Labs/Jetpack/Weekly Meeting/2011-09-06: Difference between revisions

From MozillaWiki
Jump to navigation Jump to search
(Removing redundant headers)
 
(4 intermediate revisions by 3 users not shown)
Line 1: Line 1:
= Agenda  =
= Agenda  =


* dcm: Usage update
*FlightDeck Updates  
*FlightDeck Updates  
*SDK 1.1 Status  
*SDK 1.1 Status  
Line 16: Line 17:
****This would speed up the request by narrowing the search, but I have to include all bugs because I can't filter the attachment/patch relationships in the initial request...  
****This would speed up the request by narrowing the search, but I have to include all bugs because I can't filter the attachment/patch relationships in the initial request...  
***Page script now uses HTML5 history API to create/modify history entries for every time you do a new request. (Back/forward should redraw the chart/tables from the stored state!)
***Page script now uses HTML5 history API to create/modify history entries for every time you do a new request. (Back/forward should redraw the chart/tables from the stored state!)
**Jeff: Workshops update
***On the schedule for All-hands, need to prioritize workshop content
***29/100 registrations for London!
***Waiting on CHM for final date for MV workshop, looks like the 12th or 17th of Oct.
***Hack Day event on Friday Sept 30 in London, looking for Judges. Anyone on the team interested in attending?


= Attendees =
= Attendees =
 
* mossop
* myk
* dcm
* gabor
* eddy
* brian
* jeff
* matteo
* dietrich
* alex
* irakli
* dbuc
* wes


= Minutes =
= Minutes =
== Usage Update ==
* dcm: as of August, 22% of AMO-submitted addons based on SDK
* July was really close to that as well
* on an upward curve, really great
* fligtar is going to run new queries soon
* #/% new authors, existing authors submitting new SDK-based addons
* # of SDK-based addons, # of SDK-based addons updated with or without repacks
* # of traditional addons that get updated via SDK
* not sure how long, fligtar has promised to start running numbers
* myk: all numbers i suggested?
* dcm: yes, plus one
* not sure of timeline, but he will do it
== FlightDeck Updates ==
* dbuc: we are pushing out another release this week
* another search facet going in
* a few more search additions
* 50% chance of getting in "push to AMO"
* if not this week, then next week for sure!
* getting some dev help with builder, pretty confident that we're going to feature freeze, work on stabilization, plan for 1.0 in early Q1 2012
* don't need new features, better to ship
* that's the plan, now that we have more resources
== SDK 1.1 Status ==
* myk: we spun 1.1rc1 last week
* folks nominated a few blockers, but nothing seems necessary
* one seemed to be a regression from 1.0, but after digging in, it turned out to be older
* still waiting on results of repack testing
* dbuc: we ran first test, had errata, AMO's queueing system was not processing repacks
* jeff had patch to solve 80% of failures, i.e. the queueing system ones
* once he puts that patch in it should be down to just SDK failures
* dbuc will take those on today or tomorrow
* jeff's fix might already have landed last thursday
* dbuc to check with jeff today and lace that up
* myk: can you report back to the discussion forum?
* dbuc to do so
== End Cycle Triage ==
* mossop: should drivers do more frequent triage at end of cycle?
* myk: we previously did daily triage at the end
* seems advisable, even if just five minutes to look at potential blockers
* we can leave other bugs for the regular weekly triage
* drivers to schedule this
== Bugs Update ==
=== {{Bug|682631}} ===
* wes: i'd like some feedback from myk on {{bug|682631}}
* myk: right now?
* wes: sometime this week
* myk: will endeavour to do so
=== 1.1-Targeted Bugs and Triage Process Change ===
* wes: when i checked last night, there are still 50 bugs on 1.1
* dcm: we had big conversation last week and proposed change to triage process
* a lot of these will fall into that if we decide to go that route
* mossop: anyone have concerns over proposal
* myk: new proposed process is to align with train schedule of releases
* drivers will continue to prioritize bugs but won't target most to milestone
* target milestone will be reserved for handful of important fixes
* developers will have time outside of targeted fixes to work on bugs at their discretion (but taking priority into account)
* applies to MoCo-employed developers; volunteers always have complete discretion to choose what to work on when
* general assent to new process
* irakli: it'll become clear how well it works as we go through process
* mossop: we'll go through bugs on thursday, make changes per proposal
* should clear out bugs assigned to 1.1 at that point
=== Low-Level Module Doc and E10S ===
* wes: there's a bug with patch that adds documentation to low-level module
* might e10s prototype change anything in there?
* eddy: i'm not really sure
* could you post this question somewhere in a bug so i can look at it?
* wes to do so
* wes: documentation is for jetpack runtime
* one of things it does is give information about process type
* was just wondering if anything in e10s prototype would make it untrue
* docs say process can be one of several processes
* ejpbruel: i've added new type, addon process
* dcm: post question to bug
* myk: and cc: eddy
== Roundtable ==
=== Bug Dashboard Update ===
* wes: you can now filter out bugs based on date filed or last modified
* put in email address and see only bugs that person is involved with (other than commenting and cc:ed)
* filed, assigned, QA contact, or patch attacher, requester, requestee
* now uses HTML 5 history API to create history entries when you do a search
* brian: can you put a spinner or ellipsis when getting bugs?
* wes: yes, and i'll do more error handling
=== Workshops Update ===
* jeff: signups going well, 1/3 of way there
* should have venue for one in SF soon with date
* looking like 12 or 17th of october in evening in computer history museum
* cheilman asked for volunteers for sept 30 hack day event
* night following london workshop
* i'll be in berlin by then, but if you're interested in participating, they've requested Mozilla people to show up, talk about what we're doing, help judge competition
* interested? email jeff
* overnight event: that evening until 7-8am the next morning
* myk: status of courses?
* jeff: giving three workshops
* sent out email to folks last week
* no replies; all good?
* brian: still working away on it, will have something put together
* brian: what format?
* jeff: three tracks, yours, mine, and XUL UI track
* three sessions
* in london will give them all three times, next week only once
* brian: but next week? (ed: at MoCo All Hands, where we intend to test the courses before the London event)
* jeff: no one knows
* dcm: we're still lacking details on rooms
* dietrich: highly doubt it'll be all six hundred
* brian: we have four-five parallel tracks, and we'll be one of those?
* jeff: hope we're running them in serial in the same room
* dcm: i'll send you amie tyrell's email address, you can send her what you want
* brian: in london, are we each lecturing to the whole group?
* jeff: in london, i'll talk to entire group for 20 minutes, then you'll have third of that group three times in a row
* brian: so i don't need warmup lecture?
* jeff: no
* jeff: brian, like your idea of warmup exercises, dietrich what do you think?
* dietrich: i like it
* dietrich: my only concern is being dependent on the builder
* might be prudent to have backup setup, so introduction about the SDK and setting it up
* brian: pointers to SDK download in introductory materials
* and point to the builder as well
* brian: what sort of internet connection will we get from venue?
* dcm: don't remember network problems at 500-600 person event at that venue previously
* jeff: and chm should have enough wifi
* but it's a conference, so who knows
* myk: dietrch, how is your presentation coming along?
* dietrich: ok, should have something by all-hands
* myk: jeff, how is your presentation coming along?
* jeff: good, not worried about it
* myk: everyone in contact with their assistants?
* brian: yes, anant sits next to me
* jeff: yes, i will be
* dietrich: alex, i'll be contacting you

Latest revision as of 23:48, 6 September 2011

Agenda

  • dcm: Usage update
  • FlightDeck Updates
  • SDK 1.1 Status
  • Bugs Update
    • Wes: I'd like some API feedback on the patch in Bug 682631 from myk. If the API gets locked in, I can try converting the SDK modules to use the new async read soon after it lands.
      • If anyone knows how to search the SDK for modules that currently use the sync read method, that'd be handy!
    • Wes: Lots of bugs still in the 1.1 milestone... What's the plan for these?
      • Should we get them moved this week before the All Hands eats our time?
    • Wes: Would ejpbruel know offhand if his e10s prototype work would affect the documentation being added in bug 661423? In what ways?
  • Roundtable
    • Wes: Bug dashboard update
      • It can now filter by filed/modified dates.
      • If you put an email address in the textbox, dashboard will only show bugs that have some relationship to that email address (filed by, assigned to, patch from, request to/from, etc) will be listed.
        • Still need to get the list to show which of those relationships applies to the specific bugs...
        • This would speed up the request by narrowing the search, but I have to include all bugs because I can't filter the attachment/patch relationships in the initial request...
      • Page script now uses HTML5 history API to create/modify history entries for every time you do a new request. (Back/forward should redraw the chart/tables from the stored state!)
    • Jeff: Workshops update
      • On the schedule for All-hands, need to prioritize workshop content
      • 29/100 registrations for London!
      • Waiting on CHM for final date for MV workshop, looks like the 12th or 17th of Oct.
      • Hack Day event on Friday Sept 30 in London, looking for Judges. Anyone on the team interested in attending?

Attendees

  • mossop
  • myk
  • dcm
  • gabor
  • eddy
  • brian
  • jeff
  • matteo
  • dietrich
  • alex
  • irakli
  • dbuc
  • wes

Minutes

Usage Update

  • dcm: as of August, 22% of AMO-submitted addons based on SDK
  • July was really close to that as well
  • on an upward curve, really great
  • fligtar is going to run new queries soon
  • #/% new authors, existing authors submitting new SDK-based addons
  • # of SDK-based addons, # of SDK-based addons updated with or without repacks
  • # of traditional addons that get updated via SDK
  • not sure how long, fligtar has promised to start running numbers
  • myk: all numbers i suggested?
  • dcm: yes, plus one
  • not sure of timeline, but he will do it

FlightDeck Updates

  • dbuc: we are pushing out another release this week
  • another search facet going in
  • a few more search additions
  • 50% chance of getting in "push to AMO"
  • if not this week, then next week for sure!
  • getting some dev help with builder, pretty confident that we're going to feature freeze, work on stabilization, plan for 1.0 in early Q1 2012
  • don't need new features, better to ship
  • that's the plan, now that we have more resources

SDK 1.1 Status

  • myk: we spun 1.1rc1 last week
  • folks nominated a few blockers, but nothing seems necessary
  • one seemed to be a regression from 1.0, but after digging in, it turned out to be older
  • still waiting on results of repack testing
  • dbuc: we ran first test, had errata, AMO's queueing system was not processing repacks
  • jeff had patch to solve 80% of failures, i.e. the queueing system ones
  • once he puts that patch in it should be down to just SDK failures
  • dbuc will take those on today or tomorrow
  • jeff's fix might already have landed last thursday
  • dbuc to check with jeff today and lace that up
  • myk: can you report back to the discussion forum?
  • dbuc to do so

End Cycle Triage

  • mossop: should drivers do more frequent triage at end of cycle?
  • myk: we previously did daily triage at the end
  • seems advisable, even if just five minutes to look at potential blockers
  • we can leave other bugs for the regular weekly triage
  • drivers to schedule this

Bugs Update

bug 682631

  • wes: i'd like some feedback from myk on bug 682631
  • myk: right now?
  • wes: sometime this week
  • myk: will endeavour to do so

1.1-Targeted Bugs and Triage Process Change

  • wes: when i checked last night, there are still 50 bugs on 1.1
  • dcm: we had big conversation last week and proposed change to triage process
  • a lot of these will fall into that if we decide to go that route
  • mossop: anyone have concerns over proposal
  • myk: new proposed process is to align with train schedule of releases
  • drivers will continue to prioritize bugs but won't target most to milestone
  • target milestone will be reserved for handful of important fixes
  • developers will have time outside of targeted fixes to work on bugs at their discretion (but taking priority into account)
  • applies to MoCo-employed developers; volunteers always have complete discretion to choose what to work on when
  • general assent to new process
  • irakli: it'll become clear how well it works as we go through process
  • mossop: we'll go through bugs on thursday, make changes per proposal
  • should clear out bugs assigned to 1.1 at that point

Low-Level Module Doc and E10S

  • wes: there's a bug with patch that adds documentation to low-level module
  • might e10s prototype change anything in there?
  • eddy: i'm not really sure
  • could you post this question somewhere in a bug so i can look at it?
  • wes to do so
  • wes: documentation is for jetpack runtime
  • one of things it does is give information about process type
  • was just wondering if anything in e10s prototype would make it untrue
  • docs say process can be one of several processes
  • ejpbruel: i've added new type, addon process
  • dcm: post question to bug
  • myk: and cc: eddy

Roundtable

Bug Dashboard Update

  • wes: you can now filter out bugs based on date filed or last modified
  • put in email address and see only bugs that person is involved with (other than commenting and cc:ed)
  • filed, assigned, QA contact, or patch attacher, requester, requestee
  • now uses HTML 5 history API to create history entries when you do a search
  • brian: can you put a spinner or ellipsis when getting bugs?
  • wes: yes, and i'll do more error handling

Workshops Update

  • jeff: signups going well, 1/3 of way there
  • should have venue for one in SF soon with date
  • looking like 12 or 17th of october in evening in computer history museum
  • cheilman asked for volunteers for sept 30 hack day event
  • night following london workshop
  • i'll be in berlin by then, but if you're interested in participating, they've requested Mozilla people to show up, talk about what we're doing, help judge competition
  • interested? email jeff
  • overnight event: that evening until 7-8am the next morning
  • myk: status of courses?
  • jeff: giving three workshops
  • sent out email to folks last week
  • no replies; all good?
  • brian: still working away on it, will have something put together
  • brian: what format?
  • jeff: three tracks, yours, mine, and XUL UI track
  • three sessions
  • in london will give them all three times, next week only once
  • brian: but next week? (ed: at MoCo All Hands, where we intend to test the courses before the London event)
  • jeff: no one knows
  • dcm: we're still lacking details on rooms
  • dietrich: highly doubt it'll be all six hundred
  • brian: we have four-five parallel tracks, and we'll be one of those?
  • jeff: hope we're running them in serial in the same room
  • dcm: i'll send you amie tyrell's email address, you can send her what you want
  • brian: in london, are we each lecturing to the whole group?
  • jeff: in london, i'll talk to entire group for 20 minutes, then you'll have third of that group three times in a row
  • brian: so i don't need warmup lecture?
  • jeff: no
  • jeff: brian, like your idea of warmup exercises, dietrich what do you think?
  • dietrich: i like it
  • dietrich: my only concern is being dependent on the builder
  • might be prudent to have backup setup, so introduction about the SDK and setting it up
  • brian: pointers to SDK download in introductory materials
  • and point to the builder as well
  • brian: what sort of internet connection will we get from venue?
  • dcm: don't remember network problems at 500-600 person event at that venue previously
  • jeff: and chm should have enough wifi
  • but it's a conference, so who knows
  • myk: dietrch, how is your presentation coming along?
  • dietrich: ok, should have something by all-hands
  • myk: jeff, how is your presentation coming along?
  • jeff: good, not worried about it
  • myk: everyone in contact with their assistants?
  • brian: yes, anant sits next to me
  • jeff: yes, i will be
  • dietrich: alex, i'll be contacting you