Distro/irc20050913

• date: Sep 13, 2005
• checkin time: 7-7:15pm PDT -0700 (10-10:15pm EDT)
• meeting time 7:15pm PDT -0700 (10:15pm EDT)
• meet in irc.freenode.net #freegeek-distro

Agenda

unfinished items from last meeting:

• freekbox3 packaging:
  • building new freekbox3 packages [1]
  • new freekbox3 package archive possibilities:
    • mini-dinstall
    • reprepro
    • built by hand, but similar structure to "regular" debian archive

ongoing items:

• user-oriented-design
• Prognosis: how is this project looking?
• Long-term outlook - including definition of objectives, estimation of required work, and timeline
• hosting our own popularity-contest server
• trimming a multi-profile Install CD
• svn repository hosted at svn.freegeek.org ?

new items:

• new items here

commitments from last meeting:

• skippy
  • continued research of popcon server

• gate
  • setup freekbox3-gnome package

• stillflame
  • post a note to distro asking for a facilitator

• Vagrant
  • post to distro list about new "official" freekbox3 packages
  • post to distro list about re-structuring freekbox3 package archive
  • document reprepro on the wiki

• tymp
  • start a media player page on the wiki

• everybody
  • read enrico zini's zen-paper on user-oriented-design

• anybody
  • look for similar themes in current personae, drop inappropriate debian-np personae

log / summary

Present

skippy(scott in columbus - facilitator), tymp(rob in olympia), vagrant("freegeek at large"), gate(greg, columbus), romdos(dave, pdx), stillflame(martin, pdx), WormRunner(michael, pdx), meho(matteo, pdx)

Summary

• freekbox3 packaging:
  • building new freekbox3 packages [1] (http://web.freegeek.org/freekbox3)
    • it's desirable to have new 'official' fb3 packages as the cvs version is considerable out of sync with the release packages (0.1.2). commitments to build and test new packages (vagrant, martin)
  • new freekbox3 package archive possibilities
    • hand-rolled method is most flexible, and this will be tried. commitments to implement this and document it (vagrant, michael, rob)

• user-oriented-design
  • is this a 'project' or an approach, or both?
  • how to inform the notion of the 'user'?
    • poll adoption teachers, tech support, [build teachers]
    • gather/use user feedback
  • choosing the 'user persona(e)' to design for
    • several personae
      • per local freegeeks [what makes them different?]
      • per types of use
        • 'adopter' user
        • startup fg user
        • other? ('grantbox' user, teachers, ...)
    • reducing the number of personae
      • compile/reduce existing personae
      • or ?

• Prognosis and Long-term outlook
  • how transferable is fgpdx distro to other freegeeks?
  • change in work-process due to inclusion of newer freegeeks...
    • social factors
    • technical factors

• hosting our own popularity-contest server
  • commitment to install/test by next meeting (scott)

• trimming a multi-profile Install CD
  • what should the cd include, user-reinsall cd vs 'freegeek on a disc'
    • profiles, package-sets: laptop, desktop, base-server, 'workstation', lessdisks server...
  • simplifying testing: test w/ actual cdrw burn/installs?
  • reducing size of package-set(s) to fit on cd
  • definition of objectives in making an install cd...

• svn repository hosted at svn.freegeek.org
  • commitment to get it working (martin).
  • will be setup for webdav and svn+ssh. logins for all are needed.

Commitments

martin:

• get tested a freekbox3 package release
• post instructions to pdx QC process.
• finish svn, add user access (either ssh or webdav)
• post log of meeting

scott:

• allow vagrantc to hand-roll packages until a better method is devised.
• get feedback from build teachers [on user personae].
• install and test a popcon server before next meeting.
• take to the distro list the question of objectives for CD images.

rob:

• spend 4 hours over next two weeks helping document putting packages packages in repository (michael wanted to help with this "as long as I can poke you enough where I fail to understand")
• post summary of meeting

michael:

• I will email the adoptionteachers list [to arrange a meeting to discuss developing user personae]

vagrant:

• document freekbox3 installer CD build cycle

matteo:

• attend next meeting

Full Log

[7:10pm] <vagrantc> we also, in addition to logging the meeting notes, should find someone like ideath to consolidate the notes after the meeting ... i spent probably half an hour reading through the log mostly just to get commitments from folks.
[7:10pm] <stillflame> re: notes, i'm logging and can post them up afterwards. is that enough?
[7:10pm] <stillflame> right.
[7:11pm] <WormRunner> vagrantc: thanks
[7:11pm] <stillflame> i won't have time to do the log parsing.
[7:12pm] <stillflame> at least, not in the next week.
[7:12pm] <tymp> i'll try to note the main points to post as well - though won't likely have time to parse the log soon either
[7:12pm] <stillflame> how about we agree to an easily searchable commitment format, like "COMMIT: ..."
[7:12pm] <tymp> good idea
[7:13pm] <WormRunner> yes
[7:13pm] <tymp> is anyone pseudo-facilitating?
[7:13pm] <vagrantc> stillflame: very clever
[7:13pm] <vagrantc> tymp: i think that would be skippy
[7:13pm] <tymp> great
[7:14pm] <stillflame> that also could be used for decisions ("DECISION: ...") or anything else we deem important.
[7:14pm] <skippy> okay -- it's about that time.
[7:15pm] <tymp> ..something for future agenda items too?  agenda:
[7:15pm] * stillflame nods.
[7:15pm] <skippy> check in. Who's present.
[7:15pm] <skippy> skippy, freegeek columbus.
[7:15pm] <tymp> hi, rob here in olympia..  
[7:16pm] <vagrantc> i'm here, but i haven't had a lot of time to work on distro ... i'm part of freegeek at large :)
[7:16pm] <gate> greg, columbus
[7:16pm] <stillflame> i'm coming out of freegeek portland oregon
[7:16pm] <romdos> dave, pdx
[7:16pm] * stillflame pokes WormRunner
[7:16pm] <skippy> has everyone reviewed the agenda? Any last minute changes / additions ?
[7:16pm] <WormRunner> michael, portland fg
[7:17pm] * vagrantc notices that the agenda is in the topic for this channel
[7:18pm] <tymp> skippy : i reviewed it, would say go with it ..
[7:18pm] * stillflame is good
[7:18pm] <skippy> first item: unfinished biz.
[7:18pm] <skippy> 1a) FB3 packaging.
[7:19pm] <skippy> who wants to speak about that?
[7:19pm] <tymp> ..not sure whether the listed items are that relevant at this moment..
[7:19pm] <tymp> maybe more important is the recent distro conversation between vagrantc and gate
[7:19pm] <tymp> is that the same topic?
[7:19pm] <vagrantc> i can talk about packaging, if we want to talk about it now.
[7:19pm] <tymp> oi
[7:19pm] <gate> sounds fun
[7:20pm] <skippy> go ahead, vagrantc
[7:20pm] <vagrantc> tymp: i would say it's a third topic on that theme, but it's also a new item
[7:20pm] * tymp nods
[7:20pm] <vagrantc> so, i want to get new "official" freekbox3 packages, since there have been a lot of changes in cvs not present in the released packages (0.1.2)
[7:21pm] <vagrantc> i need some people to test new packages, and ideally folks from freegeek portland.
[7:21pm] <vagrantc> that's the first item.
[7:21pm] <gate> sounds like a good idea, release before we break it
[7:21pm] <tymp> is there any procedure established for getting the packages updated?
[7:21pm] <vagrantc> it would ease the CD building considerably ... any takers?
[7:21pm] <tymp> i mean, for getting agreement to take action on that?
[7:22pm] <tymp> vagrantc : you're asking for someone to *do* it?
[7:22pm] <stillflame> vagrantc: i can offer to have available users try out installer disks.
[7:22pm] <stillflame> provided there's an iso i can burn from
[7:22pm] <vagrantc> tymp: i can build the packages, or help someone build the packages, but i need someone to test the packages.
[7:23pm] <WormRunner> I can do an install or two and see how it goes
[7:23pm] <skippy> is there a defined testing criteria / process?
[7:23pm] <vagrantc> not that i'm aware of.
[7:23pm] <skippy> is such a thing necessary?
[7:23pm] <tymp> perhaps its second nature to those in pdx?
[7:23pm] <skippy> What constitutes "testing" ?
[7:24pm] <tymp> yes, i'm not clear either
[7:24pm] <vagrantc> probably go through the QC process ?
[7:24pm] <WormRunner> yeah. for start
[7:24pm] <stillflame> okay
[7:24pm] <vagrantc> maybe have someone who's been a teacher to run through a "class" with it?
[7:25pm] <stillflame> vagrantc: adoption teacher?
[7:25pm] <vagrantc> stillflame: yeah, and build teachers too, probably.
[7:25pm] <skippy> sorry for being obtuse, but does this mean that Oly and CMH are unable to help test?
[7:25pm] <stillflame> WormRunner: can you try to pull an adoption teacher in that?
[7:26pm] <WormRunner> sure. and I have been doing some of that myself
[7:26pm] <vagrantc> anyone who wants to test can give input, but since portland relies on it crucially, i think it's crucial to have testers there.
[7:26pm] <tymp> tangent : i'm going to set about polling for feedback/approval of the gui changes i made... a procedure or suggestions how best to do this would help
[7:26pm] <stillflame> okay, COMMIT: to get tested a freekbox3 package release
[7:27pm] <skippy> but if there's no dcoumented test process, how can we best participate?
[7:27pm] <stillflame> tymp: would the distro RT work?
[7:27pm] <stillflame> skippy: ah, right. i can offer up slightly outdated QC instructions.
[7:27pm] <skippy> stillflame: that would help. please do.
[7:27pm] <vagrantc> skippy: eyeball it up, see if anything is obviously broken ... walk through the adoption class outline, do build QC process ...
[7:28pm] <vagrantc> that's a start.
[7:28pm] <stillflame> COMMIT: post instructions to pdx QC process.
[7:28pm] <tymp> stillflame : yes.  i'm also planning to spam adoption-teachers, build, and something
[7:28pm] <skippy> next item: 1b) new images
[7:29pm] <vagrantc> skippy: could you clarify?
[7:29pm] <skippy> vagrantc: "building new freekbox3 packages".
[7:29pm] <vagrantc> oh, i thought that was what we just covered ?
[7:30pm] <skippy> okay. Item 1c) mini-dinstall / reprepro / hand-rolled
[7:30pm] <vagrantc> again, i put it on the agenda, so i'll summarize
[7:30pm] <vagrantc> basically, it would be helpful to have a more standard package repository format
[7:31pm] <vagrantc> so that tools such as debpartial-mirror and reprepro can more easily mirror our freekbox3 package archive
[7:31pm] <stillflame> what's the cost of doing so?
[7:31pm] <vagrantc> our current archive is at http://web.freegeek.org/freekbox3
[7:32pm] <vagrantc> stillflame: just need to agree on a process for building the archive, and learn different commands.
[7:33pm] <vagrantc> mini-dinstall would be most useful for developers making experimental releases and cvs builds, reprepro would be best for standard archive format for our released packages, and our hand-rolled one could be as good as reprepro, but more flexible ... but it would require writing it.
[7:33pm] <stillflame> so we just need to choose between those three mentioned?
[7:33pm] * stillflame withdraws his last comment
[7:33pm] <gate> by hand is to much work
[7:33pm] <tymp> vagrantc : what do you think is best?
[7:34pm] <vagrantc> gate: "by-hand" means writing a very simple 5-15 line script to do it, and i already have most of it written for the lessdisks project
[7:34pm] <gate> ok so thats not as bad then, I though u ment copy files and then create packages.gz
[7:35pm] <vagrantc> tymp: reprepro might be the easiest to start with, but it likes to delete old packages, whereas i would like to keep them around.
[7:35pm] <tymp> vagrantc : so you are not sure?  seems like no one else is either
[7:35pm] <gate> reprepro sounds bad to me
[7:36pm] <vagrantc> mini-dinstall is the easiest, but the archive format isn't really any better.
[7:36pm] * WormRunner expresses his ignorance of these matters
[7:36pm] <vagrantc> a backported version of reprepro might also satisfy our needs.
[7:36pm] <vagrantc> but then we've got to deal with a backport ...
[7:36pm] <WormRunner> is reprepo modifiable easily?
[7:36pm] <skippy> is one demonstrably better than the others, even if it still has shortcomings?
[7:36pm] <vagrantc> WormRunner: it's written in C ... so i'd say no.
[7:37pm] <gate> if we have to deal with packports whats wrong with the hand meathod
[7:37pm] <WormRunner> ok
[7:37pm] <vagrantc> skippy: i guess i'm a little inclined towards hand-rolled scripts ...
[7:37pm] <tymp> if we said hand-rolled, what would be a resonable deadline to have that?
[7:38pm] <gate> does apt-move or any other tools help with it
[7:38pm] <tymp> ie, how urgent vs time required and available
[7:38pm] <vagrantc> the exact tool isn't even that important, in my opinion ... i guess it's more an issue of process- how do we decide what goes into the package repository, who does it, etc.
[7:39pm] <stillflame> what's the importance of serving up old package versions?
[7:39pm] <vagrantc> tymp: i could get a functional archive working in a matter of days with the hand-rolled method.
[7:39pm] <vagrantc> stillflame: historical relevence, easy to revert if we broke something badly
[7:39pm] <gate> all the stable releases should stay and maybe an auto build after a cvs commit
[7:39pm] <skippy> can we rsync or manually copy old packaes in some way?
[7:40pm] <vagrantc> we could use a method that uses both mini-dinstall and some other method ...
[7:40pm] * tymp really has no investment yet in this issue
[7:41pm] <skippy> so, can we take action, or do we need to discuss on the list some more?
[7:41pm] <vagrantc> would anyone want to work with me on these issues, getting it done, getting it documented, and just giving us the ok to just do what we want?
[7:41pm] <vagrantc> within reason :)
[7:42pm] <WormRunner> I would say if vagrantc has an idea of how to get it done and wants to, we could go with his hand-rolled idea
[7:42pm] <tymp> i *want* to, but can't commit this month
[7:42pm] <gate> I would but It will have to hold off till Oct
[7:42pm] <tymp> yay, October!
[7:42pm] <tymp> (me too)
[7:42pm] <vagrantc> ideally, i'd like to have it done by next meeting ...
[7:42pm] <tymp> i second WormRunner
[7:43pm] <tymp> vagrantc : say "commit"? :))
[7:43pm] <vagrantc> i haven't found a co-conspirator yet
[7:43pm] <skippy> commit: allow vagrantc to hand-roll packages until a better method is devised.
[7:43pm] <tymp> what investment will the co-conspiritor need to make?
[7:44pm] <vagrantc> maybe they get to do the documentation- how to put packages into the repository, mainly.
[7:45pm] <vagrantc> so that the documentation works for someone inexperienced with these matters already
[7:45pm] <tymp> i can commit 4 hours before next meeting to that .. ?
[7:45pm] <tymp> .. i think i am inexperienced enough :)
[7:45pm] <vagrantc> i can just do it, and hope that my documentation is good enough
[7:46pm] <vagrantc> and if anyone wants to help, they would be welcome to
[7:46pm] <tymp> commit : spend 4 hours over next two weeks helping document putting packages in repository
[7:46pm] <WormRunner> I would be happy to help document, as long as I can poke you enough where I fail to understand
[7:46pm] <vagrantc> sounds great!
[7:47pm] <skippy> okay; sounds like consensus to me. Vagrant will prepare packages, and WormRunner and tymp will assist documentation.
[7:47pm] <tymp> is communication on the agenda?
[7:47pm] <skippy> next section: on-going business.
[7:47pm] <tymp> is communication on the agenda?
[7:47pm] <skippy> no, tymp, it's not.
[7:47pm] <tymp> i'd like to parking-lot that
[7:48pm] <skippy> what does that mean?
[7:48pm] <tymp> (ie, if we get through the rest of the agenda...)
[7:48pm] <WormRunner> =table
[7:48pm] <skippy> ok. if we get through.
[7:48pm] <WormRunner> I presume
[7:48pm] <tymp> ha, that's simpler
[7:48pm] <skippy> item 2a) user-oriented-design.
[7:48pm] <skippy> has everyone read the Zini paper?
[7:48pm] <gate> yup
[7:48pm] <tymp> yes. 
[7:49pm] <skippy> any questions?
[7:49pm] <vagrantc> re-read it earlier today :)
[7:49pm] <tymp> skippy : can you estimate time we should spend on this item?
[7:49pm] <skippy> tymp: it's an on-going agenda item, so I'm comfortable giving this 5 - 10 minutes tops.
[7:49pm] <skippy> does anyone have any questions re: zini ?
[7:50pm] <tymp> well... re this project of doing useroriented design..
[7:50pm] <stillflame> does moving toward more user-centric design imply that we would move to distro to debian-np?
[7:50pm] <tymp> i'd like to see some timeline
[7:50pm] <stillflame> move *the* distro
[7:51pm] <tymp> that seems like something that's worth considering
[7:51pm] <vagrantc> stillflame: not at all.
[7:51pm] <vagrantc> stillflame: it is a totally separate issue.
[7:51pm] <skippy> stillflame: I think debian-np is more "business user" oriented.
[7:51pm] <tymp> or, perhaps "move" is too strong
[7:51pm] <stillflame> okay, thanks.
[7:51pm] <skippy> tymp: timeline for this in what way?
[7:51pm] <tymp> goals, dates
[7:51pm] <vagrantc> user-oriented-design is just techniques to inform our design decisions
[7:51pm] <tymp> 1 month, 1 year
[7:52pm] <WormRunner> a way of thinking
[7:52pm] <tymp> i think it's a long-term project
[7:52pm] <vagrantc> it is a life-long project, but i see some specific goals:
[7:52pm] <WormRunner> I think this is a way of thinking we just need to never forget
[7:52pm] <vagrantc> create a persona for each of the freegeek categories
[7:52pm] <vagrantc> trim the adoption program personae down to 1 or 2 personae
[7:53pm] <WormRunner> 3
[7:53pm] <tymp> does anyone else think : there are kinda two projects here 1) bottom-up, working with what we have and making it cleaner, 2) top-down, thinking about what it should be long-term
[7:53pm] <stillflame> so this is something that the adoption teachers should contribute heavily to
[7:53pm] <vagrantc> exactly how many is up for debate, but the fewer the better, or we'll have a very muddy design.
[7:53pm] <WormRunner> ok
[7:54pm] <vagrantc> stillflame: absolutely!
[7:54pm] <WormRunner> stillflame: yes, I think so
[7:54pm] <WormRunner> and support
[7:54pm] <tymp> stillflame : yea, and feedback systems that we hope to build in would contribute
[7:54pm] <vagrantc> i think the most appropriate areas of expertise are adoption teachers and tech support.
[7:55pm] <tymp> i think there's a meta-issue with design that requires a different approach
[7:55pm] <tymp> it's not very technical at all
[7:55pm] <stillflame> okay, i think a physical meeting is needed to really start these personae off, so WormRunner, romdos, meho and/or i need to get a meeting organized with teachers here to define them.
[7:56pm] <WormRunner> commit: I will email the adoptionteachers list
[7:56pm] <skippy> I've loosely fleshed out what I perceive to be an FGCMH user.
[7:56pm] <vagrantc> hopefully, the various freegeeks will be able to make use of the same personae, but there may be differences to consider.
[7:56pm] <stillflame> WormRunner: and set up a meeting to define personae
[7:56pm] <tymp> the ideal will be one persona
[7:56pm] <tymp> * to have one persona
[7:56pm] <vagrantc> but the ideal is impossible, by the very diversity of freegeek.
[7:57pm] <gate> I don't see the persona as that specific that we could not all get buy with one
[7:57pm] <stillflame> skippy: good. would it be okay for the meeting to try to absorb that persona into another?
[7:57pm] <WormRunner> I see several distinct types of users
[7:57pm] <vagrantc> it should largely be a "sterotypical" user- but defined enough to get a feel for their personality
[7:57pm] <skippy> gate: Maybe not at the moment, but I can envision, say, an Hispanic FreeGeek initiative, that might have different goals from FGCMH
[7:58pm] <skippy> stillflame: you folks are welcome to build upon my start; but I'd like to keep a separate FGCMH persona, so that we don't mis-direct our efforts once we start deploying systems.
[7:58pm] <vagrantc> WormRunner: it's likely that freegeek just has a hard design.
[7:58pm] <gate> well yes but as it currently stands I see maybe 2 at most
[7:59pm] <tymp> i hope we can keep an open mind to agreeing on one
[7:59pm] <vagrantc> skippy: well, once you start deploying systems, you may need to create a persona from scratch and forget about the old one ...
[8:00pm] <skippy> vagrantc: perhaps. But that would happen whether FG/PDX absorbs our persona or not.
[8:00pm] <skippy> gate: did you review my persona sketch ?
[8:00pm] <gate> ya
[8:00pm] <vagrantc> skippy: definitely. and if you're lucky, you'll realise that the other freegeek persona fit you just fine :)
[8:00pm] <gate> I read it a few days back
[8:01pm] Quits: meho (n=matteo@209-162-205-24.dq1mn.easystreet.com) ("leaving")
[8:01pm] <skippy> vagrantc: please excuse my skepticism. You may be completely correct; but you folks have a lot of inertia behind a lot of what you've done, and we're in a position to enjoy a fresh start with no preconceived notions. I'd like to use that as much as possible.
[8:01pm] <gate> seems a lot like the rest, just a different background but computer uses were about the same between them
[8:01pm] Joins: meho (n=matteo@209-162-205-24.dq1mn.easystreet.com) Clones: romdos & stillflame [10 users]
[8:02pm] <skippy> anyway, we're straying from the task at hand.
[8:02pm] <tymp> skippy : i think your skepticism is sensible.  but the hope would be..
[8:02pm] <vagrantc> skippy: of course. only time will give a real idea if any of the personae are good.
[8:03pm] <tymp> that's more than 10 minutes
[8:03pm] <skippy> it seems to me that the general consensus for the currently defined personae are that they are task-oriented _users_, and not tweakers or customizers.
[8:03pm] <tymp> yes
[8:03pm] <gate> as stated before get the teachers/tech support in here to tell us what really needs done
[8:04pm] <vagrantc> i think it is important to note that all of the existing adoption personae were created by a single person.
[8:04pm] <skippy> commit: get feedback from build teachers. (I think someone said that already)
[8:04pm] <WormRunner> yes
[8:04pm] <gate> just restating so we can move on
[8:04pm] <vagrantc> skippy: though that would largely be for a build persona, not the adoption persona
[8:04pm] <WormRunner> and adoption teachers and tech support
[8:05pm] <WormRunner> so now that we have committed to a process for that, what is next
[8:05pm] <skippy> next item: 2b) prognosis
[8:05pm] <skippy> which I take to include long-term outlook, objectives, and timeline
[8:06pm] <tymp> we're dead in 6 months
[8:06pm] <WormRunner> ?
[8:06pm] <skippy> are you being silly, tymp ?
[8:06pm] <tymp> just kidding
[8:06pm] <tymp> sorry
[8:06pm] <skippy> okay.
[8:06pm] <gate> I was going to say, I just say the doc and she did not tell me this
[8:06pm] <WormRunner> tech support people have no sense of humor. don't confuse us
[8:07pm] <tymp> lol
[8:07pm] <skippy> I think the addition / collaboration of new FGs will help many aspects of the project(s) overall, but it will also slow things down a fai rbit.
[8:07pm] <skippy> we don't have any real milestones for the distro project.
[8:08pm] <skippy> does anyone have any comments before moving on?
[8:08pm] <tymp> yeah, seems like there is a fundamental shift?
[8:08pm] <vagrantc> well, the freekbox3 is an already mostly finished milestone- many systems are already out there being used!
[8:08pm] <tymp> i think skippy means, the project as non-pdx-ers know it
[8:09pm] <WormRunner> and other than some minor stuff, they are working pretty well
[8:09pm] <skippy> I think it's fair to say that the FB3 is an FB/PDX-specific success; and not directly transferrable to start-ups
[8:09pm] <vagrantc> the gnome fork definitely adds a lot of complication, but is leading to a better infrastructure.
[8:09pm] <tymp> but, i do want to the non-locals to not lose sight of all the work pdx has done
[8:09pm] <skippy> PDX has a lot of investment, and it feels to me that they take a lot for granted.
[8:10pm] <skippy> that's neither good nor bad, by the way -- it's just a statement of fact.
[8:10pm] <vagrantc> skippy: any specific examples?
[8:10pm] <tymp> it's to be expected, i think
[8:10pm] <skippy> errr.. perception.
[8:10pm] <tymp> (skippy): i agree, on the whole
[8:11pm] <tymp> well, it is beginning to move -
[8:11pm] <skippy> vagrantc: this discussion shows several examples, vagrantc, as folks talk about build and adoption and teaching.
[8:11pm] <tymp> i think there has not been real motivation to open it up
[8:11pm] <skippy> We have no such hierarchy / definition, so we don't have the same kinds of approach to issues.
[8:11pm] <vagrantc> skippy: definitely.
[8:12pm] <tymp> vagrantc : an example is the installation process - which until recently was assumed to be based on cloning
[8:12pm] <tymp> ("assumed" - in practice)
[8:13pm] <WormRunner> and there has been motion for a long time toward a more standard install
[8:13pm] <vagrantc> basically, freegeek portland had a lot of stuff that basically worked, but is not necessarily easy to replicate or maintain over the long-term, and coordinating with other groups is forcing the project into a good direction.
[8:13pm] <skippy> decision making seems to be "the PDX way, or the highway", to be blunt.
[8:13pm] <WormRunner> good grief
[8:13pm] <tymp> skippy : i think that is a mis-perception.  but i have had that perception too.
[8:14pm] <vagrantc> well, the meeting time is portland-centric ...
[8:14pm] <tymp> i think the central problem is the ease of misunderstanding, rather than any specific inattention or disregard, etc
[8:14pm] <skippy> tymp: that may be the bulk of it.
[8:14pm] <skippy> anyway, do we have any more comments re: prognosis?
[8:15pm] <vagrantc> basically, any group is going to be resistant to changing something that is basically working for them.
[8:15pm] <tymp> i think it's important to respect the special "ownership" of portland over freegeek - if that word is not too ugly
[8:15pm] <tymp> vagrantc : it's just social inertia
[8:15pm] <WormRunner> It seems to me that there is a great deal of movement by portland, and that these meetings are evidence of that
[8:16pm] <vagrantc> tymp: it's more than merely social- it is technical inertia, also.
[8:16pm] <vagrantc> WormRunner: it feels to me that portland is less actively involved, at the moment, actually.
[8:16pm] <tymp> vagrantc : by 'just', i mean, just inertia - but i take it back
[8:17pm] <skippy> WormRunner: I hope you're not offended by my comment. I'm not trying to cause trouble, or whine. I'm trying to be honest, because this is something about which I'm passionate (as I assume are all of you).
[8:17pm] <skippy> are there any action items for this prognosis issue?
[8:17pm] <vagrantc> skippy: again, i'd be curious about more of the specifics ?
[8:17pm] <skippy> vagrantc: I suggest we put this on hold for the mailing list.
[8:17pm] <WormRunner> skippy: I am troubled by what I see as an adversarial approach
[8:17pm] <vagrantc> skippy: sounds perfect.
[8:18pm] <WormRunner> skippy: yes
[8:18pm] <skippy> WormRunner: feel free to expand your concerns on the mailing list. I don't think I'm being adversarial.
[8:18pm] <skippy> item 2c) popcon server.
[8:18pm] <skippy> This was my item. I've taken no action on it.
[8:19pm] <skippy> Commit: install and test a popcon server before next meeting.
[8:19pm] <tymp> sweet.  that was quick!
[8:19pm] <skippy> item 2d) multi-profile CD.
[8:19pm] <vagrantc> haven't done much on this- last CD build was 699MB, including a lot of the freekbox3-gnome trimming.
[8:20pm] <tymp> what's the date on that?
[8:20pm] <skippy> vagrantc: ought there be a distinction between a "user" multi-profile CD, and a "FreeGeek on a disc" multi-profile CD?
[8:20pm] <vagrantc> september 3rd.
[8:20pm] <gate> I made some non release CDs this week which seemed to work allright but I have not gotten back to them since sunday
[8:20pm] <vagrantc> skippy: maybe. ideally we can fit it all on one CD.
[8:20pm] <tymp> the 8/23 one has some problems - but i didn't feel the need so much to comment, as i'm thinking it's still ... alpha stage?
[8:21pm] <tymp> vagrantc : did you get my comment about the broken cd link on the wiki?
[8:21pm] <vagrantc> skippy: there is so much overlap, i don't see the advantage of the extra overhead required for two CDs.
[8:21pm] <vagrantc> tymp: nope.
[8:22pm] <tymp> in general : what is the intended product of work on multi-installer cd?
[8:22pm] <skippy> vagrantc: I disagree with the need to fit it all on one CD. Many users are not likely to be interested in replicating an FG for themselves. It seems a better investment of our time to provide users with a very focused CD; and give interested parties the extra tools to do what they need to make an FG.
[8:22pm] <gate> ya, the only reason to go two CD is if the packages grow too much
[8:23pm] <tymp> skippy : that is rather what i'd been thinking
[8:23pm] <WormRunner> I have thought it would be good to have a basic CD but a package to install an expanded set of relevant packages
[8:23pm] <vagrantc> skippy: i guess we disagree on what's a better investment of time.
[8:23pm] <stillflame> we currently need to install a few different set of packages: laptops, baseservers and freekboxes.
[8:23pm] <gate> The first time a real package comes up for dropping or adding and size become a consern make two CDs, its pretty easy with simple-ccd
[8:23pm] <stillflame> we = fg.pdx
[8:23pm] <vagrantc> gate: we're already riding on the edge of a 700MB image, though.
[8:23pm] <gate> I know
[8:24pm] <tymp> vagrantc : i still don't see clearly why it's desirable to have a single cd image with all the profiles
[8:24pm] <vagrantc> of course, according to user-oriented design, we should have a separate CD for every persona :)
[8:24pm] <tymp> lol
[8:25pm] <gate> less coasters laying around
[8:25pm] <tymp> but there should be one persona, so, for each of one ;)
[8:25pm] <tymp> gate : laying around *where*
[8:25pm] <gate> my coffee table
[8:25pm] <vagrantc> tymp: well, i don't want to have to maintain two separate images if 90% of it is overlap.
[8:25pm] <skippy> perhaps the FG-on-a-disc could even be a mini-CD?
[8:25pm] <tymp> will portland need to use this in-house?
[8:25pm] <tymp> skippy : business-card
[8:26pm] <skippy> tymp: sure, or thar.
[8:26pm] <skippy> *that
[8:26pm] <WormRunner> the CD is useful even if portland continued to clone
[8:26pm] <skippy> WormRunner: can you elaborate?
[8:27pm] <gate> vagrantc: as long as we maintain good simple-ccd profiles maintaining them is pretty easy
[8:27pm] <tymp> WormRunner : i'm wondering who is the intended user?
[8:27pm] <WormRunner> We grant boxes to people and organizations who may not be in portland and at some point need to reinstall
[8:27pm] <vagrantc> gate: it's still more maintainence, and i would like to stick them all on one CD. but if we can't, oh well.
[8:27pm] <tymp> i think the cd's highest value would be... 
[8:27pm] <tymp> WormRunner : exactly what i'm thinking
[8:27pm] <WormRunner> coming back to town is a real hassle
[8:28pm] <gate> I think the aim is one CD but if we can't fit it then o well
[8:28pm] <tymp> i think reinstall is the highest value of the cd
[8:28pm] <WormRunner> and one cd would be best if at all possible
[8:29pm] <skippy> have any packages been marked for removal from the original FreekBox3 build, to squeeze out some extra room?
[8:29pm] <WormRunner> even if it means giving up wesnoth ;-)
[8:29pm] <vagrantc> has anyone actually used two CDs with debian? it's kind of a pain, honestly.
[8:29pm] <skippy> or is the current FB3 locked-in, package wise?
[8:29pm] <tymp> i think aside from reinstallation without ethernet, the cd will not be hugely needed
[8:29pm] * vagrantc does the give up wesnoth dance!
[8:29pm] <tymp> vagrantc : but you can assume broadband
[8:29pm] <WormRunner> lol
[8:29pm] <vagrantc> tymp: that, is a very stupid assumtion.
[8:30pm] <tymp> vagrantc : how come?
[8:30pm] gate (n=gate@cpe-24-210-70-207.columbus.res.rr.com) is now known as {gate}
[8:30pm] <WormRunner> very few of our adopters have broadbanc
[8:30pm] <WormRunner> d
[8:30pm] <vagrantc> tymp: well, even if you have broadband, installing from a CD is *way* faster than a basic DSL connection.
[8:30pm] <tymp> vagrantc : but very few of your adopters will need to install lessdisks
[8:30pm] <skippy> when I said earlier "two CDs", I didn't mean a CD set, a la Debian install CDs. I meant two standalone CDs.
[8:31pm] <skippy> one CD for users, from which they could rebuild their system.
[8:31pm] <vagrantc> skippy: well, for an application server, nearly all of the freekbox3 packages are useful.
[8:31pm] <tymp> vagrantc : what i mean is, the multi-install is not needed by dialup users
[8:31pm] <vagrantc> tymp: of course not.
[8:31pm] <skippy> one CD for FG folks, who want to have a terminal server, and FGdb, and thorough documentation, etc
[8:33pm] <skippy> vagrantc: We don't use an application server (yet), so the utiilty of bundling that on a CD for users is lost on me.
[8:34pm] <tymp> .. so the overall objective is ill-defined
[8:34pm] <skippy> tymp: I'd concur with that.
[8:34pm] <tymp> ugh, too many variables
[8:34pm] <vagrantc> skippy: and having a couple extra menu items is totally unacceptable?
[8:34pm] <tymp> but i am so happy vagrantc has been working on it!
[8:35pm] <skippy> vagrantc: if the target user is not intended to use those menu options, they have no business being present if it's in our control to decide.
[8:35pm] <vagrantc> skippy: i guess, if you want to maintain a separate CD that only has freekbox3 packages, you are welcome to.
[8:35pm] * WormRunner reserves judgement
[8:36pm] <vagrantc> (or freekbox3-gnome, or whatever).
[8:36pm] <tymp> lets step back and clarify the goals - on the list
[8:36pm] <tymp> eh, distro list
[8:36pm] <skippy> well, that's a fine example of "PDX way or the high way".
[8:37pm] <vagrantc> skippy: i live in new mexico at the moment, and am interested in propegating freegeeks everywhere.
[8:37pm] <tymp> skippy : i think it is an example of vagrant is doing all the work so his perspective deserves a little more weight
[8:37pm] <WormRunner> vagrantc is a pdx refugee!
[8:37pm] <tymp> oh jeez  ;)
[8:37pm] <WormRunner> tymp: I agree
[8:37pm] <tymp> edit: * almost all the work
[8:37pm] <skippy> propogating FGs is a different issue than making CDs for users. It was my understanding that it was specifically this issue that was under discussion.
[8:37pm] <WormRunner> lol
[8:38pm] <vagrantc> skippy: people building freegeeks are users.
[8:38pm] <tymp> skippy : everything is prapagation!
[8:38pm] <skippy> tymp: I'd very much like to contribute to the work, but am finding it challenging to find the right entry vector for my current skillset.
[8:38pm] <tymp> skippy : i'm there too at the moment.  
[8:38pm] <stillflame> vagrantc: are you suggesting a personae of "Freegeek Startupist"?
[8:38pm] <{gate}> it shoudl getter beter when we have a place to store code
[8:39pm] <vagrantc> skippy: as gate said, it is not difficult to create extra CD images with simple-cdd ...
[8:39pm] <skippy> if we're all going to simply defer to vagrant, then this entrire agenda item can be entirely excised.
[8:39pm] <vagrantc> stillflame: perhaps i am.
[8:39pm] <WormRunner> that would be an interesting persona
[8:39pm] <tymp> {gate} : and a good way to get folks into that - which will follow organically i spose
[8:39pm] <tymp> {gate} : once it's there
[8:39pm] <vagrantc> i have no intention of rail-roading anything.
[8:40pm] <{gate}> I think we are a bit off topic here
[8:40pm] <WormRunner> could be
[8:40pm] <tymp> skippy : no, just i expect vagrantc has thought about this more than the rest of us
[8:40pm] <skippy> are there any actions to be taken, here?
[8:40pm] <stillflame> skippy: right, we need to decide what balance of work and options we want. vagrantc: are you only capable of handling one installer cd? if so, we need to decide what cd would best serve all freegeeks right now.
[8:41pm] <tymp> stillflame : i intend to get up to snuff on it in oct
[8:41pm] <stillflame> okay then, all of us: how many cds can we maintain?
[8:41pm] <tymp> .. i really think this will all make sense if we have some discussion about the intended purpose(s)
[8:42pm] <vagrantc> it is not impossible to make two CD images, but the most crude test cycle of each CD build takes 4-8 hours. i don't want to double that.
[8:42pm] <tymp> ah!
[8:42pm] <stillflame> tymp: yeah, i agree. i think we should ignore making more cds until we have things we're sure we want to add to the cds
[8:42pm] <vagrantc> so someone else can do it, and i will happily help them get up to speed with simple-cdd to do it.
[8:42pm] <{gate}> columbus plans on testing these things too. curned out two on sunday and tested both.
[8:42pm] <tymp> and it's quicker to do on real hardware
[8:43pm] <tymp> than emulated
[8:43pm] <vagrantc> tymp: but makes a lot of CD coasters.
[8:43pm] <stillflame> vagrantc: can you commit to making some sort of documentation on how to do that build cycle?
[8:43pm] <vagrantc> stillflame: sure!
[8:43pm] <tymp> vagrantc : use rewritables
[8:43pm] <skippy> rewritable CDs should prevent coasters.
[8:43pm] <tymp> jinx!
[8:43pm] <skippy> darnit!
[8:43pm] <tymp> heh
[8:44pm] <vagrantc> tymp: i don't have access to any CD burners
[8:44pm] <skippy> commit: take to the distro list the question of objectives for CD images.
[8:44pm] <stillflame> vagrantc: freegeek.pdx would happily donate one
[8:44pm] <tymp> vagrantc : nor a spare box, i assumed - was thinking, here's how others can help
[8:44pm] <vagrantc> stillflame: i will soon be too mobile to lug around extra hardware.
[8:44pm] <stillflame> mmm
[8:45pm] <tymp> stillflame : grant vagrantc a new laptop w/ burner :)
[8:45pm] <WormRunner> yeah!
[8:45pm] <vagrantc> so it sounds like a lot of this discussion might be better had on the list?
[8:45pm] <stillflame> that's possible
[8:45pm] <skippy> who else has commitments for this, before we move on?
[8:45pm] * vagrantc committed to something ... ?
[8:46pm] <tymp> we have fifteen minutes, and we haven't checked in on past commitments
[8:46pm] <stillflame> vagrantc: docs on freekbox3 build cycle
[8:46pm] <vagrantc> commit: document freekbox3 installer CD build cycle
[8:46pm] <skippy> item 2e) SVN
[8:46pm] <skippy> status report from PDX?
[8:46pm] <tymp> i am concerned we aren't going to go over past commitments
[8:47pm] <tymp> can we do that quickly?
[8:47pm] <vagrantc> tymp: there is considerable overlap with existing topics
[8:47pm] <stillflame> svn server is currently being put into the dmz, after which it will be accessable globally. still some configuration issues to work out with buildbot and such.
[8:47pm] <tymp> vagrantc : but to have it in one place
[8:47pm] <tymp> stillflame : eta?
[8:47pm] <vagrantc> tymp: i think there will be time.
[8:47pm] <{gate}> so when can we start getting accounts
[8:48pm] <stillflame> i'll work on it all tomorrow night, and i think that will be enough.
[8:48pm] <tymp> well, since we started... meho?
[8:48pm] <tymp> oh.. that's real old
[8:48pm] <tymp> oh, i'm looking at the wrong agenda
[8:48pm] <vagrantc> ?
[8:49pm] <tymp> gate?
[8:49pm] <stillflame> tymp: hold on while i finish with svn
[8:49pm] <tymp> ah, sorry
[8:49pm] <stillflame> it will be set up for both webdav and svn+ssh, and i'll need to get login info for everyone
[8:49pm] <stillflame> trac is installed, and will need to be configured for each project hosted
[8:50pm] <{gate}> so that means u want our pub keys
[8:50pm] <stillflame> {gate}: right
[8:50pm] * tymp apologizes for being on another planed for a moment
[8:50pm] <stillflame> if people send things to me now, i can add them in once things are up and tell everyone about it.
[8:51pm] <stillflame> so carryover commitment for me on svn
[8:51pm] <stillflame> COMMIT: finish svn, add user access (either ssh or webdav)
[8:51pm] <stillflame> oh, and i need to remember to set up ssl certs
[8:51pm] <skippy> okay. previous commitments.
[8:52pm] <vagrantc> stillflame: including anonymous access?
[8:52pm] * tymp meant "planet"
[8:52pm] <skippy> meho: popcon. still pending. gate: setup fb3-gnome packages -- in process. stillflame: request for coordinator; unnecessary now.
[8:52pm] <skippy> errr, not meho; me. sorry.
[8:53pm] <stillflame> vagrantc: anon is working, yeah
[8:53pm] <{gate}> yup, thats the status of the gnome package.
[8:53pm] <skippy> vagrantc: post about new "official" fb3 packages; post about restructuring. document reprepro.
[8:53pm] <stillflame> skippy: and i did it over email, i think.
[8:53pm] <vagrantc> flaked on all accounts.
[8:53pm] <vagrantc> we kind of discussed the main issues here for the packaging/restructuring.
[8:53pm] <stillflame> we all read the zen paper, though
[8:53pm] * stillflame applauds us.
[8:54pm] <skippy> tymp: media player on wiki?
[8:54pm] <tymp> no.  flaked.  but also, i don't know if that's the right place for it
[8:54pm] <tymp> i can make a stub for documenting options, but it really seems like a discussion topic 
[8:54pm] <WormRunner> Is that a discussion of the relative merits?
[8:54pm] <tymp> yes
[8:55pm] <WormRunner> wiki can be a good place then
[8:55pm] <vagrantc> maybe distro list?
[8:55pm] <tymp> vagrantc : yeah
[8:55pm] * vagrantc hates discussions on wikis
[8:55pm] <tymp> yes, it's not really suitable
[8:55pm] <skippy> the list better ensures that I at least will answer.
[8:55pm] <WormRunner> vagrantc: ok, then on the list
[8:56pm] <tymp> but the wiki can be good for distilling the discussion - if that's desirable
[8:56pm] <WormRunner> yes. I can commit to that part
[8:56pm] <vagrantc> sounds good.
[8:56pm] <tymp> okay, given to WormRunner
[8:56pm] <skippy> any final business?
[8:56pm] <tymp> next meeting
[8:57pm] <skippy> the 27th
[8:57pm] <tymp> agenda deadline? facilitator?
[8:58pm] <tymp> i'll facilitate, and have a rough agenda up by next tues, if no one else does it
[8:58pm] <tymp> k?
[8:58pm] <skippy> deadline should be the friday before the meeting, methinks, at the latest.
[8:58pm] <tymp> that sound good, at the latest
[8:58pm] <vagrantc> tymp: i think commitments maybe should come before on-going issues, in the future (it was my bad organization that started the trend)
[8:59pm] <tymp> yes, makes sense.  np
[8:59pm] <vagrantc> and if we've got tons of extra time, we can always agree to add new items :)
[9:00pm] <skippy> okay... last chance to say anything on the record...
[9:00pm] <tymp> is there an after-party? ;p
[9:00pm] <skippy> </meeting>
[9:00pm] <{gate}> crap
[9:00pm] <tymp> thanks you all
[9:01pm] <{gate}> I mean its bed time
[9:01pm] <{gate}> later
[9:01pm] <tymp> {gate} : cheers
[9:01pm] <skippy> g'night {gate}
[9:01pm] <vagrantc> tymp: what was the thing you wanted to discuss if we had extra time?
[9:01pm] <skippy> communication
[9:01pm] <tymp> "communication"
[9:01pm] <vagrantc> that's where things usually break down, yes.
[9:02pm] <tymp> actually, the thing in my mind was minor/simple: it can be appropriate to have off-list discussion - if it is specific ... workgroup discussion.  eg, i plan to mail you and WormRunner on the .. whatever that was we agreed to work on and document thing
[9:03pm] <vagrantc> package repository building?
[9:03pm] <tymp> though, your policy of encouraging on-list discussion is good - and i think clearly things fall off the list by accident and need to be nudged back on sometimes
[9:03pm] <tymp> yes, that should be it
[9:04pm] <tymp> but wonder if there's a way to talk about it as a broader topic too "how to improve communications (on an ongoing basis)"
[9:04pm] <vagrantc> it's important to not over-whelm the list with lots of information, but at the same time, i think it's even more important to keep everything documented and accessible.
[9:05pm] <tymp> maybe there could be some known system of subject line codes..
[9:05pm] <tymp> to make it easy to filter
[9:05pm] <tymp> like "workgroup: x"
[9:05pm] <vagrantc> i think people should just use meaningful subjects.
[9:06pm] <vagrantc> we all could probably improve that.
[9:06pm] <tymp> shouldcouldwould :)
[9:06pm] <vagrantc> and maybe including the working group in the title is part of that ...
[9:06pm] <WormRunner> being willing to change subjects too -- A (was C)
[9:06pm] * vagrantc nods
[9:07pm] <tymp> and you could say "hi, welcome to blabla, you may want to ignore emails with "signifier" in the subject.."
[9:07pm] <tymp> "..while you get acquainted"
[9:07pm] <tymp> WormRunner : truely
[9:08pm] <vagrantc> well, i need to go harvest, divvy up, and distribute food tomorrow :)
[9:08pm] <skippy> good night.
[9:08pm] <tymp> cool.  enjoy.
[9:08pm] Quits: vagrantc (n=vagrant@pcp05158478pcs.montbl01.nm.comcast.net) ("leaving")
[9:08pm] <tymp> it *feels* late here.  charms of insomnia.
[9:08pm] <WormRunner> yeah, I need to head off as well.
[9:09pm] <tymp> thanks much for coming WormRunner
[9:09pm] <tymp> !
[9:09pm] <WormRunner> my pleasure
[9:09pm] Parts: WormRunner (n=worm_@pool-71-111-140-187.ptldor.dsl-w.verizon.net) ("Now where was I?") [8 users]
[9:09pm] Quits: skippy (n=skippy@two.skippy.net) ("[BX] Wouldn't it be great if the dog could walk itself?")
[9:09pm] <tymp> stillflame : were you posting a log?  i'll put a summary up
[9:10pm] <tymp> ... pretty easy since we followed the agenda so closely
[9:10pm] <stillflame> i will
[9:10pm] <stillflame> COMMIT: post log
[9:10pm] <stillflame> =:)
[9:11pm] <tymp> commit : post summary :)
[9:12pm] * meho commits to attend next meeting
[9:14pm] Quits: romdos (n=romsos@209-162-205-24.dq1mn.easystreet.com) ("I left some time ago but left my computer on")