...
- 28 Oct 2014
Attendees
Robyn Allsman, Jacek Becla, David Ciardi, Gregory Dubois-Felsmann, Frossie Economou, Kian-Tat Lim, Donald Petravick, Ron Lambert, Robert Lupton, Simon Krughoff, Xiuqin Wu
Honored guests: Jeff Kantor, Mario Juric
Attendees, you will be considered present in this virtual meeting if you enter your vote in the table below. If you have no opinion on the change, enter "present" or "here".
Preamble
The SAT met on 2014-10-21 and decided to recommend the following proposal:
- "Open up LSST software mailing lists"Jira server JIRA serverId 9da94fb6-5771-303d-a785-1b6c5ab0f2d2 key DM-9
SAT Proposal
The proposal for the suite of public lists would be:
- data-announce@lsst.org: a mailing list for public (e.g., release) announcements
- data-users@lsst.org: the list for users of DM software (it was felt that eschewing a mailing list for SO or CQ may be too radical)
- data-devel@lsst.org: general development-related discussions (there will always be a need to reach all developers; e.g., the recent discussions on github-JIRA integration/notifications would go here)
An alternative prefix may be 'dm-', if there's pushback that 'data-' is too broad. Another option (that just occurred to me and isn't necessarily supported by either Frossie or Josh) is 'code-'. All these lists would be public, archived, and open to anyone to subscribe.
In addition to these, we would let any of the working groups (or subgroups) open mailing lists as necessary. So, for example, we could have algo-devel@lsst.org (for apps), or even diffim-devel@lsst.org (if algo-devel becomes unwieldy and needs further splitting), etc. We'll rely on common sense to not let the number of lists grow to >= number of people in DM. These lists could be private, if necessary, but the default should be to make them public.
Finally there'd be a (private) dm-staff@lsstcorp.org, for broadcasts to DM staff only.
Transition plan:
We'd ask Iain to create these lists.
Once they are created, the proposal is to retire and archive lsst-data and lsst-dm-stack-users lists after some grace period (~1 week).
The subscribers to lsst-data would be alerted to re-subscribe to either the -announce, -users, or -devel lists. The subscribers to lsst-dm-stack-users will be alerted to the name change, re-subscribed to data-users (or, if Mailman allows for that, lsst-dm-stack-users would just be renamed to data-users), and an alias will be set up to forward e-mails from lsst-dm-stack-users to data-users.
SAT Recommendation
The SAT recommended at its meeting on 2014-09-21 that Mario's proposal for data-announce, data-users, and data-devel be adopted. In addition, the SAT recommends that smaller groups be encouraged to set up their own mailing lists, ideally via self-service at lsst.org.
Goals
The TCT will now:
- first, review and comment on these proposed changes based on their impacts upon cost and and schedule; and
- secondly, present our findings to the DM Project Manager and the DM Project Scientist who consider them when determining their mandate on these three changes.
Analysis
Robyn's opinion on the impact upon cost and schedule of accepting each of these changes to the mail lists is that there would be minimal impact.
Discussion items
Who | Comment | Yes/No/Here | |
|---|---|---|---|
| Robyn Allsman | Yes. | ||
| Jacek Becla | |||
| David Ciardi | In principle, I have no objection, but I am worried a bit about the sheer number of emails and different addresses. | Yes | |
| Frossie Economou | |||
| Kian-Tat Lim | Are there any LDM documents that would need changes? | Yes | |
| Donald Petravick | |||
| Ron Lambert | |||
| Robert Lupton | I prefer the dm- prefix (we're not discussing data). I don't see the problem with having more lists (more emails, yes!) | Yes | |
| Simon Krughoff | +1 for the dm- prefix. | Yes | |
| Xiuqin Wu | I prefer dm- prefix too. | Yes | |
| Tally of votes: | Quorum is 6 members present |
Action items
- At the end of voting, Robyn Allsman will notify Jeff Kantor and Mario Juric of the TCT vote. The TCT requests that Jeff and/or Mario advise the TCT as to the final disposition of this issue so that the Standards documents can be modified if, and as, appropriate.