Dear Jonathan, I wholly support your proposal.
Many thanks, David On Fri, 22 Mar 2019 at 16:57, Jonathan Gregory <j.m.greg...@reading.ac.uk> wrote: > Dear all > > As you know, CF currently uses this email list for discussion of standard > name > proposals and all other matters except for proposals to change the > conventions. > > For the latter, we have been using trac for years, but have recently > completed > migration of that function to GitHub. The source of the CF conventions > document > is now in GitHub and version 1.7 was generated from it. In future, proposed > conventions changes will be discussed as GitHub issues on the conventions > repository, and the changes in wording will be finalised in pull requests. > This > is convenient because it provides continuity from the discussion to the > modified document. We are due to prepare and release version 1.8, freeze > trac > and publish guidelines for using GitHub in future for conventions changes. > > One of the reasons this has not yet been done is that it's important to > keep > everyone informed of conventions changes. For a while, postings to the > GitHub > conventions repository were sent to the subscribers to this email list, > but as > you remember there were complaints about the amount of traffic, and some > confusions were caused by people replying to GitHub by email, so the email > feed > was switched off while there was further discussion on the CF committee of > how > to proceed. > > Now we would like to present a new proposal, as follows: > > * To change to using GitHub for standard name proposals and all other > matters, > as issues in a different repository from the conventions (probably called > "discuss"). If we make this change, anyone who wishes to make a new > standard > name proposal or start a discussion thread about something else will need a > GitHub username, and will log on to GitHub and post it there as a new > GitHub > issue. This is not difficult (see below). It does not involve any of the > complexity of git for making updates to a repository as this will be a > place > solely to hold discussions and keep an archive of them. > > * To make sure that all existing subscribers to this list are kept in > touch, > postings to the GitHub issues in the discuss and conventions repositories > will > be distributed via email, initially to all subscribers to this email list. > Their subject lines will begin with [cf-metadata/discuss] and > [cf-metadata/conventions]. However, it will no longer be possible to reply > to > them on this email list. The archives will be kept and also copied to > GitHub. To comment on a standard name proposal or other discussion, there > will > be two options: > > * Either you can log onto GitHub and add to the issue in the discuss > repository, > > * Or you can *unsubscribe* from the broadcast emails (instructions will be > given for that), and instead *watch* the GitHub discuss repository. That > means > you will receive the updates to issues by email from GitHub, rather than > via an > email list, and you can then reply to GitHub by email. Your reply is > magically > posted on GitHub. That facility means that you can carry out a discussion > of an > issue entirely by email, once it's started, not much differently from an > email > list. > > We do not wish to lose members of the CF community who are subscribed to > this > list and remain interested in CF. The main reason for my writing this now > is > to ask whether there are subscribers who would "leave" us because it's off- > putting to use GitHub. Please say if that is what you think before 8th > April. > > Among the advantages of making this change are that > > * It would allow you to be selective, in that you could chose to be > notified > about conventions but not standard names, or vice-versa, on GitHub. Of > course, > we recommending watching both of them! > > * We won't have to maintain a CF email list, which is an administrative > complexity, especially as we currently have two different lists (for trac > and > email), which have to be kept consistent manually. Anyone can join GitHub > for > themselves and watch the CF repositories. > > Best wishes and thanks for your interest in CF > > Jonathan > > > To get a username on GitHub, go to https://github.com/ and sign up. This > is > free! Once you have a username, you can sign in. > > The issues on the CF conventions repository are at > https://github.com/cf-convention/cf-conventions/issues > This shows the list of changes to 1.7 which are being discussed. (NB the > name > of this repository will probably change, to improve consistency, and this > is > not the repository for the standard name discussions, which we have not yet > created.) If you wanted to start a new issue, which is like a new email > thread > e.g. for a standard name proposal, you would click the "New issue" button, > and > get from there to a web form where you type plain text, and then "Submit" > it. > > To comment on an issue (like replying to an email thread) you click on the > issue, go the bottom of the page, write in the box, and press "Comment". > > To receive emails containing comments on any of the issues contributed by > others, you can "watch" the repository, by clicking on the drop-down menu > labelled "Watch" at the top of the page, with an icon like an eye. When you > receive such an email, you can reply to it and your reply will be posted > to the > issue. (It tells you at the bottom of the email that you can do this, and > gives > you the URL of the comment that's been sent in the the email in case you > want > to do it on GitHub.) > _______________________________________________ > CF-metadata mailing list > CF-metadata@cgd.ucar.edu > http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata > -- David Hassell National Centre for Atmospheric Science Department of Meteorology, University of Reading, Earley Gate, PO Box 243, Reading RG6 6BB Tel: +44 118 3785183 http://www.met.reading.ac.uk/
_______________________________________________ CF-metadata mailing list CF-metadata@cgd.ucar.edu http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata
