On Mon, Mar 10, 2014 at 2:34 PM, Stephen Connolly
<[email protected]> wrote:
>
> Take care to note that in my original vision I did not assume Type 2 was
> bound to Jenkins, only bound to *a* CI system... Then I said how literate
> was my vision of solving the problem *for* Jenkins.
But conceptually, a CI system should just integrate and automate the
stuff that would already work as separate steps that you could run
without it. Or you are introducing magic instead of integration.
> This falls back to my philosophy for tools (which you may know if you follow
> the maven lists) namely, don't force the user to stick with your tool if you
> can help it. Let them try other tools and stick with you *because* you are
> the best, not because they've been locked in.
>
> That is how I see tools being successful and staying so... If I am
> successful with my vision, maven 4 will lure you back in and reward you for
> experimenting with it (even if you don't stay... Though we'd prefer if you
> do ;-) )
Experimenting is easier and more likely to go forward if it doesn't
require infrastructure changes.
>> Is there a standalone literate-builder that the developer can
>> run/test/debug on his own without jenkins support?
>
>
> Yes but very very bare bones (literate-cli) and I am not sure how far mic
> got with it!
I'd expect much faster acceptance with a full standalone version that
did exactly the same things except from a checked out local workspace.
I wouldn't expect people to write make/cmake/project(etc.) files
without the ability to run/test them locally before committing and
don't see much difference here (except maybe automatic branch tracking
for other versions).
--
Les Mikesell
[email protected]
--
You received this message because you are subscribed to the Google Groups
"Jenkins Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
