On Wed, 10 Sep 2014, Nate Finch wrote: > How much do we care about charm development on Windows (both windows charms > and development on windows machines)? > > The reason I ask is that Horacio is starting on the charm-sync task from > the pain points spreadsheet (syncing the charm files local <-> unit during > charm development & debugging). > > Obviously if we can just use scp, it's fairly trivial, but Windows servers > don't ship with an SSH server enabled by default and Windows clients don't > have scp. Go has a robust SSH server > <https://godoc.org/code.google.com/p/go.crypto/ssh> that we could easily > run on a Windows machine, but obviously that's more work than just using > the built-in stuff on Linux. > > So, how much time and effort do we want to spend to support Windows? > > My ideal would be feature parity on Windows. I know in my years as a > Windows developer, I always felt really annoyed that Windows was constantly > a second class citizen. And I know there are a *ton *of enterprise > customers that only deal with Windows, who would love to be able to deploy > their stuff with the ease Juju provides.
I know that I wish we had our initial juju-quickstart implementation back a little bit. We went down a path that made it easy for linux users, and then we were able to add OSX support pretty easily. However, to do support on widows it requires a much different take and significant work down a different implementation. I'm all for iteration and solving the needs for linux users sooner (we have them annoyed today) but I definitely think with us pushing windows support and such that whatever path moves forward needs to know how the windows solution will work ahead of time. I guess I'd love to see others learn from the juju-quickstart experience here. -- Rick Harding Juju UI Engineering https://launchpad.net/~rharding @mitechie -- Juju-dev mailing list Juju-dev@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/juju-dev
