Release Team Meeting 2004-03-04

	"The Release Team had a meeting? They're telling us what they talked
about? *shock*"

	Sorry guys - flame us to the high heavens if we ever suck this badly
again :-)

Minutes For GNOME Release Team Meeting 2004-03-04

Present:                      Missing:

Andrew Sobola                 Jonathan Blandford
Frederic Crozat
Jeff Waugh
Kjartan Maraas
Luis Villa
Mark McLoughlin
Murray Cumming


  DONE: Mark/Jonathan to try and get latest gtk+ into FC2 test2 so as to
        ensure new file chooser gets wider testing

  DONE: Luis find out what's going on with release notes (Jeff obliged)

  DONE: Mark to track string freeze adherence

  DONE: Luis to send bug lists

  PENDING: Jonathan let gnome-i18n know when filesel is going to be
           string frozen

  PENDING: Kjartan to let us know how the current i18n stats compare to
           2.4 (Gathering these stats require significant manual work)

  NEW: Luis to send more bug lists

  NEW: Jeff to track release notes progress

  NEW: Murray to setup a template for the release notes in CVS

  NEW: Mark to mail gtk team on release-team consensus about the 
       possible file chooser API break

  NEW: Jeff to announce schedule delay based on the gtk API break 

  NEW: Murray to announce code freeze. Again, pending the gtk decision

  NEW: Jeff to send out a call for tarballs for Beta 2. Also to identify
       any modules which desperately need someone to step in and do a 

  NEW: Murray to keep track of press release progress

  NEW: Luis to mail Jakub, Tuomas and Garrett about a splash screen

  NEW: Andrew to handle gathering screenshots

Decisions and discussion:

  * GNOME 2.6 Release Notes

	Some concern over the progress with the release notes. The release is
approaching fast. Jeff will talk to Colin Charles about what content he
has to date. Murray will set up a template in CVS so people can help

  * Quality

	Luis reports that 2.6 is looking pretty okay quality wise. There has
been significant testing and there aren't many embarassing stoppers.
Things look to be on track.

  * GTK+ 2.4.0 Progress

	Much discussion around the possible API breakage with GtkFileChooser
and the impact on GNOME. General uneasiness about slipping further after
already slipping 2 weeks for GTK+. Also, some discussion about the
impact for Fedora and our responsibility to meet our schedules.

	General consensus that the decision about the API break lies with the
GTK+ team and that the release-team is as concerned about the quality of
the GTK+ API as anyone else. If the API break is agreed upon, we will
announce a two week slip of GNOME 2.6.0 immediately. Much agreement on
the urgency of the decision.

  * New Modules

	Brief discussion on whether its neccessary to revisit the new modules
in GNOME 2.6 to make sure they still meet the criteria for inclusion.
Consensus that all new modules are still on track.

  * Beta Releases

	Beta 1 should be released publically later today. Proposal to push out
Beta 2 by a week given that Beta 1 is late. Decided that GNOME is seeing
enough testing for the Beta 1 delay to not be a significant issue.

	Tarballs for Beta 2 will still be due on Monday, irrespective of any
slip in the schedule. Jeff to handle this.

  * Preparations for 2.6.0 Final

	Much agreement that we should aim to have a press release prepared a
week in advance of the actual release date to avoid the confusion which
happened around the 2.4.0 press release. Murray will work on a list of
items to be included in the press release and work with Leslie. Luis has
already sent an initial list to the marketing mailing list which Murray
can start with.

	Splash screen needed. Mark has contacted Jimmac. Luis will email
Garrett also.  Murray noted that it would be good to send out a request
for splash screens well in advance next time so that anyone can

	Some discussion on the need for screenshots for the final release. Jeff
to host them on Andrew to send out a call for

  * Future Meetings

	Agreement to continue meeting at the same time on irc and that we
should start sending out minutes publically again.

<jdub> go for it
<fcrozat|mtg> markmc: go...
<luis> urp, sorry, forgot
<markmc> morning luis
<fcrozat|mtg> just in time :)
<markmc> okay, action items
<markmc>      - Mark/Jonathan to try and get latest gtk+ into FC2 test2 so as to       ensure new file chooser gets wider testing
<markmc> that's done
<markmc> and it'll continue
<markmc> in fact alex and I are keeping abreast of all of GNOME
<markmc>      - Luis find out what's going on with release notes
<markmc> jdub, you did that right ?
<jdub> no word from colin, really
<jdub> and nothing up in cvs afaik
<jdub> i'll check now
<murrayc> It's strange that he doesn't reply to the emails.
<jdub> he's travelling
<fcrozat|mtg> wasn't he out of email this week ?
<markmc> yeah, and from the fedora planet he's obviously around and doing stuff
* markmc is getting worried about this
<murrayc> I'd like to see what there is before the last minute.
<jdub> i'm less worried about not reading emails than lack of docs in releng :)
<jdub> i'll call him or something
<murrayc> Do you know him well? I never heard of him.
<jdub> meanwhile, can someone set up the 2.6 release notes template in releng?
<murrayc> OK, I can do that.
<jdub> he's from melbourne, met him at lca
<markmc> jdub, so will you continue to track this ?
<markmc> and murray will do the template
<jdub> markmc: i'll keep trying to get in touch with colin, murray can sort out releng
<markmc> main thing is that we know what's going on and we can jump in if he doesn't have the time
<markmc> okay, next
<markmc>      - Mark to send mail backing up Christian on string freeze
<markmc> decided it wasn't neccessary
<markmc> things seem to be okay there now
<markmc> right ?
<luis> that seems to have calmed down a bit.
<murrayc> yes
<markmc>      - Jonathan let gnome-i18n know when filesel is going to be string       frozen
<markmc> don't think that happened
<markmc> not sure if its string frozen yet even
<markmc> sigh
<markmc>      - Kjartan to let us know how the current i18n stats compare to 2.4
<kmaraas> it hasn't changed much lately
<murrayc> How is arabic this time?
<kmaraas> don't have good numbers yet, but I can take a look tonight
<markmc> cool
<markmc> do you have a rough feeling even ?
<murrayc> arabic would be a big win. I think it was planned for 2.6.
<markmc> or we better or worse off ?
<kmaraas> murrayc, it's around 80-90%
<jdub> kmaraas: if you get some numbers, please post to d-d-l as well as i18n/r-t -> needs exposure :)
<kmaraas> jdub, ok
<kmaraas> hmm, no, it's at 67%
<markmc> thanks kmaraas
<markmc>      - Luis to send bug lists
<markmc> that's done, right ?
<luis> yup
<luis> though it's sort of an ongoing thing
<markmc> yeah, will there be another one this week
<markmc> the last one was useful
<luis> yes, hope so, though shorter
<markmc> though I expected it to be more worrying :)
<markmc> which brings us onto
<luis> it's really not a bad release, quality wise
<markmc>   * Quality update (Luis or Andrew: could you guys prepare this?)
<luis> I'm just nervous because of coverage
<luis> there isn't a formal update, unless andrew had more time than I expected :)
<murrayc> We have 4 days before Hard Code Freeze.
<markmc> yeah, is that something to be worried about ?
<markmc> or are we okay, just about ?
<luis> but the luis gut-feeling-o-meter says 'not a bad release, we have had a fair amount of testing and nothing utterly gross has show up, I think we'll be OK'
<murrayc> Only if there is >0 big crasher.
<luis> I'm spending today focusing bugday on new-ish 2.4 and 2.5 bugs to check for 'would you be embarassed' type bus
<luis> bugs
<markmc> okay, that sounds good
<murrayc> That is when we must decide whether any of those bugs is a blocker.
<luis> if we find anything there (or even if we don't) I'll send out a 2.6.0 email tonight so that we have people working the weekend before freeze.
<markmc> its as good as we've ever felt before I guess
<markmc> now
<markmc>   * GTK+ 2.4.0 progress
<markmc> fun !
<luis> well, I felt better about 2.0, seeing as I read bugs all day every day ;)
<murrayc> It's only API bugs that are the big problem. Other things can be fixed by the time they get into distros.
<markmc> so, I presume everyone is pretty up to date with what's going on with gtk+
<kmaraas> luis, the intltool blocker has been fixed at least
<markmc> yeah
<jdub> it sounds ok from the meeting reports
<markmc> how do people feel about this API break thing ?
<jdub> if they decide to api break, i think we ought to slip
<murrayc> I want to see the filechooser API fixed, but it could definitely cause a 2 week release delay - based on the DISABLE_DEPRECATED experience.
<markmc> my feeling is, there's a time when you have to bite the bullet and just live with API problems
<markmc> I mean in 4 days time ..
<markmc> any API problems, they're going to have to live with them
<murrayc> Yes. But this might be a really big problem.
<markmc> and this problem doesn't seem like a huge problem 
<markmc> how ?
<kmaraas> I don't think that's the worst thing that could happen though, two weeks is not going to kill us
<murrayc> If it prevents an application from giving meaningful feedback to the user.
<jdub> kmaraas: that's one month total slippage
<kmaraas> and we get to polish some more if need be
<jdub> kmaraas: i would prefer to cut that out of the next cycle
<markmc> murrayc, oh don't get me wrong, I think they should fix it - but not by breaking the existing API
<jdub> kmaraas: so we shouldn't take it *too* lightly
<kmaraas> jdub, yeah, but it's not bad when marrying an unstable gtk+
<kmaraas> jdub, if we avoid that in the future we're good I think
<jdub> kmaraas: it's over-a-barrel time, for sure :)
<kmaraas> :)
* markmc thinks we'll be able to sort that out next time around
<murrayc> I personally can accept a 2 week slip, if we never ever do it again.
<murrayc> How is that for fedora though?
* murrayc hasn't looked at the fedora schedule.
<kmaraas> good question
<jdub> markmc: can we action two or three people to sort out the gtk+ decision with them?
<luis> I was under the understanding that fedora is hosed by SE Linux right now anyway?
<markmc> murrayc, lest I be accused of anything ... that's not our problem
<markmc> but, its not good for Fedora, basically
<murrayc> It should be our problem.
<markmc> whyso ?
<jdub> markmc: fedora has other issues, but we need to take responsibilty for it
<kmaraas> still we don't have to have a two week slip
<kmaraas> we can do whatever we can to minimize it
<jdub> markmc: we should care for our user as gtk+ should care for theirs.
<murrayc> Because we are time-based to make life easier for people like distros.
<markmc> jdub, sure
<markmc> yeah, okay that makes sense
<markmc> um
<fcrozat|mtg> murrayc: OTOH, distros should always take project schedule with lot of cautions :)) 
<markmc> so who do you want to action with the gtk thing jdub ?
<luis> fcrozat|mtg: well, we've told distros they don't have to take our schedule with cautions
* markmc thinks we should sort it out now
<murrayc> So, a vote? Slip for the break or have a compromise non-breaking API?
<jdub> anyway, if we get a couple of people to taskmaster this decision in the next few days, we should know about slippage requirements
<kmaraas> if we get the fix into gtk+ this week, and start rolling new tarballs for the modules that need it right away we can get away with a shorter slip than two weeks I think
<jdub> markmc: you and murray seem interested enough :)
<fcrozat|mtg> luis: and they trusted us ? well, I didn't trust myself :)
<markmc> no, lets just decide now
<markmc> 1) gtk+ team can break the freeze if they really feel its neccessary
<markmc> 2) they should decide today
<markmc> 3) if they do, we'll slip two weeks
<markmc> ?
<murrayc> I think a break will cause 2 weeks slip, and there's no avoiding it. Optimism doesn't pay.
<jdub> markmc: if they break, then yes, we will slip
<jdub> markmc: the decision is whether they break or not
<markmc> okay, that's that then
<jdub> markmc: so we need a couple of people to taskmaster that
<markmc> yeah, and that's they're decision
<murrayc> I am now undecided. I hate to make life difficult for other schedules.
<murrayc> Does anybody vote definitely yes to the API break?
<markmc> note me anyway
<jdub> murrayc: fedora have their own problems atm
<markmc> not me anyway
<jdub> murrayc: that's not our decision to make
<murrayc> jdub: Hmm.
<markmc> okay, so I'll definitely keep a close eye on this
<murrayc> So, we just tell them "Decided today. Tell us what you decided."
<markmc> and send gtk-devel a mail after this
<kmaraas> I think we have to take some responsibility for tying in with an unstable gtk+
<kmaraas> and let them do what's best for them (and us) in the long run
<jdub> murrayc: we get two people to co-operate with gtk to finalise this; sounds like that'd be you and markmc -> easy
--- murrayc is now known as murrayc_mtg
<jdub> kmaraas: yeah
<markmc> kmaraas, no, its for tying ourselves to a feature based schedule
<murrayc_mtg> OK, but we decide today, and we make it clear that it means a 2 week GNOME slip.
<markmc> kmaraas, nothing wrong with a gtk+ on a time based schedule
<kmaraas> markmc, same difference really
<kmaraas> markmc, the file chooser thing is special in some ways
<murrayc_mtg> I am very much in favour of tying ourselves to GTK+ in future, anyway.
<markmc> okay, done on this then ?
<kmaraas> we *have* to have it
<jdub> markmc: dude, we've *already* made our mistake with gtk+ - we're just in damage control now :-)
<markmc> jdub, yeah, I'm talking randomly about the future :)
* markmc smacks markmc
<markmc> right
<markmc> moving on
<markmc>   * New modules - do we need to go back and sure we're happy with each     one being in the final release?
<jdub> oh, i have some ideas about that too, but it's way out of this meeting's scope :)
<jdub> i'd say no
<jdub> the only one that we could remove is dasher
<jdub> and it's fine
<murrayc_mtg> I never thought of removing any before.
<markmc> well, it is part of the whole deal
<fcrozat|mtg> why removing dasher ?
<markmc> i.e. if we were including evo, we would definitely be revisiting it now
<murrayc_mtg> Nobody is using dasher, so it's of no importance if it had problems.
<markmc> all I'm wondering is, are there any other modules like that ?
<murrayc_mtg> markmc: Yes, it's a question we should ask every time.
<jdub> fcrozat|mtg: we're not - sounds like mark is just doing due diligence :)
<markmc> jdub says no, good enough for me :)
<fcrozat|mtg> jdub: ok :))
<jdub> markmc: nup, all fine
<markmc> okay
<markmc>   * Beta 2 preparations
<markmc>      - Code freeze advance warning
<jdub> release coming out in the next few hours
<murrayc_mtg> I can send the warning today.
<markmc> which release ?
<markmc> jdub, ?
<jdub> can we delay the code freeze warning until we get slip details sorted?
<jdub> markmc: .90
<markmc> jdub, is that not out already ?
<jdub> nup
<murrayc_mtg> jdub: Makes sense. Either a freeze warning or a slip announcement tomorrow then.
<markmc> hmm
<murrayc_mtg> Is .90 == Beta 1?
<markmc> should we not be talking about slipping anyway, then ?
<jdub> murrayc_mtg: you'll be the first to know about the slip, so go for the warning if it happens
<markmc> i.e. we need time when the betas are actually out, right ?
<jdub> markmc: well, i was basing that discussion on current status, so yes, we have ;-)
<murrayc_mtg> By 1 week or 2?
<markmc> sorry, didn't grok that
<jdub> if gtk+ breaks, we slip two
<jdub> if we want to slip one anyway, we can do that
<markmc> okay, so definitely slipping one then ?
<murrayc_mtg> I thought you said Beta1 is late, so we slipped already anyway.
* markmc thinks that's what it sounds like
<jdub> we have the option to slup
<jdub> yea or nay
* murrayc_mtg is confused.
<markmc> what was the two week slip for again ?
<jdub> markmc: gtk+
<markmc> because one of the other releases was late right ?
<jdub> sheesh
<markmc> ah
<markmc> sorry, just confused
<jdub> if we have a gtk+ break, we slip two weeks
<jdub> we have the option to slip one week based on tarball release lag
<murrayc_mtg> OK, do we want to take that option?
<markmc> yes, I think we need to do that
<markmc> everybody else ?
<murrayc_mtg> I think so. release should not be late. It's exceptional.
<luis> hrm
<luis> why do we think we need to slip, exactly?
<murrayc_mtg> Because betas should be tested.
<markmc> because Beta 1 is a week late
<markmc> and we need to have real time when people can test them
<markmc> does that make sense ?
<jdub> i think the gtk+ break is fairly likely anyway
<jdub> let's see what happens, and slip one week or two?
<kmaraas> ok, so we slip two more weeks to allow for gtk breakage and more testing?
<murrayc_mtg> But this could mean 1 (release late) + 2 (GTT+ break) = 3 weeks slip
<jdub> not additive
<markmc> no, lets decide on the first week now
<murrayc_mtg> jdub: Maybe not.
<markmc> and not pretend that we're doing it because of gtk+
<luis> speaking as the testing nazi, I'm not terribly worried about testing at this point; people are building and testing things regularly. More testing would be nice but I don't think it is a reason in and of itself to slip.
<jdub> i mostly agree with luis
<murrayc_mtg> So, you think that jhbuild testing is enough, and we don't get much more from tarball testing?
<jdub> in terms of really vitally useful testing right now, yes
<luis> I think we've gotten alot of tarball testing, and the churn in the codebase has been fairly low lately
* murrayc_mtg will defer to the bugmaster on this.
<markmc> okay, I'm willing to go with luis on this
<luis> the biggest issue is pushing people to test gtk
<kmaraas> me too
<luis> that's the scary one
<markmc> that's fine, just presumed we would need more testing
<markmc> alright
<kmaraas> and since that's in fedora testing now it should be good I think
<murrayc_mtg> I think GTK+ functionality and stability is fine.
<markmc> back to what we were talking about
<markmc>  * Beta 2 preparations
<markmc> so, murray's doing whatever announcement needs to happen
<markmc>      - Call for tarballs
<jdub> the code freeze warning
<jdub> markmc: will do that as soon as the release is out
<murrayc_mtg> or the slip
<markmc> jdub, as soon as Beta 1 is out
<markmc> murrayc_mtg, yeah, exactly 
<markmc> jdub, ?
<markmc> is that what you meant ?
<jdub> yes
<markmc> cool, thanks
<murrayc_mtg> There'll be a Bindings Beta 1 today too, by the way.
<markmc>      - Any module in need of someone to jump in do a release?
<markmc> murrayc_mtg, cool
<jdub> if we decide to slip, i would prefer to do the announcement for that
<murrayc_mtg> jdub: OK
<jdub> markmc: i'll answer that immediately after the release too
<markmc> okay, cool
* markmc notes all this
<jdub> thanks
<markmc> let me know if you need someone to do a release of some random module
* markmc loves distchecking :/
<markmc> so, final item on that
<markmc> do you have the bandwidth to handle the Beta 2 release jdub ?
<jdub> yeah
<jdub> flow has imploded now
<murrayc_mtg> this gives you lots of time?
<markmc> so, you'd expect tarballs due Monday, release Wednesday or something ?
<jdub> yeah
<jdub> yeah
<markmc> it pretty important they happen quickly at this stage ..
<markmc> okay, cool
<murrayc_mtg> jdub: And you'll not go hungry?
<jdub> no
<markmc> heh
<markmc> moving on ?
<markmc>   * Preparations for final release:
<markmc>      - Release notes
<markmc> covered that
<markmc>     - Press release
<murrayc_mtg> Who normally bugs who about that?
<fcrozat|mtg> leslie..
<jdub> okay, well, leslie is partially useless
<murrayc_mtg> How bad was it for 2.4.0? I was away.
<jdub> what we need to do is have a list of points to make in the press release
<jdub> ready to give to her
<jdub> and we can do that by next week, i'm sure
* markmc would be much more conformtable if we aimed to have the press release ready a week in advance
<murrayc_mtg> But let's give her a warning now, and get confirmatoin that's she's ready.
<markmc> okay, so there's two action items
<jdub> markmc: if we have our end ready next week, we can put timelines on her table
<markmc> 1) draw up a list of bullet points for the press release
<markmc> 2) liase with leslie
<murrayc_mtg> luis: You have already thought about major improvements. Can you send her and us a prelimary list?
<markmc> takes ?
* markmc isn't sure what kind of stuff to put in a press release
<jdub> markmc: it's hard for this release
<markmc> yeah
<jdub> actually, kmaraas, when were you going to have i18n stats?
<murrayc_mtg> The usual "we did it again". Nautilus speed and improvements.
<murrayc_mtg> File Manager.
<murrayc_mtg> File Chooser, I mean.
<markmc> murrayc_mtg, you sound like you're handling (1) then ? :-)
<kmaraas> jdub, I have to calculate them manually so it could take some time
<jdub> leslie has already had pingage about the release notes
<jdub> press release, rather
<kmaraas> jdub, since the status pages have a lot more strings than just developer platform and desktop
<murrayc_mtg> jdub: Did you CC us?
<jdub> it was on m-p and r-t
<murrayc_mtg> OK, I'll find it.
<jdub> confusingly called "2.6 Press Release"
<markmc> so, murrayc_mtg, am I putting you do for this then ?
<luis> murrayc_mtg: I've already sent it to the marketing list
<murrayc_mtg> OK, I want to be more involved with the release notes anyway.
<luis> (which includes leslie)
<murrayc_mtg> This means I will be bothering leslie and colin a bit, politely.
<jdub> see subject in r-t and m-l
<murrayc_mtg> ok
<markmc> alrighty
<markmc> next !
<markmc>      - Splash screen
<murrayc_mtg> 3) I or jdub set up start/2.6?
<markmc> murrayc_mtg, good point !
<jdub> murrayc_mtg: just worry about releng notes for now
<murrayc_mtg> jdub: OK
<markmc> jdub, it'll be good to have a list of 2.6 release things and who will do them when the time comes
<luis> markmc: hrm, splash screen I should poke jimmac
<jdub> markmc: need to action someone to bug tigert and jimmac
<markmc> yeah, I mailed jimmac a few days ago
<markmc> no response
<jdub> markmc: it can mostly be done by one person, but i'll document the process again (there's a list in r-t archives already)
<markmc> so I don't know whether to repok him or one will magically dissappear
<murrayc_mtg> We should maybe ask for splash screens earlier in future, from everybody.
<jdub> luis: screenshots hosting doable again?
* markmc notes that one doesn't get any response from artists other than artwork :)
<murrayc_mtg> jdub: That would be cool.
<jdub> luis: was considering just using
<jdub> murrayc_mtg: already been done a number of times
<luis> jdub: shouldn't be a problem, though ideally having more than a few hours to get the shots into place would be good. :)
<jdub> luis: that would be easier to co-ordinate
<murrayc_mtg> archive searching is difficult.
<jdub> aes: have you got many screenshots yet?
<jdub> luis: ok, i'll just do f.g.o
* markmc doesn't understand all the screen shot stuff
<jdub> markmc: perhaps action aes to do a screenshots reminder?
<markmc> "action: Jeff to handle screenshots"
<murrayc_mtg> It's on gnomedesktop.
<markmc> oh, aes ?
<jdub> aes to gather
<jdub> i'll just ship them
<markmc> ah, cool
<markmc> right, done
<markmc>      - How much notice for sysadmin/mirrors?
<jdub> a couple of days
<jdub> just a matter of pinging stric
<markmc> okie
<markmc>      - Anything else?
<jdub> he doesn't watch for schedule changes
<murrayc_mtg> Is that more than just the sync trigger script?
<markmc> in terms of final release preperations
<jdub> murrayc_mtg: is what?
<luis> hrm
<murrayc_mtg> mirror updating?
<luis> we should poke garret for splash too
* luis does that
<jdub> markmc: not really
<kmaraas> I have to leave, do you need me more?
<jdub> murrayc_mtg: that only pushes to two mirrors
<jdub> murrayc_mtg: but we point everyone to f.g.o anyway
<jdub> kmaraas: nup, thanks!
* kmaraas bows out
<-- kmaraas has quit (Leaving)
<markmc> ahrr
<jdub> luis: good call
<markmc> oh well
<murrayc_mtg> ah,
<markmc> alrighty
* markmc moves it along
<markmc>   * Schedule review
<markmc> we've covered that enough ?
* jdub thinks so
<markmc> okay, final item
<markmc>   * Next weeks meeting time.
<jdub> same is ok by me
<markmc> it is worthwhile continuing this right ?
<markmc> but is this time okay ?
<murrayc_mtg> yes. me too.
<luis> same is fine with me
<markmc> luis, not good for you I take it ?
<markmc> oh, cool
<fcrozat|mtg> ok for me too
<luis> markmc: it's fine by me, I'm just likely to forget unless someone wanders into #bugs and pokes me
* luis usually shows up for bugday fairly blearyeyed
<markmc> great
<markmc> done then
<markmc> anything else people ?
<jdub> markmc: "action... entire release team to poke luis in #bugs" ;-)
* markmc will send proto minutes to r-t and then send them on to d-d-l tommorrow morning
<murrayc_mtg> thank you thank you thank you.
<markmc> ====================== meeting adjourned =====================

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]