Hi, On Sunday 06 September 2015 00:54:30 [email protected] wrote: > Hi, > > There are some support requests, and regular spam-cleaning to do. > > Then there's the SVN repo for project "admin" with documentation and > fully automated Puppet rules. There's also a no-lesser-automated > Vagrant environment for development/deployment from scratch.
I ll start by looking at this. > System is up-to-date and can work stably as-is for the next 2-3 years. > > > (Also, what is your opinion about how much gna is relevant these days? I > > must admit often using github, it is kind of pratical. > > Wrt github, do I need to point to: > http://fsfe.org/news/2001/article2001-10-20-01.en.html > and > http://about.gna.org/ > ? :) I was talking from a purely technical standpoint. We all agree that the availability of a libre platform matters. Savane is aging and since CERN stopped using it (it was used in order to coordinate software development about LHC being built - LHC is up since almost ten years now), I m not sure that are still many users using its most advanced part, while they most basic one are... well... aging. I did not follow through development that have been made or was planned to be made since a few years. I ve read a few times about rewrites. How did that go? I guess the notion of rewriting something is always tempting but that is hard work and most of the time it ends to be regression, while the time spent could have been used to make actual progress. My point is, since no one probably has time or plan (but I may be wrong, correct me) to make a business activity to work on this software, there is little point in writing one that will always lag. Or using one that will always do. There is nothing wrong using a software that get old, except we know that PHP crap is pain to maintain, to keep secure, and every new version is likely to break stuff. So I was wondering what s the feeling about it. Checking forge from https://en.wikipedia.org/wiki/Forge_%28software%29 - redmine: seems ok but no so far from what we already have - launchpad: history of licensing issue - kallithea: git and mercurial, no issue tracking, but otherwise interesting - gitlab: interesting software but the notion of Community/Enterprise = Free beer/proprietary makes it irrelevant - gforge: irrelevant - fusionforge: ditto - apache allura: nice features but still the same thing If I had to imagine what I thought would be practical, it would be a minimalistic platform that would work with a mail interface or an ssh wrapper interface. Similar in the essence to debian BTS, but I guess with authentication through GPG signature. I dont think email will stop being relevant in the next 10 years. The design of web interface is something else. Also, we kill the entirely annoying notion of dealing with password on our side: people would use their own GPG or SSH keys to authentify themselves (add SSH key by GPG signed mail / add GPG public key over SSH). It puts down the web interface to a few generated static pages and cut down many useless feature. What do you think about such concept? > > > - You setup the project@ mailing list, I don't remember changing the > > > conf. > > > > Then I misconfigured it. That should be changed. > > Maybe, provided there's timely moderation. Moderation of non list members. > > Let's forget this and join for a beer or a glass of wine. > > This is a nice idea, sure, though we're all hundreds of kilometers > away from each others and with limited time. Are you far from Paris? > Would you could come to > the FOSSa event this month, that's the next Free Software event? Depends on the date and location. _______________________________________________ Project mailing list [email protected] https://mail.gna.org/listinfo/project
