Thread: BOF at pgCon?
Koichi,all, Dan says that nobody has sent him a request for a BOF slot. We wanted to hold this on Friday night or Tuesday night? What was the plan for this? --Josh
I think Tuesday night is better. I remember plan includes review of cluster feature discussion since Tokyo meeting and setting up cluster feature set to help people to select the product. I'd like to talk with my colleagues to collect other issues. Regards; ---------- Koichi Suzuki 2010/5/8 Josh Berkus <josh@postgresql.org>: > > Koichi,all, > > Dan says that nobody has sent him a request for a BOF slot. We wanted > to hold this on Friday night or Tuesday night? What was the plan for this? > > --Josh > > -- > Sent via pgsql-cluster-hackers mailing list (pgsql-cluster-hackers@postgresql.org) > To make changes to your subscription: > http://www.postgresql.org/mailpref/pgsql-cluster-hackers >
On Fri, May 7, 2010 at 6:30 PM, Josh Berkus <josh@postgresql.org> wrote: > Koichi,all, > > Dan says that nobody has sent him a request for a BOF slot. We wanted > to hold this on Friday night or Tuesday night? What was the plan for this? Friday strikes me as a *terrible* time to have it, as: a) It's after end of conference, and I know I'm always pretty baked by that point. b) People start leaving for the airport and such by this point. c) Of those that remain, "vegging out" at the Royal Oak is much more desirable than an energetic discussion. Tuesday seems like a way, way, way better time for it. That time suffers from core-ish folks being somewhat "baked" from the developer summit, but it's surely not as bad as Friday that way. -- http://linuxfinances.info/info/linuxdistributions.html
Christopher Browne wrote: > On Fri, May 7, 2010 at 6:30 PM, Josh Berkus <josh@postgresql.org> wrote: > > Koichi,all, > > > > Dan says that nobody has sent him a request for a BOF slot. ?We wanted > > to hold this on Friday night or Tuesday night? ?What was the plan for this? > > Friday strikes me as a *terrible* time to have it, as: > a) It's after end of conference, and I know I'm always pretty baked by > that point. > b) People start leaving for the airport and such by this point. > c) Of those that remain, "vegging out" at the Royal Oak is much more > desirable than an energetic discussion. > > Tuesday seems like a way, way, way better time for it. > > That time suffers from core-ish folks being somewhat "baked" from the > developer summit, but it's surely not as bad as Friday that way. The developer summit is Wednesday. The tutorials are Tuesday. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com
On 5/7/2010 10:33 PM, Bruce Momjian wrote: > Christopher Browne wrote: >> On Fri, May 7, 2010 at 6:30 PM, Josh Berkus <josh@postgresql.org> wrote: >> > Koichi,all, >> > >> > Dan says that nobody has sent him a request for a BOF slot. ?We wanted >> > to hold this on Friday night or Tuesday night? ?What was the plan for this? >> >> Friday strikes me as a *terrible* time to have it, as: >> a) It's after end of conference, and I know I'm always pretty baked by >> that point. >> b) People start leaving for the airport and such by this point. >> c) Of those that remain, "vegging out" at the Royal Oak is much more >> desirable than an energetic discussion. >> >> Tuesday seems like a way, way, way better time for it. >> >> That time suffers from core-ish folks being somewhat "baked" from the >> developer summit, but it's surely not as bad as Friday that way. > > The developer summit is Wednesday. The tutorials are Tuesday. Not everyone (me for example) is attending the tutorials. I don't know by what time I'm going to arrive on Tuesday night. Jan -- Anyone who trades liberty for security deserves neither liberty nor security. -- Benjamin Franklin
> Not everyone (me for example) is attending the tutorials. I don't > know by what time I'm going to arrive on Tuesday night. I know I'm arriving about 6pm Tuesday, so it's possible to do something that night. I'm leaving Friday evening, so that night is *not* possible. Even I were staying, I'd not find it a good night. What about during the day, perhaps a lunch? Or if we could all find a slot where we aren't speaking and/or ;) not interested in the talks for that block... Talks don't start until 10AM on Friday, so <10 is an option too. -- Greg Sabino Mullane greg@endpoint.com End Point Corporation PGP Key: 0x14964AC8
Attachment
Greg Sabino Mullane wrote: > What about during the day, perhaps a lunch? Or if we could all find a slot where we > aren't speaking and/or ;) not interested in the talks for that block... > Talks don't start until 10AM on Friday, so <10 is an option too. > It's tough to just do lunch given the fairly narrow time window for it and how many talks I'm seeing from people here around that time. Having a talk and a break worth of time, or something else that cuts across more time, would be more likely to work. And I'm not real optimistic about being productive early in the morning after the EDB party. Since I'm happy to be the jerk who announces which slots I'd consider missing for a cluster meeting instead, I'll list what I see as schedule possibilities during the conference itself, given the regulars I know of on this list and what talks I expect to be popular: 1) Friday, 5:30PM-? Take over the room potentially as soon as the Postgres-XC talk ends and miss "Closing sessions" if we're concerned about starting early for people who have to leave on Friday night, continue into the BOF #2 time from 6:00PM-8:00PM. What's nice about this is we have a room to grab and everything. 2) Wednesday, 6:00PM-?, after developer meeting. I expect everyone to be tired of talking at this point, but it's a possibility. Not sure where to meet though. 3) Friday, 9:00AM-10:00AM. Could borrow one of the session rooms. Not quite enough time and I'd expect attendance to be low for this one. 4) Thursday, 11:00AM-1:30PM. Nice follow up to the "Built-in replication in PostgreSQL 9.0" session. Would miss "Hypothetical Indexes towards self-tuning in PostgreSQL", "PostgreSQL in Mission-Critical Financial Systems", "Efficient k-nn search with GiST and other development"; not a lot of overlap in topics/speakers there I think. 5) Friday, 10:00AM-11:30AM. Leads into the "Replication Panel" on Friday at 11:30AM. Would miss "PL/Parrot", "PgMQ", "Identifying, Monitoring, and Controlling PostgreSQL Traffic [through pgBouncer]". Realistically, I think only (1) is feasible here, but there's the other options I see. -- Greg Smith 2ndQuadrant US Baltimore, MD PostgreSQL Training, Services and Support greg@2ndQuadrant.com www.2ndQuadrant.us
Thanks, Greg, that about matches with what I was thinking too. My votes are for: * 2 (post-developer's meeting: maybe over dinner somewhere?**) * 4 (Thursday slot before lunch), * 5 (Friday slot before lunch) * 3 (Friday before talks start) ** How many are we, Josh? -- Greg Sabino Mullane greg@endpoint.com End Point Corporation PGP Key: 0x14964AC8
Attachment
On Mon, May 10, 2010 at 20:53, Greg Smith <greg@2ndquadrant.com> wrote: > Greg Sabino Mullane wrote: >> >> What about during the day, perhaps a lunch? Or if we could all find a slot >> where we aren't speaking and/or ;) not interested in the talks for that >> block... >> Talks don't start until 10AM on Friday, so <10 is an option too. >> > > It's tough to just do lunch given the fairly narrow time window for it and > how many talks I'm seeing from people here around that time. Having a talk > and a break worth of time, or something else that cuts across more time, > would be more likely to work. And I'm not real optimistic about being > productive early in the morning after the EDB party. > > Since I'm happy to be the jerk who announces which slots I'd consider > missing for a cluster meeting instead, I'll list what I see as schedule > possibilities during the conference itself, given the regulars I know of on > this list and what talks I expect to be popular: > > 1) Friday, 5:30PM-? Take over the room potentially as soon as the > Postgres-XC talk ends and miss "Closing sessions" if we're concerned about > starting early for people who have to leave on Friday night, continue into > the BOF #2 time from 6:00PM-8:00PM. What's nice about this is we have a > room to grab and everything. FYI, Dan is looking into getting us a general "developer room" available more or less the whole conference. Depending on the size of the group, it should be possible to use that. I don't think we've seen confirmation if we'll have a room yet, though. -- Magnus Hagander Me: http://www.hagander.net/ Work: http://www.redpill-linpro.com/
> 1) Friday, 5:30PM-? Take over the room potentially as soon as the
> Postgres-XC talk ends and miss "Closing sessions" if we're concerned
> about starting early for people who have to leave on Friday night,
> continue into the BOF #2 time from 6:00PM-8:00PM. What's nice about
> this is we have a room to grab and everything.
I, personally, cannot miss the closing session, and Bruce & Tom were
talking about having 10 minutes of speaking.
Note that the *other* BOFs are Friday night.
> 2) Wednesday, 6:00PM-?, after developer meeting. I expect everyone to
> be tired of talking at this point, but it's a possibility. Not sure
> where to meet though.
I could arrange a space, maybe even dinner. Agreed that there would be
the talked-out issue. Because of a schedule conflict Jan and I have,
though, we'd need to decide this by Wednesday so I could reschedule that.
> 3) Friday, 9:00AM-10:00AM. Could borrow one of the session rooms. Not
> quite enough time and I'd expect attendance to be low for this one.
With the EDB party the night before? This seems highly unrealistic.
> 4) Thursday, 11:00AM-1:30PM. Nice follow up to the "Built-in
> replication in PostgreSQL 9.0" session. Would miss "Hypothetical
> Indexes towards self-tuning in PostgreSQL", "PostgreSQL in
> Mission-Critical Financial Systems", "Efficient k-nn search with GiST
> and other development"; not a lot of overlap in topics/speakers there I
> think.
>
> 5) Friday, 10:00AM-11:30AM. Leads into the "Replication Panel" on
> Friday at 11:30AM. Would miss "PL/Parrot", "PgMQ", "Identifying,
> Monitoring, and Controlling PostgreSQL Traffic [through pgBouncer]".
The problem with doing this during a session is that I assumed that we
wanted to attact other interested attendees, and not just the people on
this list. Don't we?
--
-- Josh Berkus
PostgreSQL Experts Inc.
http://www.pgexperts.com
It's sounding like early votes are leaning toward a dinner and talk session after the developer's meeting. Josh, if youcan work out the logistics, I can arrange to throw some funds toward booking somewhere too if that helps. It might be nice to extend the invitation here to a larger group. But to be honest I'm not sure anybody who's not alreadyactive in replication development will get much out of it besides massive overload of detail. That's why I thoughtsomething low-key during other sessions was worth considering. Sent via BlackBerry from T-Mobile
Josh Berkus wrote: > > > 1) Friday, 5:30PM-? Take over the room potentially as soon as the > > Postgres-XC talk ends and miss "Closing sessions" if we're concerned > > about starting early for people who have to leave on Friday night, > > continue into the BOF #2 time from 6:00PM-8:00PM. What's nice about > > this is we have a room to grab and everything. > > I, personally, cannot miss the closing session, and Bruce & Tom were > talking about having 10 minutes of speaking. The idea of Tom and I speaking at the end was never confirmed so I assume it is not happening. It was just an idea I threw out. However, I agree core members can't miss the closing session. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com
All,
Given the total lack of "I can make it / I can't make it" responses on
this thread, I'm going to suggest that we hold our BOF on Friday night
at the same time as the other BOFs.
Of course, if people are around on Tuesday Night, I will be too. Heck,
I'll be around on Monday night but working on my tutorial.
--
-- Josh Berkus
PostgreSQL Experts Inc.
http://www.pgexperts.com
> * 2 (post-developer's meeting: maybe over dinner somewhere?**)
> * 4 (Thursday slot before lunch),
> * 5 (Friday slot before lunch)
> * 3 (Friday before talks start)
>
> ** How many are we, Josh?
That depends on whether we're looking to attract othere developers than
just the ones on this list.
--
-- Josh Berkus
PostgreSQL Experts Inc.
http://www.pgexperts.com
On Wednesday 12 May 2010 21:14:39 Josh Berkus wrote: > > * 2 (post-developer's meeting: maybe over dinner somewhere?**) > > * 4 (Thursday slot before lunch), > > * 5 (Friday slot before lunch) > > * 3 (Friday before talks start) > > > > ** How many are we, Josh? > > That depends on whether we're looking to attract othere developers than > just the ones on this list. I am sure there are others than me interested in it, only lurking here or not even reading this list - especially as the visibility of it wasnt that big in the past. Andres
> Given the total lack of "I can make it / I can't make it" responses on > this thread, I'm going to suggest that we hold our BOF on Friday night > at the same time as the other BOFs. > > Of course, if people are around on Tuesday Night, I will be too. Heck, > I'll be around on Monday night but working on my tutorial. I thought we were leaning towards a dinner meeting on Wednesday night? That's far better for me than Friday evening. -- Greg Sabino Mullane greg@endpoint.com End Point Corporation PGP Key: 0x14964AC8
Attachment
> I thought we were leaning towards a dinner meeting on Wednesday night?
>
> That's far better for me than Friday evening.
Jan and I can't do both a BOF on Friday and a dinner meeting on
Wednesday. So we'd have to pick one *or* the other.
Wednesday night won't work for attracting people other than the ones on
this list.
So it depends on the purpose of the meeting.
--
-- Josh Berkus
PostgreSQL Experts Inc.
http://www.pgexperts.com
Josh Berkus wrote: >> I thought we were leaning towards a dinner meeting on Wednesday night? >> >> That's far better for me than Friday evening. > > Jan and I can't do both a BOF on Friday and a dinner meeting on > Wednesday. So we'd have to pick one *or* the other. > > Wednesday night won't work for attracting people other than the ones on > this list. > > So it depends on the purpose of the meeting. > The purpose of the meeting will also help in picking a better venue. Do we want a lecture room at the university with a white board and chairs so when one person speaks everyone can hear them or do we want a restaurant that will be geared more towards lots of concurrent informal discussions. What would the agenda for the BOF look like? or do we just want an informal social gathering? -- Steve Singer Afilias Canada Data Services Developer 416-673-1142
Josh Berkus wrote: > Jan and I can't do both a BOF on Friday and a dinner meeting on > Wednesday. So we'd have to pick one *or* the other. > Wednesday night won't work for attracting people other than the ones on > this list. > So it depends on the purpose of the meeting. > Clustering and replication projects are really important, right? And there's a lot to talk about from both the development and the user side of things. The idea that there will be one big meeting that covers everything is looking really impractical, and I'm not so sure that's a bad thing. It's not like we're getting 100% attendance from around the world anyway; eventually I'd like to get more input from Marko and Hannu on a few of these topics, just to pick the first from the regulars on this list not in attendance that come to my mind. We've scheduled to make some progress at the developer's meeting. I think it's important to have a BOF session on this topic that is on the schedule, and that all the users are invited to attend. Catering to that audience will preclude some of the really serious tech talk anyway, so still useful even if there's a "hackers" session (or more!) before that. It sounds like that will include only a subset of the people we'd like to be there, but that's how it goes. I know what Bucardo use cases look like well enough to steer people there even if we can't get the project's esteemed leader to attend. From what I'm hearing here, that's all it's really practical to start the conference with. I think further planning beyond the BOF slot will only drive us all crazy. We can regroup in Ottawa and just all try to be on the lookout to advance all the open projects here usefully. Search for opportunities to have one or more informal meetings, in as large of a subset as can be arranged, and just do it. Possibilities include: -Meet after Josh's talk that ends at 3:30PM on Tuesday. I'll certainly be there by the end of that, flying in that morning. Doubt there's anything happening in that space afterwards. -Meet after the Developer's session on Wednesday. Josh and Jan have other plans, and it doesn't sound like there's a definite enough alternative to ask them to cancel over; we really need them at the BOF instead. Maybe this one just turns into an informal session without them, when everybody else who's arrived already gets together at the Royal Oak to grab our registrations. -Blow off one of the less relevant (to us!) session blocks to spend some time on this instead, like the two I mentioned I wasn't likely to attend. One of the reasons I tried to get the wiki page filled out better before the conference, and that Josh started asking about "claimed" features, was so that it's easier to figure out which features involve which subsets of people. I intend to visit the conference with a chart of all that I'll fill in as I find useful groups we can divert onto these topics, and Josh takes good meeting notes too. As long as the useful advances make their way back onto the wiki or somebody's published project plan, it's OK if not everybody is at every discussion. I hope to attend all the sessions here, to continue helping with the public documentation, and I'll do what I can to help coordinate informal meetups too. Should be a busy and productive conference for everyone. -- Greg Smith 2ndQuadrant US Baltimore, MD PostgreSQL Training, Services and Support greg@2ndQuadrant.com www.2ndQuadrant.us
On Thu, 2010-05-13 at 12:12 -0700, Josh Berkus wrote: > So it depends on the purpose of the meeting. There'll be a Workshop meeting about Synchronous Replication at CHAR(10) on July 3, which has a specific focus. A clear focus for this list and the meeting at PGcon would be useful. I'm not completely clear what use case we're trying to solve here or whether we're trying to solve multiple use cases. -- Simon Riggs www.2ndQuadrant.com
> A clear focus for this list and the meeting at PGcon would be useful.
> I'm not completely clear what use case we're trying to solve here or
> whether we're trying to solve multiple use cases.
Yeah, my thought is that we already have some clustering topics on the
agenda for the developer meeting, and we have your Char(10) conference.
So the need for an internal discussion among clustering geeks is not
as important as the need to find more people who are interested in
working on clustering projects and/or documentation.
Hence, Friday night rather than Wednesday night.
BTW, why "CHAR10"? Because the word "clustering" has 10 characters in it?
--
-- Josh Berkus
PostgreSQL Experts Inc.
http://www.pgexperts.com
Josh Berkus wrote: > Yeah, my thought is that we already have some clustering topics on the > agenda for the developer meeting, and we have your Char(10) > conference. So the need for an internal discussion among clustering > geeks is not as important as the need to find more people who are > interested in working on clustering projects and/or documentation. The amount of time allocated to these topics at the developer's meeting is much larger than I had expected it to be. Has someone officially requested the BoF slot yet? > BTW, why "CHAR10"? Because the word "clustering" has 10 characters in > it? Hint: next years conference will be CHAR(11). -- Greg Smith 2ndQuadrant US Baltimore, MD PostgreSQL Training, Services and Support greg@2ndQuadrant.com www.2ndQuadrant.us
> The amount of time allocated to these topics at the developer's meeting
> is much larger than I had expected it to be. Has someone officially
> requested the BoF slot yet?
Yes, it's on the schedule.
> Hint: next years conference will be CHAR(11).
Oh, I see. Figures that I'd think of it in DB terms rather than as a
acronym.
--
-- Josh Berkus
PostgreSQL Experts Inc.
http://www.pgexperts.com
On Sun, 2010-05-16 at 11:20 -0700, Josh Berkus wrote: > BTW, why "CHAR10"? Because the word "clustering" has 10 characters in it? Clustering, High Availability & Replication, 2010 -- Simon Riggs www.2ndQuadrant.com
So, did this BOF happen? I had to leave on Friday evening before the BOFs. If it did happen, can someone give a report here? -- Greg Sabino Mullane greg@endpoint.com End Point Corporation PGP Key: 0x14964AC8
Attachment
Not actually. The last session (mine) took longer and the closing session took even longer. So we moved to the pub directly. Many clustering people gathered and we had very good time. I talked about XC thing with several other guys there but I don't know what was going on in total. Sorry for poor information. (2010年06月10日 01:06), Greg Sabino Mullane wrote: > So, did this BOF happen? I had to leave on Friday evening before > the BOFs. If it did happen, can someone give a report here? > >
----- Mensagem original ----- > Many clustering people gathered and we had very good time. I talked > about XC thing with several other guys there but I don't know what was > going on in total. I can help here a bit. There were a lot of people interested in clustering, but since there are many different solutions for different tasks theBOF turned into 3 or 4 groups discussing each technology. I remember people interested in Slony, pgpool, PostgresXC (including me), londiste, pgbouncer and plproxy. The discussion about XC was interesting. Fellows from NTT and EnterpriseDB are open to discuss how it will improve and howdesign changes and decisions can make a better software. I think the BOF started as a great way to keep in touch with interesting people and try to find ways to help those projects. Flavio Henrique A. Gurgel tel. 55-11-2125.4786 cel. 55-11-8389.7635 www.4linux.com.br FREE SOFTWARE SOLUTIONS
Greg,
As you've already read, the BOF turned into more of a social thing. The
closing session ran over, and by the time we started the BOF we had to
move to the pub or we'd have lost most of the attendees to hunger.
As it was, we had about 25 people in one wing of the pub. Several
clustering users introduced themselves and why they were interested in
clustering. Tellingly, even in this small sample (around 12 users) we
had a full range of reasons to cluster: failover, continuous HA, read
scalability, write scalability, burst traffic tolerance.
Then the various clustering projects present (Slony, PostgresXC, Skype,
pgPool2) introduced themselves. At that point, we divided into groups
discussing different clustering technologies.
--
-- Josh Berkus
PostgreSQL Experts Inc.
http://www.pgexperts.com