Ubuntu Wiki

   10:00 mdz        agenda is very lightweight, especially if lifeless doesn't turn up
   === Seveas [n=seveas@ubuntu/member/seveas] has joined #ubuntu-meeting
   10:00 freeflying mdz: hi
   === sivang checks the agenda
   10:01 mdz        freeflying: hi
   10:01 mdz        mjg59 said that he would not be able to make it
   10:01 mdz        sabdfl doesn't seem to be around either, so just me and Keybuk tonight
   10:02 Keybuk     "the terrible twosome"
   === sivang lols
   10:02 freeflying mdz: I'm in ture of apply for dev
   10:02 mdz        freeflying: pardon?
   10:02 freeflying mdz: zpply for ubuntu-developer membership
   10:03 freeflying s/zpply/apply
   10:03 mdz        freeflying: what is your launchpad username?  I don't see you on the list
   10:03 freeflying mdz: zhengpeng-hou
   10:03 mdz        oh, you applied a long time ago apparently
   10:04 mdz        is this the first meeting you've been able to attend since January?
   10:04 Keybuk     mdz: uh, did someone move "Proposed Members"
   === Keybuk doesn't see it
   10:04 mdz        Keybuk: there aren't any for core
   10:04 Keybuk     mdz: ah!
   10:04 mdz        it's there on ubuntu-dev
   10:04 freeflying mdz: this is the second
   10:05 mdz        hmm, I think you applied during the meeting where I wasn't here
   10:05 mdz        what was the outcome?
   10:05 Keybuk     mdz: we asked him to work more closely with the other scim maintainers and the other MOTU
   10:05 Keybuk     as well as increase his usage of Malone
   10:05 freeflying mdz: was rejected one month ago , due to the low karma value
   10:06 Keybuk     s/rejected/deferred/
   10:06 Keybuk     but yes
   10:06 mdz        freeflying: is there still difficulty between you and minghua?
   10:06 freeflying mdz: i don't think we have
   10:07 mdz        dholbach: any feedback from MOTU about freeflying's involvement since?
   10:07 mvo        mdz: I worked with him on various CJK related issues (input methods, fonts) and he is very active
   10:07 dholbach   mdz: i can't say I worked much with freeflying - I think he was more involved with mvo and the KDE guys
   10:07 mdz        mvo: have you sponsored packages or patches for him?
   10:07 mvo        I sponsored some of his patches
   10:08 mdz        freeflying: who have you been working with primarily?
   10:08 freeflying mdz:  Riddell
   10:08 mdz        Riddell: around?
   10:09 Riddell    hi
   10:09 mdz        Riddell: have you sponsored packages for freeflying? any feedback?
   10:09 Riddell    freeflying's packages have been getting steddily better and for scim-bridge which I think was the last I reviewed I just had to correct the English on the description
   10:10 mdz        freeflying: I must say that cooperation is very important in our teams; I'm concerned about the fact that there was such a heated conflict over scim in the past
   10:10 mvo        I sponsored the initial scim-bridge upload and was happy with the quality
   10:11 mdz        freeflying: if you were in the same position again, what might you do differently?
   10:12 freeflying mdz: we had disagreement before on scim
   === sabdfl [n=mark@ubuntu/member/pdpc.silver.sabdfl] has joined #ubuntu-meeting
   10:12 freeflying mdz: but that don't mean I will not cooperate
   10:12 sabdfl     ello there
   10:12 ogra       hey sabdfl
   10:12 sabdfl     sorry to be late
   10:12 sabdfl     where are we?
   === mvo waves to sabdfl
   10:12 sabdfl     hey michael
   10:13 ogra       sabdfl, first ubuntu-dev wannabe
   10:13 ogra       (freeflying)
   10:13 mdz        sabdfl: see paste in /msg
   10:13 mdz        freeflying: if you had such a disagreement today, what might you do differently?
   10:14 freeflying mdz: now we haven't any disagreeement
   === j_ack [n=nico@p508D9043.dip0.t-ipconnect.de] has joined #ubuntu-meeting
   10:14 freeflying mdz: you know scim is in main now , and is the default for cjk users
   10:14 sabdfl     freeflying: i saw some of the correspondence that flew around, and it struck me that you were very confrontational in your approach
   10:15 sabdfl     the other correspondent was bein cautious and reasonable
   10:15 sabdfl     i like your energy, and your willingness to learn and contribute
   === cbx33 [n=pete@84-45-238-195.no-dns-yet.enta.net] has joined #ubuntu-meeting
   10:15 mdz        freeflying: perhaps my question was unclear.  if you had a different disagreement, perhaps with someone else, would you handle the situation in the same way?  or would you do something differently?
   === crimsun [i=crimsun@pdpc/supporter/silver/crimsun] has joined #ubuntu-meeting
   10:15 sabdfl     but i'm a bit concerned about your approach to dealing with folks who are more cautious or who see things differently
   10:15 freeflying mdz: I'd talk with him
   10:16 sabdfl     glad to hear Riddell and others are impressed with the improving quality of your work
   10:16 freeflying sabdfl: thx
   10:17 mdz        Keybuk: his earlier deferral was in part based on your concerns; do you have any questions?
   10:18 Keybuk     I'm glad he's taken up my advice to work more with Launchpad and Malone, so I'm happy with his improvement there
   10:18 Keybuk     But I still don't think he's shown improvement in how he deals with others, especially minghua
   === lbm [n=lbm@x1-6-00-13-10-7a-d1-e4.k233.webspeed.dk] has joined #ubuntu-meeting
   10:18 Keybuk     freeflying: have you spoken to minghua since the last meeting?  worked out your differences?
   10:19 freeflying Keybuk: we have talked something about improve the support for scim
   10:19 mdz        freeflying: according to Launchpad, you've signed the code of conduct, however, I do not feel that your earlier exchange regarding scim honored the CoC
   10:20 mdz        freeflying: however, you have made substantial code contributions which we appreciate, and I would like to recognize your continued involvement
   10:20 sabdfl     freeflying: we don't want to beat up on you for a past approach which you've now gotten over
   10:20 sabdfl     i recognise that you're on a steep learning curve, and doing very well
   10:20 sabdfl     but this interpersonal approach is very, very important in ubuntu
   10:21 freeflying sabdfl: I see , I just appreciate this , so Iinvolved in ubuntu community
   10:22 sabdfl     ok
   10:22 Riddell    I didn't see much of the scim disagreement, but freeflying has always been very couteous and good to work with on #kubuntu-devel
   10:22 Keybuk     I think for me, he's shown enough of an improvement in his developer skills to join the ubuntu-dev team; and that by joining the team he'll improve his communication
   10:22 Keybuk     he's certainly shown great willingless to improve
   10:22 sabdfl     freeflying: do we have your commitment to think carefully before you flame someone with whom you disagree?
   10:22 mvo        I have a good working relationship with him too
   === mvo didn't knew about the scim disagrement
   10:23 Riddell    seems he and others in the Chinese ubuntu community were quite upset about the fridge story that was posted today, but freeflying was very good at explaining his concerns
   10:23 sabdfl     mvo: it was more a disagreement on the maturity of the code, and as it happens, freeflying's approach of "lets just get it in" won the day
   10:24 sabdfl     fridge story?
   10:24 freeflying sabdfl: I will not flame with any disagreement
   10:24 sabdfl     freeflying: ok, then +1 from me, and a big thank you for your contributions so far!
   10:24 freeflying sabdfl: I  think anything can be solved by commnuication
   10:24 freeflying sabdfl: thanks
   10:25 Riddell    sabdfl: looks like it has been removed now
   10:25 sabdfl     hmm... sounds like i should know a bit about htat
   10:25 mdz        +1, but with a strong urging to remember the CoC and resolve disagreements amicably, avoiding personal conflict
   10:25 sabdfl     anyhow, keybuk? mdz?
   === wold [n=wold@ev-217-129-81-225.netvisao.pt] has joined #ubuntu-meeting
   10:26 Keybuk     +1 from me
   10:26 freeflying mdz: thanks , and I will
   10:26 mdz        freeflying: if you find yourself in a difficult position, feel free to peaceably escalate the issue to the tech board or community council; we're here to help
   10:26 freeflying Keybuk: thanks
   10:26 freeflying mdz: ya
   10:26 mdz        freeflying: welcome to the team
   10:27 mvo        congrats freeflying!
   10:27 ogra       congrats freeflying
   10:27 freeflying thansk  mvo  Riddell  sabdfl  mdz Keybuk
   10:27 dholbach   congratulations freeflying
   10:27 freeflying ogra: thanks
   10:27 freeflying dholbach: thx
   10:27 mdz        is there anyone else here applying for ubuntu-dev?
   10:27 mdz        I pinged lifeless earlier, but he doesn't seem to be around
   10:28 sabdfl     freeflying: what's your launchpad nick?
   10:28 mdz        sabdfl: https://launchpad.net/people/zhengpeng-hou
   10:28 LaserJock  congrats freeflying
   10:28 mdz        (I updated LP already)
   10:28 cbx33      congratulations freeflying
   10:28 sabdfl     ah
   10:28 sabdfl     thanks mdz :-)
   10:28 mdz        not hearing from any other ubuntu-dev applicants, sladen has the first discussion item
   10:28 freeflying cbx33: thx
   10:28 sabdfl     lifeless is hopefully fast asleep right now
   10:28 sabdfl     or writing tests
   10:28 mdz        sladen: after seeing the agenda item, I commented on the bug with my feelings on the matter
   10:29 Keybuk     sabdfl: I thought he did that in his sleep
   10:29 mdz        https://launchpad.net/distros/ubuntu/+source/ubuntu-meta/+bug/36838/+index
   10:29 Ubugtu     Malone bug 36838 in ubuntu-meta ubuntu-minimal "ubuntu-minimal/Server does not depend on acpi-support" [Normal,Rejected]
   10:29 Keybuk     first up
   10:29 Keybuk     obvious problem
   10:29 Keybuk     even if we wanted to, we can't change ubuntu-minimal right now
   10:29 Keybuk     unless kamion has overcome that bug
   === olemke [n=olemke@p5489461B.dip.t-dialin.net] has joined #ubuntu-meeting
   10:30 mdz        we shouldn't change it at this point in the release cycle anyway, but we can answer the question of what it is meant to be
   10:30 Keybuk     why doesn't ubuntu-minimal provide acpi-support?
   10:30 Keybuk     seems logical to me for that to be there
   10:31 mdz        we do say that minimal is meant to provide hardware support, but I see power management as a special case
   10:31 mdz        especially in the context of servers
   10:31 Keybuk     true, most servers don't tend to need hibernation
   10:32 mdz        and they usually don't want powernowd or the like etiher
   10:32 sabdfl     do they not cycle down into "slow and power saving" mode?
   10:32 mdz        and they definitely don't want to automatically power down
   10:32 Keybuk     mdz: unless the power goes out?
   10:32 sabdfl     i thought heat generation was the #1 issue in data centers right now?
   10:32 sabdfl     do Xeon's support speed steppings?
   10:32 mdz        sabdfl: I think the #1 issue continues to be $ per unit of work done
   10:33 elmo       mdz: not really
   10:33 sabdfl     mdz: where some not insignificant chunk of $ goes to air conditioning
   10:33 elmo       sabdfl is right
   10:33 sabdfl     QUOTES PAGE
   10:33 elmo       most vendors are incorporating speedstep stuff into their server line
   10:33 elmo       precisely because of heat/power issues
   10:33 sivang     sabdfl: IBM's pServers do that :)
   === llolp [n=mark@80-42-155-192.dynamic.dsl.as9105.com] has joined #ubuntu-meeting
   10:34 sivang     (move into low workload state with less power consumption and low heat)
   10:34 mdz        elmo: my point is that that is only one part of the larger equation
   10:34 mdz        and fwiw, acpi-support has nothing to do with frequency scaling afaik
   10:34 sabdfl     that's powernowd etc.
   10:35 sabdfl     do we install the frequency scaling stuff in desktops by default?
   10:35 mdz        it's mostly infrastructure for suspend and hibernate
   10:35 mdz        sabdfl: yes, since warty or so
   10:35 sabdfl     and is it a low-level, daemon like thing that could also go on ubuntu server?
   10:35 mdz        yes
   10:35 sabdfl     can it cause problems with hardware that doesn't implement specs correctly?
   10:35 mdz        I believe all of its dependencies are already in -minimal
   10:35 mdz        unknown
   10:36 mdz        note that it's near impossible to exclude packages from it
   10:36 mdz        if the package does trigger a problem, the user is in trouble
   10:36 mdz        but I don't know of any such issues with cpufreq
   10:36 sabdfl     too late for us to get meaningful feedback pre-dapper
   10:36 mdz        agreed
   10:36 sladen     mdz: ta.  yup, I closed it after that
   10:37 mdz        on the larger question of which bits of power management infrastructure might be appropriate for servers, I think we should revisit that post-Dapper
   10:37 mdz        and I think it would mean an ubuntu-server metapackage rather than expanding -minimal
   10:38 ogra       ++
   10:38 cbx33      sounds like a good idea to me
   10:38 mdz        sladen: you're satisfied with the specific answer to your question, though, right?
   10:38 sladen     mdz: yeah.  I rather have somebody else decide the policy :)
   10:38 mdz        I've added the server power management question to my post-dapper list
   10:39 mdz        next agenda item is from janimo
   10:39 janimo     there are a handful of packages I'd very much like to get into xubuntu. They are gtk-only builds of regular gnome apps. But the gtk only support is either not (yet) upstream, or not taken advantage of in debian packaging
   10:39 janimo     having them separately packaged conflicts with ReducingDuplication
   10:39 mdz        janimo: is it possible to merge your changes into the existing packages, so that they build an alternative set of binaries?
   10:39 mdz        so one source package, 2xN binary packages
   10:39 janimo     mdz, yes that would be ideal
   10:40 mdz        reducing-duplication is only concerned with source
   10:40 janimo     but since most (all?) of thsoe use CDBS
   10:40 janimo     packaging either needs to be very much tweaked
   10:40 janimo     or cdbs1 hacked into supporting multibuild
   10:40 janimo     the former need agreement of respective mainatiners
   10:40 janimo     the latter I am not sure how feasible it is
   10:41 mdz        have you talked with the desktop team about this?  they know cdbs quite well
   10:41 janimo     mdz, have not yet.
   === sladen had a similar issue when wanting to build -i386-win32 binaries for qemu. ideally those would be in the archive too, but I guess the important thing is just argumenting the source-package to always produce two binary packages.
   10:41 mdz        perhaps they will have some ideas about how it can be done non-intrusively
   10:41 janimo     ok I'll talk to them
   10:41 mdz        but understand that we are very much in feature freeze now
   10:42 mdz        and if these changes are to receive an exception, they will need to be done with minimal interference with the existing, working packages
   10:42 janimo     mdz, sure. But packaging changes while ugly , should not affect the outcome of the gnome packages at all
   10:42 sabdfl     janimo: good work on xubuntu btw
   10:42 janimo     sabdfl: thanks
   10:42 sabdfl     i have it running on a machine here, and am testing it on low-end hardware
   10:42 sabdfl     very impressive
   10:42 sabdfl     very clean, very fast
   10:43 janimo     sabdfl: great, if you have any suggestions, -> malone :)
   10:43 mdz        janimo: which applications do you want to do this with?  do you have a list?
   10:43 janimo     evince, gnome-system-tools, gnumeric, cups-manager
   10:43 sabdfl     janimo: and -> blueprint ;-)
   10:43 mdz        gnome-system-tools without gnome? ;-)
   10:43 janimo     gnumeric maintainer said he's accept patches in debian, but is slow to respond
   10:44 janimo     mdz, you'd be surprised to see how little gnome code is in there
   === sivang seconds janimo
   10:44 janimo     nautilus share and a few gconf bits
   10:44 ogra       yeah, they originally were the ximian-system-tools
   10:44 janimo     I actually have xubuntu-system-tools in universe now
   10:44 sabdfl     janimo: if you're willing to handle merges from debian, go ahead and update the ubuntu package
   10:44 mdz        janimo: do you have a patch for gnumeric?  I don't think it uses cdbs
   10:44 janimo     works the same as the original
   10:44 janimo     mdz, gnumeric actually has --disable-gnome in upstream
   10:45 janimo     as it runs on win
   10:45 janimo     and we (Gauvain Pocentek an dmyself) have worked on the pacthes for gnimeric and libgoffice
   10:45 mdz        janimo: but in debian and ubuntu, it's built for gnome
   10:45 janimo     right
   10:45 janimo     but we have pacthes to buold both of them
   10:45 janimo     and yes this is debhelper so it's closer to the goal
   10:46 mdz        perhaps the ideal solution would be to maintain a xubuntu branch of those packages with different packaging, but we don't yet have the infrastructure to support that
   10:47 mdz        janimo: as sabdfl said, we can certainly trial the patches in Ubuntu if the risk is minimal (build failures are very disruptive close to release, though)
   10:47 janimo     mdz, sure. I just want to know (preferably from each maintainer) which way to go
   10:47 seb128     please don't fork totally the packaging away from Debian
   10:47 janimo     convert to debhelper?
   10:47 seb128     that's going to mean extra work to merge, etc
   10:47 seb128     like cdbs to debhelper
   10:47 janimo     seb128: you are involved with most of these packages I think
   10:48 mdz        seb128: agreed
   10:48 janimo     what do you prefer
   10:48 seb128     yeah, and as already discussed I'm not happy to fork the packages like that
   10:48 janimo     I put this on agenda to find a way we are all happy with
   10:48 seb128     if we switch from cdbs to debhelper the package is totally different from the Debian one and it means extra work for dholbach and me updating them, syncing with Debian, etc
   10:49 janimo     seb128, could be a temporary switch fro dapper, hoping for dapper+1 either cdbs improves or something else saves us
   10:49 seb128     grumpf
   10:49 janimo     either way I said I'd help maintain all packages I'd have to tweak
   10:49 mdz        janimo: it's especially important late in the release cycle that we stay agile
   10:49 ogra       thats a bad idea for a 3 year support cycle
   10:49 mdz        so that if we need last-minute fixes we can merge them quickly
   10:49 janimo     I am comfortable with any solution you propose as long as these packages can go into default xubnutu
   10:50 seb128     I'm not happy to redo totally those package to use debhelper now
   10:50 seb128     one week before dapper beta
   10:50 mdz        it's truly not a good time to add something like this to dapper
   10:50 janimo     seb128: I don't like debhelper much either
   10:50 mdz        we can consider an exception, but only if it can be done very simply and safely
   10:50 janimo     I;ll continue looking into cdbs multibuild then
   10:50 mdz        which in the cdbs case,  it sounds like it cannot
   10:50 sabdfl     agreed, refactoring the packages totally now is not worth it
   10:51 janimo     Kamion proposed a workaround will have to try it.
   10:51 mdz        what was his proposal?
   10:51 janimo     not sure about the details, have to look it up
   10:51 janimo     said it on irc
   10:51 mdz        ok, I'm happy to review it via email
   10:51 janimo     something using a rules file which calls existing one and a new one
   10:52 janimo     so debhelper is not an option for packages using cdbs right?
   10:52 janimo     I am ok with that
   10:52 mdz        janimo: I don't think that would be best
   10:52 mdz        janimo: unfortunately we don't have a perfect solution at this point, but we can continue to discuss alternatives
   10:52 janimo     if we cannot get clean packaging to do this, how bad is having different source packages
   10:52 ogra       and it would likely upset debian maintainers again#
   10:53 ogra       twice the work if you have a security patch
   10:53 mdz        janimo: sorry we don't have a magic bullet to offer, but are you satisfied enough with the discussion to take the remainder to email?
   10:53 janimo     sure
   10:53 mdz        thanks
   10:53 mdz        is there any other business?
   10:53 janimo     ubuntu-devel ?
   10:54 mdz        janimo: sure, but CC me or bring it to my attention or I may not see it there
   10:54 mdz        ubuntu-devel is very noisy these days
   10:54 janimo     ok
   10:55 mdz        last call for other business
   10:55 mdz        ok, adjourned
   10:55 mdz        thanks, everyone