Summary of yesterdays EPEL SIG meeting

Thorsten Leemhuis fedora at leemhuis.info
Thu Apr 19 16:32:46 UTC 2007


= Meeting 20070419 =

== Attending ==

>From the Steering Committee:

 * dgilmore
 * mmcgrath
 * nirik
 * quaid
 * stahnma
 * thl
 * thimm

Other contributors that joined the meeting:

 * entr0py, rdieter

== Summary ==

 * RHEL5 is now on the builders thx to dgilmore; note that the RHEL5-ppc
tree is based on the RHEL5-server version that that misses some packages
that are part of the RHEL-desktops (which are only available for x86 and
x86_64); see this mail from thimm for details:
http://www.redhat.com/archives/epel-devel-list/2007-March/msg00250.html
 of the differences

 * Mass rebuild for RHEL5 final -- we give contributors a chance to
rebuild their stuff over the next week until the next SIG meeting; arch
packages should be build, noarch don't have to; we discuss in the next
meeting what to do with those packages that didn't get rebuild

 * Elect chairmen and discuss responsibilities  -- six out of seven
Steering Committee members want a chairmen; we discuss responsibleness,
expectations and powers of the next week on the list

 * Relations to 3rd party distro  -- hard to write a summary; see 0:44
an later for details;

== Full Log ==

{{{
00:00            --- | thl has changed the topic to: EPEL meeting
00:00 <         thl> | ping dgilmore quaid stahnma thimm nirik mmcgrath
00:00              * | nirik is here.
00:00              * | mmcgrath here
00:01              * | thimm also
00:02 <     entr0py> | rabble rabble
00:02 <         thl> | hmm, that makes four people
00:03 <         thl> | well, maybe others show up over time, so let's start
00:03            --- | thl has changed the topic to: EPEL Meeting --
RHEL5 on the builders -- dgilmore/mmcgrath
00:03 <         thl> | dgilmore told be we have RHEL5 on the builders now
00:03 <    mmcgrath> | AFAIK we're good to go.
00:03 <         thl> | but he said something else:
00:03 <    mmcgrath> | We still don't have a good way to keep the
builders up to date.
00:03 <         thl> | "my only concern is that some ppc builds may fail
since there is only a server version for ppc"
00:04 <         thl> | mmcgrath, k, I'll put that up the schedule to
make sure it doesn't get forgotten
00:04 <       thimm> | mmcgrath: what is the issue?
00:04 <         thl> | but I'd say it should be fine for now
00:04 <    mmcgrath> | thimm: actually getting the RPM's to a location
to sync to in an automated fashion.  We could use satsync but thats nasty.
00:04 <     entr0py> | so go ahead and build but look for problems?
00:04 <    mmcgrath> | did RH not release a PPC client version?
00:05 <         thl> | entr0py, yes
00:05 <         thl> | mmcgrath, no, they didn't
00:05            --> | stahnma (Michael Stahnke)  has joined
#fedora-meeting
00:05 <         thl> | so some packages might be missing
00:05 <         thl> | and builds might fail
00:05 <         thl> | is " go ahead and build but look for problems "
okay for everyone?
00:05              * | stahnma is here (meeting moved)
00:05 <         thl> | I asked dgilmore to provide a "ls -l" of the trees
00:06 <         thl> | then we can analyse of what to do
00:06 <    mmcgrath> | thl: yep
00:06 <       thimm> | Out of client, the only thing I would be missing
is fribidi
00:07 <       thimm> | So not a very big deal
00:07            --> | silug (S. Ill. Linux UG - http://www.silug.org/)
 has joined #fedora-meeting
00:07 <    mmcgrath> | heh, thats funny.
00:07 <         thl> | thimm, maybe, I'm not sure of the exact differences
00:07            --> | FrancescoUgolini (Francesco Ugolini)  has joined
#fedora-meeting
00:07 <    mmcgrath> | We can always compile that on our own.
00:07 <       thimm> | thl, I have the diffs here
00:07 <       nirik> | so we are going to tell people to rebuild? or
just mass rebuild?
00:07 <         thl> | thimm, can you send them my way please? tia!
00:08            --- | thl has changed the topic to: EPEL Meeting --
mass rebuild for RHEL5 -- thl
00:08 <         thl> | nirik, well, I'd say we go further as planed last
friday
00:08            --> | rdieter (Rex Dieter)  has joined #fedora-meeting
00:08 <         thl> | e.g. tell contributors to rebuild their stuff
00:08 <         thl> | we wait 3|4|7 days
00:08 <         thl> | and then we add a .1 to every pacakge that didn#t
get rebuild
00:08 <       thimm> | thl: they are on the epel-devel list ...
00:08 <         thl> | commit, tag, build
00:09 <    mmcgrath> | Playing devil's advocate: Would it be horrible if
we didn't require a rebuild and just let people file bugs if any are found?
00:10 <         thl> | mmcgrath, I'd prefer a clean build
00:10 <         thl> | thimm, "they are on the epel-devel list" -> what
do you mean?
00:10 <    mmcgrath> | I would to, but its not as simple as snapping
ones fingers :-/
00:10 <         thl> | thimm, I wanted to annouce it on epel-devel list
00:10 <       thimm> | thl and rest: diffs server/client on
http://www.redhat.com/archives/epel-devel-list/2007-March/msg00250.html
00:10 <       thimm> | already done
00:10 <       thimm> | if you like so
00:10              * | dgilmore is somewhat here
00:11 <       thimm> | mmcgrath: is also an option
00:11 <         thl> | I got told a lot got changed, z00dax said that
and suggested we rebuild everything
00:11              * | quaid is here now
00:11 <       thimm> | I can't imagine RHEL5beta2 being further away
than RHEL5, than FC6-F7
00:12 <         thl> | thimm, it was RHEL5beta1...
00:12 <       thimm> | Even so
00:12 <    mmcgrath> | thl: I'd defer to z00dax on that.  He'd know
better than I.
00:12 <       thimm> | (who is zoodax?)
00:12 <    mmcgrath> | thimm: centos.karan.org
00:12 <         thl> | thimm, from centos and http://centos.karan.org/
00:12 <       thimm> | Karanbir Singh?
00:13 <    mmcgrath> | <nod>
00:13 <         thl> | thimm, yes
00:13 <       thimm> | k
00:13 <     stahnma> | do we expect a lot of people to not rebuild if we
ask them to nicely?
00:13 <    dgilmore> | stahnma: most will
00:13 <    mmcgrath> | especially if we start to nag.
00:13 <         thl> | stahnma, it will be a whole lot quicker if we do
it outselfes after some days
00:13 <    mmcgrath> | I'm fine with either.
00:14 <     stahnma> | so, a time limit of 5 or 7 days seems reasonable
00:14 <         thl> | I'd say we give them some days, and then to it
ourselfs
00:14 <         thl> | +1 for 5 to 7 days
00:14 <     stahnma> | if they can't meet that, they can always select a
co-maintainer also
00:14 <       thimm> | I would be upset if I get my spefiles forked ...
00:14 <     stahnma> | some people are very concered about forked spec
files and some seem not to be, what is the general consensus?
00:14 <         thl> | thimm, yeah, but 7 days should be enough normally
00:15 <    mmcgrath> | thimm: I feel the same way but we can't say we
didn't warn them :-)
00:15 <         thl> | mmcgrath, +1
00:15 <     stahnma> | if we announce today, and put a due date of next
wednesday, how would that be?
00:16 <     stahnma> | we can have a status at this meeting next week?
00:16 <         thl> | stahnma, should be fine
00:16 <     stahnma> | pick up stragglers
00:16 <       thimm> | What happens if I know a package of mine needs no
rebuild
00:16 <         thl> | stahnma, I can annouce the rebuild later
00:16 <     stahnma> | then and beat them with 1s and 0s
00:16 <       thimm> | For example it is a python app
00:16 <         thl> | I actually started to prepare a mail already
00:16 <    dgilmore> | thimm: noarch could be excluded
00:17 <         thl> | dgilmore, +1
00:17 <       thimm> | Can we leave that to packager's discretion?
00:17 <     stahnma> | does that make sense in all cases?
00:17 <     stahnma> | I am just asking becuase I am not 100% sure
00:17 <    dgilmore> | noarch should need no compilation and work
regardless
00:17 <       thimm> | It need to be checked on a package by package basis
00:17 <     stahnma> | that being dgilmore's noarch exclusion above :)
00:17 <         thl> | thimm, are you volunteering to do this?
00:18 <       thimm> | What?
00:18 <    dgilmore> | if it really needs a rebuild it should not be no
arch
00:18 <         thl> | otherwise we'll never get this of the table
00:18 <         thl> | thimm, look "on a package by package basis"
00:18 <       thimm> | Yes, for my packages ...
00:18 <       thimm> | "Can we leave that to packager's discretion?"
00:18 <     stahnma> | that brings us back to a voluntary rebuild from
each contributor though right?
00:19 <       thimm> | yes
00:19              * | thl wonders thy this comes up now, as the
decisions to have a rebuild was done weeks ago
00:19 <       thimm> | Consider it the following way:
00:19 <    dgilmore> | thimm: they could try sell us on why it doesnt
need a rebuild
00:19              * | stahnma was also thinking that
00:19 <         thl> | we could have saved a lot of trouble if we would
have decided to ahve a colunettered mass rebuild only
00:19 <    mmcgrath> | Could we ask a voluntary rebuild and meet about
it next week to decide what to do next?
00:19 <         thl> | we could use the "needs.rebuild" cvs trick again
00:20 <         thl> | mmcgrath, +1
00:20 <       nirik> | mmcgrath: +1
00:20 <    mmcgrath> | I just want something to happen in epel, it feels
like no actual action has taken place in weeks :-/
00:20              * | nirik has several noarch packages that are very
unlikely to be needing rebuild.
00:20 <     stahnma> | mmcgrath: ++
00:20 <       thimm> | mmcgrath: I think dgilmore is doing a lot of work
00:20 <         thl> | that makes four for "ask a voluntary rebuild and
meet about it next week to decide what to do next"
00:20 <    mmcgrath> | he got the RHEL5 builders going yes.
00:20 <    dgilmore> | mmcgrath: thats fine
00:21 <         thl> | five
00:21 <    mmcgrath> | dgilmore aside though, nothing has happened :-/
00:21 <     stahnma> | yay dgilmore !
00:21 <         thl> | I'll annouce that on the list then if that's okay
for everybody
00:21 <       thimm> | thl and I fight every day, does that count? ;)
00:21 <     stahnma> | thl +1
00:21 <         thl> | thimm, we leave out some days
00:21 <     stahnma> | it adds to my amusement :)
00:21              * | dgilmore is unlikely to be  able to be here the
next 2 weeks
00:21 <         thl> | thimm, today iirc ;-)
00:22 <         thl> | (well, until now at least)
00:22 <       thimm> | day is still young
00:22 <    mmcgrath> | :)
00:22 <       nirik> | perhaps some days count double. ;)
00:22 <         thl> | anyway, anything else?
00:22 <       thimm> | OK, 4 of 7 have agreed, go ahead
00:22 <         thl> | or move on?
00:22 <    dgilmore> | move on
00:22 <       nirik> | moveon++
00:22 <    mmcgrath> | Ok, so that email will get sent and then we will
discuss next week what to do.
00:22            --- | thl has changed the topic to: EPEL Meeting --
Elect chairmen and discuss responsibilities
00:22 <         thl> | thimm, spot is the defacto chairmen of FPC
00:23 <         thl> | thimm, he actually started that group and is
responsible for it afaik
00:23 <       thimm> | He doesn't veto votes ...
00:23 <         thl> | afaik
00:23 <         thl> | thimm, the chairmen can't veto votes
00:23 <       nirik> | wheres the wiki page with proposed duties of the
chair?
00:23            --> | GeroldKa (GeroldKa)  has joined #fedora-meeting
00:23 <     stahnma> | I don't think anyone was proposing veto of votes
00:23 <         thl> | he  has just to coordinate them
00:23              * | nirik is unconvinced we need one.
00:23            <-- | MauricioPretto has quit ("Leaving")
00:23 <       thimm> | nirik++
00:23 <    dgilmore> | thimm: chairperson is to make sure things are
happening
00:23 <       thimm> | what would a chairman solve?
00:23 <         thl> | last week the decisions was to have one
00:23 <       thimm> | ???
00:24 <       thimm> | What decision?
00:24 <    dgilmore> | thimm: it would solve things seeming to be
disorganised
00:24 <         thl> | that was afaics the conclusion of last week
00:24 <     stahnma> | I thought it was more of an administrivia role
and not so much formal power
00:24 <    mmcgrath> | a chairman would be someone that has a vision and
can make decisions.
00:24 <         thl> | stahnma, +1
00:24 <    dgilmore> | well because they are
00:24 <    mmcgrath> | If we don't like his decisions, we don't re-elect
him.
00:24 <     stahnma> | and beat him!
00:24 <       thimm> | So it is an empowerment position, not administrative
00:25 <       thimm> | E.g. not a chairman bit a president
00:25              * | thimm don't like that
00:25 <    mmcgrath> | The fact is even the few people we have involved
right now are bumping heads regularly.
00:25 <    dgilmore> | thimm: bit of both
00:25 <         thl> | dgilmore, +1
00:25 <    dgilmore> | sometimes things just need doing
00:25 <       thimm> | That's why we vote
00:25 <    mmcgrath> | thimm: voting is slow.
00:25 <       thimm> | We din't get off because there was reluctance to
a committee
00:25 <    mmcgrath> | and voting on everything shows a lack of vision.
00:26 <       thimm> | So fesco/fpc have a lack of vision?
00:26 <    mmcgrath> | I'm not talking about fesco or fpc.
00:26 <    mmcgrath> | I'm talking about EPEL, specifically the
bickering that has started over the last couple of months that is A)
counter productive and B) making sure that nothing gets done.
00:27 <       thimm> | And a chairman will stop flames on the list?
00:27 <    dgilmore> | mmcgrath: indeed
00:27 <    mmcgrath> | Not everyone can be happy all the time.
00:27 <    mmcgrath> | a chairman will make sure the flames only last a
day or so.
00:27 <       thimm> | And then? what does he do?
00:28 <    mmcgrath> | people on the list are treating every little
decision as if they don't get their way means that epel will cease to
exist in a few weeks and thats just not the case.
00:28 <       thimm> | He contacts the mailman admin to shutdown the
thread?
00:28 <       thimm> | Or he make a decision on his own?
00:28 <    mmcgrath> | No, he makes the decision and tries to find
someone to take action unless he takes the action himself.
00:28 <     stahnma> | ask to present your case formally, and have both
sides present
00:28 <       nirik> | or he brings it to a vote?
00:28 <    mmcgrath> | he can bring it to a vote if he feels wants.
00:28 <     stahnma> | a well informed EPEL group then makes the call
(vote possible)
00:28 <       nirik> | so this is more of a manager position that a
admin of meetings and such?
00:28 <    mmcgrath> | we talked about only having a chariman for 6
months anyway right?
00:29 <    mmcgrath> | so lets just do it and see where we stand in 6
months.
00:29 <         thl> | mmcgrath, yes
00:29 <         thl> | mmcgrath, +1
00:29            --- | couf_afk is now known as couf
00:29 <       quaid> | nirik: facilitator
00:29 <       quaid> | that's the main role of chair, IME
00:29 <    dgilmore> | thimm: im the mailman admin for epel  im not
going to censor any threads
00:29 <       quaid> | and occasionally person to make decisions when a
decision needs to be done and a consensus is not clear, yet no formal
vote is needed
00:30 <       nirik> | well, if he's making decisions by himself (or
herself) then they are more than a facilitator...
00:30 <       quaid> | ok, then
00:30 <       quaid> | 70% facilitator
00:30 <       thimm> | quaid: And *who* decides whether a formal vote is
needed?
00:30 <         thl> | nirik, we have this committee for decisions
00:30 <       quaid> | 29% front face
00:30 <       quaid> | 1% decider
00:30 <       thimm> | decider -----
00:30            <-- | FrancescoUgolini has quit (Remote closed the
connection)
00:30 <       thimm> | Let's look at the history
00:31 <       thimm> | chairman is supposed to solve thl's and my
diagreements?
00:31 <       quaid> | thimm: I've never made a decision as a chair that
wasn't from the FDSCo members asking me to decide
00:31 <    dgilmore> | thimm: i dont know if anyone can solve them
00:31 <       thimm> | s/solve/end/
00:31 <       nirik> | I guess I would be ok with that, as long as their
decisions could be overruled/revisted...
00:31 <    mmcgrath> | They won't end or solve them.
00:31 <       quaid> | thimm: what is your suggestion as to how to break
stalemates?
00:31 <    mmcgrath> | but they'll make sure it doesn't get in the way
of epel progress.
00:31 <         thl> | mmcgrath, +1
00:31 <       thimm> | quaid: votes
00:32 <       quaid> | chair is to help move things forward
00:32 <       thimm> | The issues last week or so was that thl didn't
even want my proposed votes to get on the table
00:32 <    mmcgrath> | IMHO, someone like a chairman would be perfect to
get EPEL off the ground.  THen in 6 months we can re-evaluate.
00:32 <       quaid> | thimm: how do you know when to stop debate and
have a vote?
00:32 <       thimm> | I don't see how having thl as a chair will help
there
00:32 <    mmcgrath> | then don't vote for him.
00:32 <       quaid> | thimm: in that case, it was more like ....
00:33 <       quaid> | from my perspective ...
00:33              * | quaid tries to think how to put this
00:33 <    dgilmore> | thimm: the cair will be responsible for
coordinating all vote's  as proposed by everyone
00:33 <       quaid> | thimm: without a fair chair to mitigate, it felt
to me as a committee member that a vote was forced that I was never
clear was needed.
00:33 <       thimm> | Let's show that votes can be effective
00:33 <       quaid> | at least with a chair, I could vote for that
person and trust their judgment to know when to vote or not vote.
00:33 <       thimm> | Let's just vote on whether we want a chair or not
00:34 <         thl> | chair +1
00:34 <    dgilmore> | +1 for a chair
00:34 <       thimm> | -1
00:34 <    mmcgrath> | Ok,  Vote on the table: do we want a chair
00:34 <    mmcgrath> | +1 for chair for 6 months.
00:34 <     stahnma> | +1 chair
00:34 <       quaid> | +1 for chair for 6 months (trial)
00:34 <       thimm> | OK, vote's done, discussion is over
00:34 <       nirik> | +1 as long as we can override or revist their
decisions
00:34 <       quaid> | again, the biggest value is not in voting
00:34 <       quaid> | it is in having a facilitator who is fair
00:35 <     stahnma> | quaid: +1
00:35 <    mmcgrath> | quaid: exactly.
00:35 <       quaid> | if they are not fair, we call them on it - "no,
we won't stop discussing, you are wrong"
00:35 <    dgilmore> | so we will have a chair for 6 months
00:35 <    dgilmore> | stahnma, and thl have been nominated
00:35 <    dgilmore> | anyone else wish to nominate someone
00:36 <    mmcgrath> | Who are the current nominees?  thl and?
00:36 <    dgilmore> | stahnma
00:36 <       thimm> | What are the powers of the cahir now?
00:36 <       thimm> | People here have different POV on that.
00:36 <       quaid> | good question
00:37 <         thl> | thimm, I wrote something to the list yesterday
00:37 <    mmcgrath> | I'll nominate thimm.  He's passionate about this
and should be in the running.
00:37 <       thimm> | what you wrote was purely admin stuff
00:37 <       quaid> | do we need to have a week's worth of discussions
first?
00:37 <         thl> | I added that to the wiki as proposal at
http://fedoraproject.org/wiki/EPEL/SteeringCommittee
00:37 <       nirik> | yeah, no decision making in that list
00:37 <       thimm> | mmcgrath: Thanks, but I cannot step into
something I argue shouldn't exist
00:37 <       quaid> | ah, the irony!
00:38 <    mmcgrath> | thimm: totally fine, I'll remove the nomination.
 Thought I'd throw it out there if you were interested ;-)
00:38 <       thimm> | No, thanks, I'm not
00:38 <       thimm> | My nominee was "no chairman", he's out ;)
00:38 <    mmcgrath> | So powers of the chair.
00:39 <         thl> | so, elect a chairmen today or next week?
00:39 <       thimm> | "issue and coordinate the votings in the meetings
and in the wiki, to prevent chaos that might arise if anybody would
issue votings"
00:39 <       thimm> | -1000
00:39 <    mmcgrath> | thl: who's going to vote on the chair?  Just us
or are we going to use the voting app?
00:39 <       thimm> | Anyone in the SC should be able to issue votings
00:39 <         thl> | mmcgrath, I'd say us
00:39 <    mmcgrath> | k
00:39 <         thl> | FESCo elects its chair itself, too
00:40 <         thl> | thimm, there needs to be some coordination
00:40 <       quaid> | yeah, *SCo elects its own chair
00:40 <         thl> | the last wiki votings showed why IMHO
00:40 <       thimm> | So the chairman can mute SC memebers?
00:40 <         thl> | thimm, no
00:40 <    mmcgrath> | no one can be muted.
00:40 <    dgilmore> | thimm: youu are free to propose them the
chairperson is a conduit
00:41 <       thimm> | So the chairman auto-issues a vote if one of us
proposes one?
00:41              * | spot mutes everyone
00:41 <    mmcgrath> | k
00:41              * | dgilmore slaps spot
00:41 <    mmcgrath> | :)
00:41 <         thl> | thimm, I wrote what I think is important in the wiki
00:41 <         thl> | everything else is not written
00:41 <         thl> | and thuis doesn#t exists imho
00:41 <    dgilmore> | thimm: yes if its proposed it needs to be brought
up for discussion
00:41 <         thl> | so there are no such powers like "So the chairman
auto-issues a vote if one of us proposes one?" afaics
00:41 <    mmcgrath> | thl: lets take a week to figure out exactly what
we're looking for in a chairman and vote next week?
00:41 <       nirik> | thl: what is the wiki page for that?
00:42 <       thimm> | dgimore and thl contradict in the interpretation.
00:42 <       quaid> | mmcgrath: +1
00:42 <         thl> | mmcgrath, fine
00:42 <         thl> | nirik, ?
00:42 <         thl> | nirik, see
http://fedoraproject.org/wiki/EPEL/SteeringCommittee#head-b5f8a5f5e875abbd584ba657548c4f5e9fa9a0d2

00:42 <       nirik> | ah, there it is. Thanks.
00:43 <    mmcgrath> | Would people be opposed to writing up
individually what they're looking for in /wiki/YourWikiName/chair ?
00:43 <       thimm> | You mean 7 lists to choose from?
00:43 <         thl> | mmcgrath, send it to the list
00:43 <    mmcgrath> | k
00:43 <         thl> | mmcgrath, that a whole lot easier imho
00:43              * | mmcgrath was trying to keep the flames down :-)
00:43 <    mmcgrath> | list works though.
00:43 <       quaid> | ... and do it even if you hate the idea
00:44 <         thl> | k, so move on?
00:44 <    dgilmore> | yup
00:44 <     stahnma> | yes
00:44            --- | thl has changed the topic to: EPEL Meeting --
Relations to 3rd party distro
00:44              * | thl send his opinion to the list
00:44            <-- | GeroldKa has quit (Remote closed the connection)
00:45 <       nirik> | would it be possible to invite players from the
other repos to this meeting? or another irc meeting?
00:45 <    mmcgrath> | Do we have relations with any 3rd party repos
besides thimm's?
00:45 <       nirik> | I would like to hear how we might be able to
cooperate more...
00:45 <     stahnma> | z00dax has been in #epel quite a bit
00:45              * | mmcgrath atrpms user.
00:45 <     stahnma> | and offering insight and having questions about
items
00:46 <       nirik> | yeah, perhaps a invite to dag/dries/rf folks
might be good?
00:46 <    mmcgrath> | Some 3rd party repos probably won't want to sign
the CLA and stuff.  Thats unfortunate but shouldn't really hurt us from
working with them.
00:46 <         thl> | nirik, dag was on the list
00:46 <     stahnma> | should we have a special meeting to work with
their discussions/conerncs/improvements etc?
00:46 <       thimm> | Dag brought up the repotag issue
00:47 <       nirik> | yeah, I would like for there to be
communication... I don't know what they would like if they don't tell us...
00:47 <         thl> | nirik, agreed, but we can't write anything
fomalized down afaics
00:47 <         thl> | at least nothing with repo names in it
00:47 <       quaid> | no CLA is required for epel-devel-list though
00:47 <         thl> | quaid, +1
00:47 <       nirik> | right. But we should open some communication and
find out their concerns and comments.
00:47 <    mmcgrath> | thl: we can always create a 3rd party repo page
though.
00:48 <         thl> | mmcgrath, there is someting about it in the faw
already
00:48 <       thimm> | nirik: Dag told us, then there were infinite-size
threads on his requests and finally his requets was voted down
00:48 <    mmcgrath> | <nod>
00:48 <       quaid> | I think the point is, we are finally ready to
formally engage Fedora with all the third-party repos, and this is the
place to do it
00:48 <    mmcgrath> | thimm: he had requests?  or just the repo-tag?
00:48 <       quaid> | so anyone who doesn't want to come here is not
really interested in cooperating directly, which is OK
00:48 <         thl> | quaid, +1
00:48 <       thimm> | s/requests/request/
00:49 <       nirik> | yeah, repotag is all I saw... perhaps he has
other things we could look at working with him on?
00:49 <       thimm> | AFAIK Dag's been "sent" back home.
00:49 <       nirik> | ?
00:49 <     stahnma> | part of cooperation is not all request items will
be honored.  They will be considered though, and hopefully have some
thought-provoking ideas around the situation
00:49 <    mmcgrath> | thimm: thats unfortunate.
00:50 <         thl> | dag he reqested some things nobody stepped on to
work on
00:50 <    mmcgrath> | We can't just do everything 3rd party repos tell
us, any more then 3rd party repos will do whatever we ask them.
00:50 <       thimm> | But was predictable
00:50 <       thimm> | correct
00:50 <         thl> | e.g. some kind of "build different branches from
one spec file"
00:50 <       nirik> | he also chimed in on the macros, and some work
has been done on that...
00:50            --- | BobJensen-Away is now known as BobJensen
00:51 <         thl> | nirik, rpmforge servers a whole differente user
base imho anyway
00:51 <         thl> | it has more new stuff
00:51 <         thl> | and replaces "core" pacakges
00:51 <         thl> | I think there are users for it
00:51 <     rdieter> | afaict, rpmforge and centos/Extras folks consider
that they were in this space first, so we(epel) should be going to them
for help, answers, cooperation.
00:51 <       nirik> | sure, but if we have communication we can try and
not conflict, etc.
00:51 <         thl> | and for our solution "no replaces, only important
updates"
00:51 <       quaid> | ah, Eitch was the other admin
00:51 <     rdieter> | imo, mostly ego, perception.
00:51              * | quaid wrong channel, sorry
00:52 <       thimm> | thl: Are you sure rpmforge replaces "core"?
00:52 <     rdieter> | thimm: yes.
00:52            --> | wpc4 (William Curley)  has joined #fedora-meeting
00:52 <     rdieter> | thimm: dag does anyway.
00:53 <       nirik> | I talked with z00dax the other day about centos
extras... I would really prefer if we don't conflict there, but we will
see...
00:53 <    mmcgrath> | nirik: with centosplus you mean or centos.karan.org?
00:53 <       thimm> | Doesn't epel and cnetos extras already conflict?
00:53 <         thl> | thimm, well, there are for example subversion
packages at http://dag.wieers.com/rpm/packages/subversion/
00:53 <         thl> | thimm, build for el5, which has subversion already
00:53 <       nirik> | mmcgrath: there is a centos "extras" repo... has
things like Xfce in it, etc...
00:53 <    mmcgrath> | ahh, yes.
00:54 <       thimm> | From Dag's page:
00:54 <       thimm> | A3. Why should I use RPMforge repositories ?
00:54 <       thimm> | There are many good repositories that you can
use. Here are someadvantages to our repositories: 	 We don't replace
base libraries or important core packages for repositories that are not
EOL. 	
00:54 <       nirik> | http://wiki.centos.org/Repositories
00:54 <    mmcgrath> | So what I'm hearing is "Yes, we want to work with
3rd party repos but their concerns are taken one at a time and right now
we have no idea what they want" ?
00:54 <       quaid> | rdieter: I can agree with that, and I don't think
EPEL is bringing ego into things, but the situation is that Fedora work
needs to be open/visible where all Fedorans can see it, and that means
not on some third-party list site somewhere
00:54 <       nirik> | right. I think we want to just tell them "hey, we
would like to hear your concerns/ideas and ask if you would like to join
us"
00:55 <     stahnma> | we also need to take the goals of each repo and
be sure we understand them
00:55 <     rdieter> | quaid: +1
00:55 <    mmcgrath> | is there anything we want from 3rd party repos?
00:55 <     stahnma> | that will help us be better and make sure we
don't conflict as much as safely possible
00:55 <       nirik> | BTW, in centos5 yum is in the extras repo I think
(replacing core)
00:55 <       quaid> |
http://fedoraproject.org/wiki/EPEL/CommunicationPlan#head-be699493fb37a6eb90a4841fbc9eaec4358a0044

00:56              * | thl has to leave soon
00:56 <       quaid> | we could improve on the communication plan for
third-party repos
00:56              * | stahnma tried to work on a few drafts last night,
but wasn't sure where to start
00:56 <       quaid> | input from you all in that is heartily welcomed!
00:56 <       quaid> | yeah, I'm not familiar enough with the people
00:56 <       quaid> | what i wrote there is what I feel :)
00:56 <     stahnma> | thimm I think you are a valuable resource when
talking about and on behalf of 3rd party repos
00:56 <       thimm> | centos had a legal battle with RH a couple of
years ago
00:56 <    mmcgrath> | If 3rd party repos are going to replace our
packages, is there an easy way to ensure that there won't be a conflict?
00:57 <         thl> | mmcgrath, no
00:57 <       nirik> | I think the next step is to mail the maintainers
of those repos and invite them to our meetings/mailing lists
00:57 <       thimm> | stahnma: I'm not empowered to talk about any 3rd
party
00:57 <    mmcgrath> | even with coordination?
00:57 <         thl> | nirik, they were invited in the past iirc
00:57 <       thimm> | Anything I say is my personal viewpoint
00:57 <       nirik> | thl: ah, ok. I didn't know that...
00:57 <     rdieter> | nirik: been there, done that, you're welcome to
keep trying tho.
00:58 <         thl> | rdieter, agreed
00:58 <         thl> | we really should move on imho
00:58 <     stahnma> | thimm: fair enough.  I think you can be more help
than you realize though :)
00:58 <       nirik> | well, I can try and get more input from z00dax.
He seems interested in talking sometimes at least. ;)
00:58 <         thl> | try to be nice to everyone (including 3rd party)
00:58 <    dgilmore> | thimm: you are your own
00:58 <       nirik> | moving on++
00:58 <       thimm> | dgilmore: ???
00:58 <    dgilmore> | thimm: i think its a great thing that you are
working with the fedora community as you are :)
00:59 <    dgilmore> | thimm: your own 3rd party repo
00:59            --> | giallu (Gianluca Sforna)  has joined #fedora-meeting
00:59 <       thimm> | Ah, OK, even that is not 100% my repo anymore,
but I have the lion's parts
00:59 <    mmcgrath> | thl: whats next on the list?
00:59 <    mmcgrath> | we're coming up on an hour.
01:00              * | nirik happily uses mythtv from atrpms every day. ;)
01:00 <       thimm> | nirik: the the RHEL5 build :)
01:00 <         thl> | mmcgrath, I#d say we are finished for today
01:00 <         thl> | the other two issues should be brought top the
list first
01:00 <    dgilmore> | ---------- Meeting
01:00 <    dgilmore> | Closed --------------------
}}}




More information about the epel-devel-list mailing list