For minutes of previous meetings, please see FoundationsTeam.




  • Any business from activity reports
  • UDS Karmic
  • Good news
    • Mike Terry (mterry), from Canonical's OEM team, will work with our team for the Karmic Cycle
    • Muharem Hrnjadovic (al-maisan), from Canonical's Launchpad team, will work with our team for 3months
  • AOB

Outstanding actions from last meeting

Actions from this meeting

Activity reports

Colin Watson


  • Fixed rootskel-gtk build failure.
  • Documented debian-installer move-to-updates procedure on ArchiveAdministration.

  • Merged: pkgsel, zlib, debian-installer-utils
  • Synced: nautilus-share
  • Spent most of Thursday fixing findutils #373214 (find -prune fails to request stat information).
  • Various updates (installer, seeds, livecd-rootfs) for new ia64 kernel names.
  • Got oem-config and ubiquity built for karmic.
  • Looked into new dpkg-vendor code upstream; added appropriate line to /etc/dpkg/origins/ubuntu in base-files so that it will know that Ubuntu is derived from Debian.
  • Fixed up ocaml build, broken as a result of overly Debian-specific Replaces field.
  • Rebuilt sqlite3 against libicu40 to help reduce CD size.
  • Ported installer to blkid.
  • Sponsorship:
    • console-common #375068.
    • db-defaults #375192 (sync).
    • gnome-desktop-sharp2 #375365.


  • Performance reviews (just about made it).
  • Worked on AllHands talk.

Evan Dandrea

  • Continued working on usb-creator refactoring in ~evand/usb-creator/future.
  • Received a 9.04 USB key from our UK shop supplier, verified the image and tested it. They seem to have sorted out their issues from 8.10 (well, everything but shipping test copies out in a timely manner) and I was able to sign off on it in short order.
  • Released a new Wubi into Karmic.
  • Started researching DeviceKit, as I found out via Scott that Martin uploaded it over the previous weekend. Tons of thanks to Martin, it removes lots of headaches from dealing with HAL.

  • Started writing a DeviceKit backend for usb-creator.

  • Finished up UDS specifications, started sending out emails requesting people's presence and making sure Wubi was sorted.
  • Did some research on Windows 7 registry locations and how it maps drive letters to volumes for migration-assistant.
  • Wrote an XML plist (iTunes, OSX software settings) parser for migration-assistant. Still need to finish the binary plist parser.

James Westby

Lars Wirzenius

Luke Yelavich


  • Further writing of documentation relating to the speech work that I will likely be doing in the karmic cycle


  • Audio bug triaging
  • Updated alsa and pulseaudio packages in my PPA to whats in karmic.
  • Started merging alsa driver, utils, lib, plugins for version 1.0.20, however would rather wait till they are in Debian to ensure we share the same orig tarballs.
  • Requested a sync of esound, since Debian has all the changes we made, yay!


  • Worked with Robert COlins to work out how to get his dmraid array working properly, which turns out to be the result of a hpa at the end of one or more of his disks. The resulting patch will mean users don't need to fiddle with libata options to alter the hpa setting.


  • Accidentally merged policykit, not seeing a bug filed for a merge to be sponsored. The apology was accepted by the bug filer, who managed to get a sponsored upload out of it anyway, due to another patch that had to be included to fix another bug, which was going to be an SRU for jaunty.
  • Gave Evolution's calendar another go with Orca, to find that things work somewhat better now. I also discovered a way to use Google Calendar with CalDav in Evolution, so I don't have to use the web UI to create events. I just need to get my calendar set up. I'll likely need one given the work I'll be doing in the next 12 months.

Matthias Klose

Michael Vogt

Mike Terry

Karmic Merging

  • Went on merge rampage to acquire sweet, sweet MOTU points
  • Merge meld (373388)
  • Merge acpi (373397)
  • Merge awn-extras-applets (373685)
  • Merge ttf-wqy-zenhei (373710)
  • Merge gravitywars (373858)
  • Merge trac-git (374946)
  • Merge quodlibet + crash patch (373827, 357351)
  • Merge boost + gcc-4.4 header fix (375047, 375043)
  • Merge nicotine (317063)
  • Merge nip2 (375435)
  • Merge gdesklets + python2.6 fixes (375461, 350562, 344079)
  • Sync wesnoth (364069)
  • Sync gnomad2 (373778)
  • Sync topshelf (374937)
  • Sync nemiver (375156)
  • Sync mod-dnssd (375440)

Tool Fixes in Support of Above Merges

  • Fixed bug in karmic cdbs python support when a package has

architecture dependent code (374892)

Scott James Remnant


  • Migrated from using vol_id/libvolume-id to blkid/libblkid, which itself has moved from e2fsprogs to util-linux-ng and merged much of the code of the other.
    • This finally solves many of the ext4 upstream's complaints about differing responses between the two libraries, by only having one!
  • e2fsprogs:
    • updated to 1.41.5
    • reworked the control and rules file to allow optional disabling of blkid/fsck/libblkid and use of the external, sent patches to Ted
    • made the option default to "disable" on Ubuntu, "enable" on Debian
  • util-linux:
    • updated to 2.15
    • updated packaging to build libblkid, copying all the various bits from the e2fsprogs package
    • later discussion with Colin about a udeb to contain the blkid binary
  • udev:
    • updated to GIT HEAD
    • changed to use blkid instead of vol_id
    • dropped building of vol_id and libvolume-id
  • devmapper:
    • changed to use blkid instead of vol_id
    • proper merge still outstanding due to the merge into lvm2
  • lvm2:
    • updated dependencies for transition
    • proper merge still outstanding due to the merge of devmapper
  • mdadm:
    • changed to use blkid instead of vol_id
    • proper merge still outstanding
  • initramfs-tools
    • changed to use blkid instead of vol_id
    • proper merge still outstanding


  • dropped forced creation of /dev/kmem, which is disabled in our kernels
  • udev-extras updates to include pitti's keymap work


  • Updated/merged to 1.2.14
  • Applied the timeout patches I worked on at the sprint-before-last
  • Broke HAL, had applied a broken version of one of the patches from the wrong directory


  • Updated to 3.8


  • still working on core code


  • git-core, autoconf, automake, libtool, autoconf-nonfree
    • (dholbach may turn feint at the fact I've actually done some <g>)

Steve Langasek

FoundationsTeam/Meetings/2009/0513 (last edited 2009-05-27 10:37:35 by robbie.w)