upon us. and if we don't do it we still need to rework how we
do our RSE integration/build. Anyone volunteering to help out ?
> why are we using their dedicated repo and not the release train
> do for other things ? To get source for Dev to or?
Yes, I think so, on Rob's request.
As long as we are actually using the release train version and not
"latest/greatest" build from RSE that should be fine.