On 07/28/2015 05:08 AM, Andrei Alexandrescu wrote: > So how's this working? Simply reply to this email? -- Andrei
Yes, unless you come up with a better idea. Let me start again. > On 7/25/15 8:33 AM, Martin Nowak via dmd-internals wrote: >> Agenda: >> >> - review last 2 weeks We finally got some traction towards 2.068.0 and fixed most outstanding blockers, thanks a lot Kenji and Walter. Unfortunately we only got a lot of new bug reports with the second beta. I guess it'll take a while to get used to more regular releases, but it's a bit annoying to get another week or so delay b/c of little testing of the first beta. Trying to squeeze Unique/RefCounted into 2.068.0 was a bad idea. I spent several days on that (not working on the release), but the feature turned out much more complex b/c of attribute hell. https://trello.com/c/pTlDuyBD/16-finish-smartref-implementation Let's please resist any temptation to add features in the last minute. It'll only cause delays and we'll end up with something that wasn't quite ready. If something should make it into 2.069.0 you'll have to start now to make that happen, not when the first beta get's released. From now on I'll regularly announce/remind of the release schedules in the newsgroup so that nobody gets surprised by a release. >> - open issues blocking release Way too many, please help to test the beta and file/reduce/fix bugs. http://d.puremagic.com/issues/buglist.cgi?query_format=advanced&bug_severity=regression&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED If you work on a bug, please assign it to yourself on Bugzilla or Trello (https://trello.com/b/XoFjxiqG/active) so that we avoid double efforts. >> - work on changelog As nobody takes responsibility for the changelog we need a mechanical solution IMO. I propose to create a CHANGELOG.md in the dmd/druntime/phobos repos and reject any PR that comes without an adequate explanation for D end users. This would also be an opportunity to improve the current changelog page (e.g. one page per release). >> - discuss ddmd plans and 2H15 update There've been some fruitful plans for ddmd. But I think we should not loose track of other priorities. A self-hosted compiler might help with dmd development and provide learning insights for the compiler devs, but this work doesn't improve D as a language or ecosystem in the short- to mid-term. Looking at http://wiki.dlang.org/Vision/2015H1 it seems we're on track, but then again we only made little progress on each point. It might be necessary to reduce the number of priorities and derive concrete stories to achieve them. This is even more mid-term planning than vision. >> - eventually plan other work -Martin
signature.asc
Description: OpenPGP digital signature
_______________________________________________ dmd-internals mailing list [email protected] http://lists.puremagic.com/mailman/listinfo/dmd-internals
