minutes of ESC call ...

* Present
  + Andras, Joel, Bjoern, Stephan, David, Astron, Thorsten,
    Michael M, Caolan, Eike, Cedric, Kendy, Petr

* Completed Action Items
  + poke website guys wrt. feedback menu-item / page (Michael)
  + switch ESC time to CET to avoid changes (Michael)
        + Bugzilla attachments not set to autodetect (Tollef)
        + instead have the global headers in one central global directory
                + check behaviour of git-cherry-pick (Miklos)
        + SLES11 / temporary PTF for the server ? using SP3 toolchain ? (Kendy)
    + Markus worked around it on his own machine
  + find a slot in the schedule for 3.6.7 (Petr)
  + poke UX / users list suggesting removal of inherited number formats (Markus)

* Pending Action Items
  + Icons for Lionel + template manager (Astron/UX)
    + Issa produced something - pending Lionel's input
        + disable Rhino / Beanshell unless in experimental mode (Michael M)
        + look at gradient / clipart issues (Michael)
        + helping out with code-pointers for UI bugs (Kendy)
        + need design for copying styles between templates (Astron/UX)
                + either in that dialog or a new dialog
                + also issue with only editing templates that are in the mgr
        + decided: change the default install directory to
          'LibreOffice 4' on master (Pmladek)
        + instead have the global headers in one central global directory
                + write up for the mailing list (Bjoern)
      + [ half way through ]
  + Personas - update / de-couple built-in URL (Kendy)
    + legacy link will continue to work for a while.
        + buy Windows build hardware / hosting (Norbert)
    + still investigating options
  + update mac SDK configure check on master (Norbert)
  + change -4-0 calc default to not use cached values (Eike, Kohei, Markus)
    [ Markus plans to submit on Monday ]

* Release Engineering update (Petr)
        + 4.0.2 retrospective
    + it's out, few complex fixes left waiting for rc2
    + otherwise seems to have gone well
        + 3.6.6 rc2 update
    + tagged yesterday, builds for linux available,
      Windows still building.
    + looks promising.
  + 4.0.3 rc1 - ~two weeks out April 15th
  + 3.6.7 date: proposed for July 17
  + 4.1 deadlines
    + feature-freeze May 20th, Alpha1 two weeks before

* WikiHelp for 4.0 (Andras)
  + what's the status of it, lots of complaints it's not on-line
  + help appreciated (Kendy)
AI: + get wikihelp credentials to Andras so he can do it (Kendy)

* UX input (Astron)
  + still pending a style copying design
  + looking at a design-bugs initiative: easy-hacks focused on design
  + would demand for rotation in writer be satisfied by
    rotation of 90 degree multiples ?
    + a very very common case; already have mirroring

* Removing C++ examples from the SDK (Bjoern)
  + tried to get the C++ examples to work, tried on Windows
    + a real pain to use that on Windows
    + someone approaching the project from far away
  + problem: people may try extensions, find it hard to use and leave
    + whereas it is far easier to hack on the core itself
  + where was the trouble, compiling SDK, or extensions ? (Stephan)
    + example, compiling the C++ - problem
    + setup of the SDK needed regardless of language (Stephan)
      + needs attention & love
      + from there a minor step to having C++ working
    + is it a gbuild related regression (Michael?)
      + SDK uses it's own build-system: dmake
        tweaked to work with gnumake, still build.pl
        based
      + worked for Stephan on Windows recently fine.
  + tried a native visual-studio environment, but failed (Bjoern)
    + immediate crash on run; as well.
    + build system we have is not ideal, building your
      own doesn't work so well either.
  + python does not require an SDK at all, no IDLs etc. (Bjoern)
    + getting started is easier
  + in practise it is always easier to go to the core; Java or C++ (Thorsten)
  + making things easier for building extensions:
    + provide a way to build extensions in our core build ?
    + don't want to bundle extensions though (Stephan)
  + easier to build the whole of LibreOffice instead of with the SDK
    + could we not make it easy to build extensions inside
      the main build ?
    + with gnumake deps - don't need to build all of it
  + we need someone to care for and love the SDK (Stephan)
    + native platform builds
    + the build-system is not set in stone -
      it needs to change.
    + eg. pre-compiling the headers
    + external makes it harder to use internal impl. details.
  + liblibreoffice might help (Michael)

* GSOC update (Cedric)
  + keep improving the Ideas page, update on Monday

* An understanding on adding new committers (Thorsten)
  + typical practise was to give a small domain of code
    to commit to, asking for non-trivial commits on the
    list etc.
  + would like a feed for new people with commit rights (Stephan)
  + can we make those lists public (Bjoern)
  + mail the ESC when people are added, for people where we
    have any doubts mail first.
AI: + mail list of committers to ESC (Bjoern)

* gerrit auto-building - how to trigger that etc. (Thorsten)
  + Mac is being setup, still hunting for a nice/cheap
    win service.
  + be interested in how to trigger that (Michael)

* Python unit tests (Bjoern)
  + although hard to debug, never have to debug most of them
    + rare occurences they fail.
  + story sprang out of Dresden sprint (Thorsten)
    + 2x volunteers wanted to help write unit tests,
      but only really comfortable in Python.
    + David hacked something nice up.
    + concerned wrt. level of push-back
  + the purpose of unit tests is to break & then to show you
    something useful (Michael)
  + we can only gain from knowing something broke (Bjoern)
    + but tests that fail for unclear reasons are
      worse than useless, distracting, time consuming (Stephan)
    + the Java tests are exactly like this, code written
      by a non-core developer, problem with the test => very
      hard to debug.
    + the effect is that you disable the old/ugly tests
      they randomly fail, so just disable them.
    + concern that tests from people only capable of writing
      tests in python might create a similar problem.
  + Java tests written at Sun: two categories (Bjoern)
    + UNO API tests - awful - can't debug, synthetic
      synarios all over the place.
    + specific tests - much more useful
  + concern wrt. un-helpful push-back to new changes (Thorsten)
  + should get it into master
    + however - introducing a test framework means:
      + helping people debug the problems involved
      + supporting and maintaining your unit tests
    + for now we should treat python tests as subsequenttests
      alongside the Java ones with similar debugging issues.
    + and evolve that as we see how they pan out.
    + thank-you to David O for stimulating the discussion
      & doing the work

* results/status of automated load testing (Markus)
  + commitments from Kendy, Eike to fix some of the few
    bugs found
AI: => publish the results as/when (Markus)

* QA update (Joel)
        + 108 unconfirmed vs. 4.0, help appreciated with triage.
                http://tinyurl.com/a3csc2o (cf. topic in #libreoffice-qa)
  + over 1000 bugs reported in February
    + over 6000 comments in a month
    + fighting to keep up with bug reports.
  + trying to work out what to do about extensions:
    + they're not our bugs, we don't control them
    + look under 'extensions and filters' component.
    + come up with a solution for users.

* Open 4.1 MAB / regressions
        + 4 (of 10) older 5/10 5/10 4/8 4/8 2/5 1/2 3/3
        + https://bugs.freedesktop.org/showdependencytree.cgi?id=60270&hide_resolved=1

* Open 4.0 MAB / regressions
        + 18 (of 112) older 17/111 19/109 19/106 17/105 14/99 15/96 16/94 16/91 13/79
             16% 15% 18% 18% 16% 14% 16% 17% 18% 22%
        + https://bugs.freedesktop.org/showdependencytree.cgi?id=54157&hide_resolved=1

* 3.6 most annoying bugs ...
        + 73 (of 218) older 74/216 78/217 77/215 77/214 75/211 76/210 57/189 44/175
             33% 34% 36% 36% 36% 36% 36% 30% 25%
        + https://bugs.freedesktop.org/showdependencytree.cgi?id=44446&hide_resolved=1

* Bibisected bugs open: whiteboard 'bibsected'
        + 24 (of 83) older 27/83 25/78 29/78 32/79 27/72
        + re-built the query to avoid pre bibisected mess
                http://bit.ly/VQfF3Q

* all bugs tagged with 'regression'
        + 291(+1) bugs open of 1414(+13)total

        * ~Component count net *
        + Writer - 96 (-3)
        + Presentation - 29 (+0)
        + Spreadsheet - 28 (+1)
        + LibreOffice - 28 (+1)
        + Database - 24 (+1)
        + Drawing - 20 (+0)
        + Crashes - 16 (-2)
        + Borders - 15 (+0)
        + Migration - 3 (-2)
        + Basic - 2 (+0)

        + https://bugs.freedesktop.org/buglist.cgi?keywords=regression%2C%20&keywords_type=allwords&resolution=---&query_format=advanced&product=LibreOffice&list_id=36764
        + Migration: https://bugs.freedesktop.org/showdependencytree.cgi?id=43489&hide_resolved=1