| gary_poster | bac benji (are you really here?) danilos gmb: kanban now-ish, skype almost now-ish | 13:29 |
|---|---|---|
| danilos | quack-quack | 13:29 |
| gmb | Baa. | 13:30 |
| gmb | gary_poster had a farm, ee | 13:30 |
| gmb | 13:30 | |
| gmb | i ee i o. | 13:30 |
| gary_poster | lol | 13:30 |
| danilos | :) | 13:30 |
| gary_poster | https://dev.launchpad.net/LEP/BugLinking | 13:36 |
| gary_poster | https://bugs.edge.launchpad.net/launchpad/+bug/95419 | 13:36 |
| _mup_ | Bug #95419: Record dependencies between bugs <lp-bugs> <oem-services> <Launchpad itself:Triaged> < https://launchpad.net/bugs/95419 > | 13:36 |
| danilos | gmb, gary_poster: I assume bug 674815 is "fix released" (Rob added it to our bug "view")? | 13:38 |
| _mup_ | Bug #674815: When a bug is marked as duplicate, you are invited to subscribe to the master bug <lp-bugs> <story-better-bug-notification> <Launchpad itself:Triaged> < https://launchpad.net/bugs/674815 > | 13:38 |
| gary_poster | benji, please make sure you keep me/us apprised as to what you are working on so we can take over if the germs win today's battle | 13:49 |
| benji | gary_poster: sure; at the moment I'm fixing a bug I found in my 777786 branch, but once that's in the rube goldberg machine of landing I'll start another one. | 13:50 |
| gary_poster | cool | 13:50 |
| benji | if there's one in particular you like, let me know | 13:50 |
| gary_poster | bug 784575 is the new one | 13:53 |
| _mup_ | Bug #784575: Change bug notification email to point to new bug +subscriptions page to manage subscriptions <story-better-bug-notification> <Launchpad itself:Triaged by yellow> < https://launchpad.net/bugs/784575 > | 13:53 |
| gary_poster | benji, you could prepare a branch for that, even though we could land that; or just pick one. bug 777789 and bug 781987 are the other good candidates that I see. | 13:55 |
| _mup_ | Bug #777789: "Other subscriptions" description of direct team subscription makes no sense <story-better-bug-notification> <Launchpad itself:Triaged> < https://launchpad.net/bugs/777789 > | 13:55 |
| _mup_ | Bug #781987: when a bug has no subscribers the list is shown as 'None' <story-better-bug-notification> <Launchpad itself:Triaged> < https://launchpad.net/bugs/781987 > | 13:55 |
| benji | sounds good | 13:56 |
| gary_poster | "even though we could land that" -> "even though we would land that at a later date" | 13:56 |
| gary_poster | bac benji danilos gmb, I've mentioned this to some of you in our pdr calls, but I was not consistent about it. I thought of another goal that could be shared across the squad: we deliver value consistently within two week increments. We will be evaluated by this collectively (as a squad), so it might be reasonable to call it out in our individual goals as well. I won't require it, but it's a good idea, I think. | 14:18 |
| benji | I like it. | 14:18 |
| gary_poster | cool | 14:18 |
| gmb | So do I. | 14:18 |
| gary_poster | très bien | 14:19 |
| gary_poster | benji, extra points for a branch that removes the feature flags at the same time :-P | 16:28 |
| benji | will I get a 1up too? I could use an extra life right about now. | 16:29 |
| gary_poster | heh, not in my power to grant :-) | 16:29 |
| gary_poster | extra points: ...I think.... Eagerly getting rid of the feature flags does make me a bit nervous, but AIUI that's what we are asked to do | 16:30 |
| benji | well, in this case we pretty much have to; the email sending can't be feature flagged (as far as I know) so everyone's emails will be sending them to a place that they might not be able to see unless we open up the feature flag to everyone, which is equivelent to just removing them | 16:31 |
| benji | ("them" == "feature flags", not the people) | 16:32 |
| gary_poster | yeah makes sense | 16:57 |
| gmb | Wouldn't it be safer to just open the feature flags up to everyone, land the branch that adds the link to the emails and then wait a week or so to remove the flags from the code? | 16:59 |
| gmb | I'm thinking about the cost of rolling back if we need to | 16:59 |
| gmb | (Though I guess with bzr it's pretty much moot) | 16:59 |
| gmb | (Aside from LOSA time) | 16:59 |
| gmb | benji, gary_poster ^^ | 17:00 |
| gmb | (Which I say with no context or knowledge of anything you may have said about this recently) | 17:00 |
| gary_poster | gmb, I think this is the logic: | 17:00 |
| gary_poster | if we send emails with links to the page, then the feature flag better be open to everyone or gone | 17:01 |
| gary_poster | links are not feature flagged | 17:01 |
| gary_poster | the email links I mean | 17:01 |
| gary_poster | So here's a bad story | 17:01 |
| gary_poster | followed by a good story | 17:01 |
| gary_poster | The Bad Story | 17:02 |
| gary_poster | 1) we open up feature flags for everyone | 17:02 |
| gary_poster | 2) we land a branch that adds a link to the email | 17:02 |
| gary_poster | 3) oh noes! Bad Things Happen! | 17:02 |
| gary_poster | 4) we turn off the feature flag for everyone (but don't revert the branch) | 17:03 |
| gary_poster | Result: emails are being sent out to a page that doesn't work | 17:03 |
| gary_poster | that is, that is feature flagged | 17:03 |
| gary_poster | The Slightly Better Story (known here as The Good Story, In Comparison) | 17:03 |
| gary_poster | 1) we open up feature flags for everyone | 17:04 |
| gary_poster | 2) we land a branch that adds a link to the email | 17:04 |
| gary_poster | 3) oh noes, etc. | 17:04 |
| gary_poster | (oh, sorry, #2 was supposed to also remove feature flags) | 17:05 |
| gary_poster | 4) We revert branch for #2 | 17:05 |
| gary_poster | Result: emails are sent with the old, working link | 17:05 |
| gary_poster | Admittedly, it is somewhat slower | 17:05 |
| gary_poster | but the end result is better | 17:05 |
| gary_poster | We can wait a bit after step 1 (in either story) to see if we can prevent #3 from happening, of course | 17:06 |
| gary_poster | The othe advantage of The Slightly Better Story is that, if #3 never happens, we have done what we are supposed to do (remove the feature flags) | 17:06 |
| gary_poster | So anyway, I suppose one could argue either way, but I'm happy with going with the second approach | 17:07 |
| gmb | gary_poster: Okay. I'm officially ±0 as far as one way of doing it over the other. Like I said, bzr reduces the cost, except for LOSA time. | 17:07 |
| gary_poster | gmb, fair enough, and agreed | 17:07 |
| * gary_poster going for a lunch/walk, since my coding project is making me grumpy :-P | 17:09 | |
| gmb | :) | 17:09 |
| bac | hi gary_poster | 19:34 |
| bac | gary_poster: you post-grumpy? | 19:34 |
| gary_poster | :-) | 19:34 |
| gary_poster | mostly but on call | 19:35 |
| bac | np | 19:35 |
| gary_poster | bac, wassup? | 19:53 |
| bac | gary_poster: not much. was just going to run a text change by you before i submitted my branch. too late now. :) | 19:53 |
| bac | idle chit chat mainly | 19:53 |
| gary_poster | heh, ok :-) | 19:53 |
| bac | s/None/No subscribers./ | 19:54 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!