linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
* [LSFMMBPF TOPIC] Killing LSFMMBPF
@ 2020-03-06 14:35 Josef Bacik
  2020-03-06 15:29 ` Jason Gunthorpe
                   ` (9 more replies)
  0 siblings, 10 replies; 30+ messages in thread
From: Josef Bacik @ 2020-03-06 14:35 UTC (permalink / raw)
  To: lsf-pc, Linux FS Devel, linux-mm, linux-xfs, Btrfs BTRFS, bpf,
	linux-ext4, linux-block

Hello,

This has been a topic that I've been thinking about a lot recently, mostly 
because of the giant amount of work that has been organizing LSFMMBPF.  I was 
going to wait until afterwards to bring it up, hoping that maybe it was just me 
being done with the whole process and that time would give me a different 
perspective, but recent discussions has made it clear I'm not the only one.

LSFMMBPF is not useful to me personally, and not an optimal use of the 
communities time.  The things that we want to get out of LSFMMBPF are (generally)

1) Reach consensus on any multi-subsystem contentious changes that have come up 
over the past year.

2) Inform our fellow developers of new things that we are working on that we 
would like help with, or need to think about for the upcoming year.

3) "Hallway track".  We are after all a community, and I for one like spending 
time with developers that I don't get to interact with on a daily basis.

4) Provide a way to help integrate new developers into the community with face 
time.  It is far easier to work with people once you can put a face to a name, 
and this is especially valuable for new developers.

These are all really good goals, and why we love the idea of LSFMMBPF.  But 
having attended these things every year for the last 13 years, it has become 
less and less of these things, at least from my perspective.  A few problems (as 
I see them) are

1) The invitation process.  We've tried many different things, and I think we 
generally do a good job here, but the fact is if I don't know somebody I'm not 
going to give them a very high rating, making it difficult to actually bring in 
new people.

2) There are so many of us.  Especially with the addition of the BPF crowd we 
are now larger than ever.  This makes problem #1 even more apparent, even if I 
weighted some of the new people higher who's slot should they take instead?  I 
have 0 problems finding 20 people in the FS community who should absolutely be 
in the room.  But now I'm trying to squeeze in 1-5 extra people.  Propagate that 
across all the tracks and now we're at an extra 20ish people.

3) Half the people I want to talk to aren't even in the room.  This may be a 
uniquely file system track problem, but most of my work is in btrfs, and I want 
to talk to my fellow btrfs developers.  But again, we're trying to invite an 
entire community, so many of them simply don't request invitations, or just 
don't get invited.

3) Sponsorships.  This is still the best way to get to all of the core 
developers, so we're getting more and more sponsors in order to buy their slots 
to get access to people.  This is working as intended, and I'm not putting down 
our awesome sponsors, but this again adds to the amount of people that are 
showing up at what is supposed to be a working conference.

4) Presentations.  90% of the conference is 1-2 people standing at the front of 
the room, talking to a room of 20-100 people, with only a few people in the 
audience who cares.  We do our best to curate the presentations so we're not 
wasting peoples time, but in the end I don't care what David Howells is doing 
with mount, I trust him to do the right thing and he really just needs to trap 
Viro in a room to work it out, he doesn't need all of us.

5) Actually planning this thing.  I have been on the PC for at least the last 5 
years, and this year I'm running the whole thing.  We specifically laid out 
plans to rotate in new blood so this sort of thing stopped happening, and this 
year we've done a good job of that.  However it is a giant amount of work for 
anybody involved, especially for the whole conference chair.  Add in something 
like COVID-19 to the mix and now I just want to burn the whole thing to the 
ground.  Planning this thing is not free, it does require work and effort.

So what do I propose?  I propose we kill LSFMMBPF.

Many people have suggested this elsewhere, but I think we really need to 
seriously consider it.  Most of us all go to the Linux Plumbers conference.  We 
could accomplish our main goals with Plumbers without having to deal with all of 
the above problems.

1) The invitation process.  This goes away.  The people/companies that want to 
discuss things with the rest of us can all get to plumbers the normal way.  We 
get new blood that we may miss through the invitation process because they can 
simply register for Plumbers on their own.

2) Presentations.  We can have track miniconfs where we still curate talks, but 
there could be much less of them and we could just use the time to do what 
LSFMMBPF was meant to do, put us all in a room so we can hack on things together.

3) BOFs.  Now all of the xfs/btrfs/ext4 guys can show up, because again they 
don't have to worry about some invitation process, and now real meetings can 
happen between people that really want to talk to each other face to face.

4) Planning becomes much simpler.  I've organized miniconf's at plumbers before, 
it is far simpler than LSFMMBPF.  You only have to worry about one thing, is 
this presentation useful.  I no longer have to worry about am I inviting the 
right people, do we have enough money to cover the space.  Is there enough space 
for everybody?  Etc.

I think this is worth a discussion at the very least.  Maybe killing LSFMMBPF is 
too drastic, maybe there are some other ideas that would address the same 
problems.  I'd love to hear the whole communities thoughts on this, because 
after all this is supposed to be a community event, and we should all be heard. 
Thanks,



^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [LSFMMBPF TOPIC] Killing LSFMMBPF
  2020-03-06 14:35 [LSFMMBPF TOPIC] Killing LSFMMBPF Josef Bacik
@ 2020-03-06 15:29 ` Jason Gunthorpe
  2020-03-06 15:30 ` [Lsf-pc] " Amir Goldstein
                   ` (8 subsequent siblings)
  9 siblings, 0 replies; 30+ messages in thread
From: Jason Gunthorpe @ 2020-03-06 15:29 UTC (permalink / raw)
  To: Josef Bacik
  Cc: lsf-pc, Linux FS Devel, linux-mm, linux-xfs, Btrfs BTRFS, bpf,
	linux-ext4, linux-block

On Fri, Mar 06, 2020 at 09:35:41AM -0500, Josef Bacik wrote:

> 1) The invitation process.  This goes away.  The people/companies that want
> to discuss things with the rest of us can all get to plumbers the normal
> way.  We get new blood that we may miss through the invitation process
> because they can simply register for Plumbers on their own.

At last year at plumbers there were many people who could not get
tickets, it has been full the last few years, I think.

IMHO LPC is about at the size now where it is almost as large as it
can be in a mid-sized hotel setting..

> 4) Planning becomes much simpler.  I've organized miniconf's at plumbers
> before, it is far simpler than LSFMMBPF.  You only have to worry about one
> thing, is this presentation useful.  I no longer have to worry about am I
> inviting the right people, do we have enough money to cover the space.  Is
> there enough space for everybody?  Etc.

LPC does a great job at making miniconfs 'easy' - really fantastic
actually. I really appreciate how great a job they do on getting video
out and trying hard to mic everything so the freewheeling discussions
are audible to everyone.

Maybe something to think about is to keep the LSFMMBPF time slot but
instead of building a conference from scratch, copy LPC - same venue,
contracts, format, etc, etc. Ie two LPC style conferences a year, but
without having to plan two completely different sites from scratch.

IIRC when LPC was in Vancouver the LF used the same venue for several
conferences, I wonder how that worked out?

Jason


^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [Lsf-pc] [LSFMMBPF TOPIC] Killing LSFMMBPF
  2020-03-06 14:35 [LSFMMBPF TOPIC] Killing LSFMMBPF Josef Bacik
  2020-03-06 15:29 ` Jason Gunthorpe
@ 2020-03-06 15:30 ` Amir Goldstein
  2020-03-06 15:55 ` Josef Bacik
                   ` (7 subsequent siblings)
  9 siblings, 0 replies; 30+ messages in thread
From: Amir Goldstein @ 2020-03-06 15:30 UTC (permalink / raw)
  To: Josef Bacik
  Cc: lsf-pc, Linux FS Devel, Linux MM, linux-xfs, Btrfs BTRFS, bpf,
	Ext4, linux-block

On Fri, Mar 6, 2020 at 4:35 PM Josef Bacik <josef@toxicpanda.com> wrote:
>
> Hello,
>
> This has been a topic that I've been thinking about a lot recently, mostly
> because of the giant amount of work that has been organizing LSFMMBPF.  I was
> going to wait until afterwards to bring it up, hoping that maybe it was just me
> being done with the whole process and that time would give me a different
> perspective, but recent discussions has made it clear I'm not the only one.
>
> LSFMMBPF is not useful to me personally, and not an optimal use of the
> communities time.  The things that we want to get out of LSFMMBPF are (generally)
>
> 1) Reach consensus on any multi-subsystem contentious changes that have come up
> over the past year.
>
> 2) Inform our fellow developers of new things that we are working on that we
> would like help with, or need to think about for the upcoming year.
>
> 3) "Hallway track".  We are after all a community, and I for one like spending
> time with developers that I don't get to interact with on a daily basis.
>
> 4) Provide a way to help integrate new developers into the community with face
> time.  It is far easier to work with people once you can put a face to a name,
> and this is especially valuable for new developers.
>

5) There is another unspoken benefit that people wanted to get from LSF/MM (*)
and you mentioned it below that is to get the high level VFS/MM maintainer
in the room.

I think that was not always the case with Plumbers (not sure?), but if LF is
going the make sure that Plumbers stays co-located with the maintainers
summit and we "nominate" Plumbers as the official replacement for LSF/MM,
then this will probably sort itself out.

(*) I've intentionally left out BPF, because I think it always has a miniconf
of its own in Plumbers anyway.

> These are all really good goals, and why we love the idea of LSFMMBPF.  But
> having attended these things every year for the last 13 years, it has become
> less and less of these things, at least from my perspective.  A few problems (as
> I see them) are
>
> 1) The invitation process.  We've tried many different things, and I think we
> generally do a good job here, but the fact is if I don't know somebody I'm not
> going to give them a very high rating, making it difficult to actually bring in
> new people.
>
> 2) There are so many of us.  Especially with the addition of the BPF crowd we
> are now larger than ever.  This makes problem #1 even more apparent, even if I
> weighted some of the new people higher who's slot should they take instead?  I
> have 0 problems finding 20 people in the FS community who should absolutely be
> in the room.  But now I'm trying to squeeze in 1-5 extra people.  Propagate that
> across all the tracks and now we're at an extra 20ish people.
>
> 3) Half the people I want to talk to aren't even in the room.  This may be a
> uniquely file system track problem, but most of my work is in btrfs, and I want
> to talk to my fellow btrfs developers.  But again, we're trying to invite an
> entire community, so many of them simply don't request invitations, or just
> don't get invited.
>
> 3) Sponsorships.  This is still the best way to get to all of the core
> developers, so we're getting more and more sponsors in order to buy their slots
> to get access to people.  This is working as intended, and I'm not putting down
> our awesome sponsors, but this again adds to the amount of people that are
> showing up at what is supposed to be a working conference.
>
> 4) Presentations.  90% of the conference is 1-2 people standing at the front of
> the room, talking to a room of 20-100 people, with only a few people in the
> audience who cares.  We do our best to curate the presentations so we're not
> wasting peoples time, but in the end I don't care what David Howells is doing
> with mount, I trust him to do the right thing and he really just needs to trap
> Viro in a room to work it out, he doesn't need all of us.
>
> 5) Actually planning this thing.  I have been on the PC for at least the last 5
> years, and this year I'm running the whole thing.  We specifically laid out
> plans to rotate in new blood so this sort of thing stopped happening, and this
> year we've done a good job of that.  However it is a giant amount of work for
> anybody involved, especially for the whole conference chair.  Add in something
> like COVID-19 to the mix and now I just want to burn the whole thing to the
> ground.  Planning this thing is not free, it does require work and effort.
>
> So what do I propose?  I propose we kill LSFMMBPF.
>
> Many people have suggested this elsewhere, but I think we really need to
> seriously consider it.  Most of us all go to the Linux Plumbers conference.  We

Some of us have had to choose whether to go to LSF/MM or to Plumbers in a
given year. I know that merging them will make it easier for me.

Thanks,
Amir.


^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [LSFMMBPF TOPIC] Killing LSFMMBPF
  2020-03-06 14:35 [LSFMMBPF TOPIC] Killing LSFMMBPF Josef Bacik
  2020-03-06 15:29 ` Jason Gunthorpe
  2020-03-06 15:30 ` [Lsf-pc] " Amir Goldstein
@ 2020-03-06 15:55 ` Josef Bacik
  2020-03-06 15:56 ` Theodore Y. Ts'o
                   ` (6 subsequent siblings)
  9 siblings, 0 replies; 30+ messages in thread
From: Josef Bacik @ 2020-03-06 15:55 UTC (permalink / raw)
  To: lsf-pc, Linux FS Devel, linux-mm, linux-xfs, Btrfs BTRFS, bpf,
	linux-ext4, linux-block

On 3/6/20 9:35 AM, Josef Bacik wrote:
> Hello,
> 
> This has been a topic that I've been thinking about a lot recently, mostly 
> because of the giant amount of work that has been organizing LSFMMBPF.  I was 
> going to wait until afterwards to bring it up, hoping that maybe it was just me 
> being done with the whole process and that time would give me a different 
> perspective, but recent discussions has made it clear I'm not the only one.
> 
> LSFMMBPF is not useful to me personally, and not an optimal use of the 
> communities time.  The things that we want to get out of LSFMMBPF are (generally)

It has been pointed out to me that this appears to make it sound like I think 
the whole conference is useless.  I would like to make it clear that this is not 
the case at all.  This is still the only conference that I make sure to make it 
to every year, because all of the reasons I list.

The point of me posting this is to get us to put some real thought into what 
would be the most optimal way to accomplish the same things in a different way. 
Maybe I should have titled it "Make LSFMMBPF great again!" instead.

I feel that there is a lot of fat to trim here, and many voices not being heard 
because of the way the conference is organized.  If I'm the only one then that's 
cool, but if I'm not I'd like people to think really hard about what the ideal 
meetup looks like, and how we can move in that direction.  Thanks,

Josef


^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [LSFMMBPF TOPIC] Killing LSFMMBPF
  2020-03-06 14:35 [LSFMMBPF TOPIC] Killing LSFMMBPF Josef Bacik
                   ` (2 preceding siblings ...)
  2020-03-06 15:55 ` Josef Bacik
@ 2020-03-06 15:56 ` Theodore Y. Ts'o
  2020-03-06 16:08   ` Josef Bacik
                     ` (2 more replies)
  2020-03-06 16:04 ` [LSFMMBPF TOPIC] Killing LSFMMBPF Nikolay Borisov
                   ` (5 subsequent siblings)
  9 siblings, 3 replies; 30+ messages in thread
From: Theodore Y. Ts'o @ 2020-03-06 15:56 UTC (permalink / raw)
  To: Josef Bacik
  Cc: lsf-pc, Linux FS Devel, linux-mm, linux-xfs, Btrfs BTRFS, bpf,
	linux-ext4, linux-block

On Fri, Mar 06, 2020 at 09:35:41AM -0500, Josef Bacik wrote:
> This has been a topic that I've been thinking about a lot recently, mostly
> because of the giant amount of work that has been organizing LSFMMBPF.  I
> was going to wait until afterwards to bring it up, hoping that maybe it was
> just me being done with the whole process and that time would give me a
> different perspective, but recent discussions has made it clear I'm not the
> only one.....

I suggest that we try to decouple the question of should we have
LSF/MM/BPF in 2020 and COVID-19, with the question of what should
LSF/MM/BPF (perhaps in some transfigured form) should look like in
2021 and in the future.

A lot of the the concerns expressed in this e-mails are ones that I
have been concerned about, especially:

> 2) There are so many of us....

> 3) Half the people I want to talk to aren't even in the room.  This may be a
> uniquely file system track problem, but most of my work is in btrfs, and I
> want to talk to my fellow btrfs developers....

> 4) Presentations....

These *exactly* mirror the dynamic that we saw with the Kernel Summit,
and how we've migrated to a the Maintainer's Summit with a Kernel
centric track which is currently colocated with Plumbers.

I think it is still useful to have something where we reach consensus
on multi-subsystem contentious changes.  But I think those topics
could probably fit within a day or maybe a half day.  Does that sound
familiar?  That's essentially what we now have with the Maintainer'st
Summit.

The problem with Plumbers is that it's really, really full.  Not
having invitations doesn't magically go away; Plumbers last year had
to deal with long waitlist, and strugglinig to make sure that all of
the critical people who need be present so that the various Miniconfs
could be successful.

This is why I've been pushing so hard for a second Linux systems
focused event in the first half of the year.  I think if we colocate
the set of topics which are currently in LSF/MM, the more file system
specific presentations, the ext4/xfs/btrfs mini-summits/working
sessions, and the maintainer's summit / kernel summit, we would have
critical mass.  And I am sure there will be *plenty* of topics left
over for Plumbers.

Cheers,

						- Ted


^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [LSFMMBPF TOPIC] Killing LSFMMBPF
  2020-03-06 14:35 [LSFMMBPF TOPIC] Killing LSFMMBPF Josef Bacik
                   ` (3 preceding siblings ...)
  2020-03-06 15:56 ` Theodore Y. Ts'o
@ 2020-03-06 16:04 ` Nikolay Borisov
  2020-03-06 16:15 ` James Bottomley
                   ` (4 subsequent siblings)
  9 siblings, 0 replies; 30+ messages in thread
From: Nikolay Borisov @ 2020-03-06 16:04 UTC (permalink / raw)
  To: Josef Bacik, lsf-pc, Linux FS Devel, linux-mm, linux-xfs,
	Btrfs BTRFS, bpf, linux-ext4, linux-block



On 6.03.20 г. 16:35 ч., Josef Bacik wrote:
> Hello,
> 

<snip>

> 
> 1) The invitation process.  We've tried many different things, and I
> think we generally do a good job here, but the fact is if I don't know
> somebody I'm not going to give them a very high rating, making it
> difficult to actually bring in new people.

FWIW I've always found the invitation-only requirement somewhat
off-putting hence haven't attended.

<snip>


^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [LSFMMBPF TOPIC] Killing LSFMMBPF
  2020-03-06 15:56 ` Theodore Y. Ts'o
@ 2020-03-06 16:08   ` Josef Bacik
  2020-03-06 19:48     ` Theodore Y. Ts'o
  2020-03-06 18:30   ` Rik van Riel
  2020-03-07 18:54   ` [LSFMMBPF TOPIC] LSFMMBPF 2020 COVID-19 status update Luis Chamberlain
  2 siblings, 1 reply; 30+ messages in thread
From: Josef Bacik @ 2020-03-06 16:08 UTC (permalink / raw)
  To: Theodore Y. Ts'o
  Cc: lsf-pc, Linux FS Devel, linux-mm, linux-xfs, Btrfs BTRFS, bpf,
	linux-ext4, linux-block

On 3/6/20 10:56 AM, Theodore Y. Ts'o wrote:
> On Fri, Mar 06, 2020 at 09:35:41AM -0500, Josef Bacik wrote:
>> This has been a topic that I've been thinking about a lot recently, mostly
>> because of the giant amount of work that has been organizing LSFMMBPF.  I
>> was going to wait until afterwards to bring it up, hoping that maybe it was
>> just me being done with the whole process and that time would give me a
>> different perspective, but recent discussions has made it clear I'm not the
>> only one.....
> 
> I suggest that we try to decouple the question of should we have
> LSF/MM/BPF in 2020 and COVID-19, with the question of what should
> LSF/MM/BPF (perhaps in some transfigured form) should look like in
> 2021 and in the future.
> 

Yes this is purely about 2021 and the future, not 2020.

> A lot of the the concerns expressed in this e-mails are ones that I
> have been concerned about, especially:
> 
>> 2) There are so many of us....
> 
>> 3) Half the people I want to talk to aren't even in the room.  This may be a
>> uniquely file system track problem, but most of my work is in btrfs, and I
>> want to talk to my fellow btrfs developers....
> 
>> 4) Presentations....
> 
> These *exactly* mirror the dynamic that we saw with the Kernel Summit,
> and how we've migrated to a the Maintainer's Summit with a Kernel
> centric track which is currently colocated with Plumbers.
> 
> I think it is still useful to have something where we reach consensus
> on multi-subsystem contentious changes.  But I think those topics
> could probably fit within a day or maybe a half day.  Does that sound
> familiar?  That's essentially what we now have with the Maintainer'st
> Summit.
> 
> The problem with Plumbers is that it's really, really full.  Not
> having invitations doesn't magically go away; Plumbers last year had
> to deal with long waitlist, and strugglinig to make sure that all of
> the critical people who need be present so that the various Miniconfs
> could be successful.

Ah ok, I haven't done plumbers in a few years, I knew they would get full but I 
didn't think it was that bad.

> 
> This is why I've been pushing so hard for a second Linux systems
> focused event in the first half of the year.  I think if we colocate
> the set of topics which are currently in LSF/MM, the more file system
> specific presentations, the ext4/xfs/btrfs mini-summits/working
> sessions, and the maintainer's summit / kernel summit, we would have
> critical mass.  And I am sure there will be *plenty* of topics left
> over for Plumbers.
>

I'd be down for this.  Would you leave the thing open so anybody can register, 
or would you still have an invitation system?  I really, really despise the 
invitation system just because it's inherently self limiting.  However I do want 
to make sure we are getting relevant people in the room, and not making it this 
"oh shit, I forgot to register, and now the conference is full" sort of 
situations.  Thanks,

Josef


^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [LSFMMBPF TOPIC] Killing LSFMMBPF
  2020-03-06 14:35 [LSFMMBPF TOPIC] Killing LSFMMBPF Josef Bacik
                   ` (4 preceding siblings ...)
  2020-03-06 16:04 ` [LSFMMBPF TOPIC] Killing LSFMMBPF Nikolay Borisov
@ 2020-03-06 16:15 ` James Bottomley
  2020-03-06 16:28   ` Christian Brauner
       [not found] ` <20200306160548.GB25710@bombadil.infradead.org>
                   ` (3 subsequent siblings)
  9 siblings, 1 reply; 30+ messages in thread
From: James Bottomley @ 2020-03-06 16:15 UTC (permalink / raw)
  To: Josef Bacik, lsf-pc, Linux FS Devel, linux-mm, linux-xfs,
	Btrfs BTRFS, bpf, linux-ext4, linux-block

On Fri, 2020-03-06 at 09:35 -0500, Josef Bacik wrote:
> Many people have suggested this elsewhere, but I think we really need
> to seriously consider it.  Most of us all go to the Linux Plumbers
> conference.  We could accomplish our main goals with Plumbers without
> having to deal with all of the above problems.

[I'm on the Plumbers PC, but not speaking for them, just making general
observations based on my long history helping to run Plumbers]

Plumbers has basically reached the size where we can't realistically
expand without moving to the bigger venues and changing our evening
events ... it's already been a huge struggle in Lisbon and Halifax
trying to find a Restaurant big enough for the closing party.

The other reason for struggling to keep Plumbers around 500 is that the
value of simply running into people and having an accidental hallway
track, which is seen as a huge benefit of plumbers, starts diminishing.
 In fact, having a working hallway starts to become a problem as well
as we go up in numbers (plus in that survey we keep sending out those
who reply don't want plumbers to grow too much in size).

The other problem is content: you're a 3 day 4 track event and we're a
3 day 6 track event.  We get enough schedule angst from 6 tracks ... 10
would likely become hugely difficult.  If we move to 5 days, we'd have
to shove the Maintainer Summit on the Weekend (you can explain that one
to Linus) but we'd still be in danger of the day 4 burn out people used
to complain about when OLS and KS were co-located.

So, before you suggest Plumbers as the magic answer consider that the
problems you cite below don't magically go away, they just become
someone else's headache.

That's not to say this isn't a good idea, it's just to execute it we'd
have to transform Plumbers and we should have a community conversation
about that involving the current Plumbers PC before deciding it's the
best option.

James



^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [LSFMMBPF TOPIC] Killing LSFMMBPF
  2020-03-06 16:15 ` James Bottomley
@ 2020-03-06 16:28   ` Christian Brauner
  2020-03-06 16:31     ` Josef Bacik
  0 siblings, 1 reply; 30+ messages in thread
From: Christian Brauner @ 2020-03-06 16:28 UTC (permalink / raw)
  To: James Bottomley
  Cc: Josef Bacik, lsf-pc, Linux FS Devel, linux-mm, linux-xfs,
	Btrfs BTRFS, bpf, linux-ext4, linux-block

On Fri, Mar 06, 2020 at 08:15:10AM -0800, James Bottomley wrote:
> On Fri, 2020-03-06 at 09:35 -0500, Josef Bacik wrote:
> > Many people have suggested this elsewhere, but I think we really need
> > to seriously consider it.  Most of us all go to the Linux Plumbers
> > conference.  We could accomplish our main goals with Plumbers without
> > having to deal with all of the above problems.
> 
> [I'm on the Plumbers PC, but not speaking for them, just making general
> observations based on my long history helping to run Plumbers]
> 
> Plumbers has basically reached the size where we can't realistically
> expand without moving to the bigger venues and changing our evening
> events ... it's already been a huge struggle in Lisbon and Halifax
> trying to find a Restaurant big enough for the closing party.
> 
> The other reason for struggling to keep Plumbers around 500 is that the
> value of simply running into people and having an accidental hallway
> track, which is seen as a huge benefit of plumbers, starts diminishing.
>  In fact, having a working hallway starts to become a problem as well
> as we go up in numbers (plus in that survey we keep sending out those
> who reply don't want plumbers to grow too much in size).
> 
> The other problem is content: you're a 3 day 4 track event and we're a
> 3 day 6 track event.  We get enough schedule angst from 6 tracks ... 10
> would likely become hugely difficult.  If we move to 5 days, we'd have
> to shove the Maintainer Summit on the Weekend (you can explain that one
> to Linus) but we'd still be in danger of the day 4 burn out people used
> to complain about when OLS and KS were co-located.
> 
> So, before you suggest Plumbers as the magic answer consider that the
> problems you cite below don't magically go away, they just become
> someone else's headache.
> 
> That's not to say this isn't a good idea, it's just to execute it we'd
> have to transform Plumbers and we should have a community conversation
> about that involving the current Plumbers PC before deciding it's the
> best option.

It's unlikely that this could still be done given that we're also facing
a little uncertainty for Plumbers. It seems like a lot of additional
syncing would be needed.
But the main concern I have is that co-locating both is probably quite
challenging for anyone attending both especially when organizing
something like a microconference.


^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [LSFMMBPF TOPIC] Killing LSFMMBPF
  2020-03-06 16:28   ` Christian Brauner
@ 2020-03-06 16:31     ` Josef Bacik
  0 siblings, 0 replies; 30+ messages in thread
From: Josef Bacik @ 2020-03-06 16:31 UTC (permalink / raw)
  To: Christian Brauner, James Bottomley
  Cc: lsf-pc, Linux FS Devel, linux-mm, linux-xfs, Btrfs BTRFS, bpf,
	linux-ext4, linux-block

On 3/6/20 11:28 AM, Christian Brauner wrote:
> On Fri, Mar 06, 2020 at 08:15:10AM -0800, James Bottomley wrote:
>> On Fri, 2020-03-06 at 09:35 -0500, Josef Bacik wrote:
>>> Many people have suggested this elsewhere, but I think we really need
>>> to seriously consider it.  Most of us all go to the Linux Plumbers
>>> conference.  We could accomplish our main goals with Plumbers without
>>> having to deal with all of the above problems.
>>
>> [I'm on the Plumbers PC, but not speaking for them, just making general
>> observations based on my long history helping to run Plumbers]
>>
>> Plumbers has basically reached the size where we can't realistically
>> expand without moving to the bigger venues and changing our evening
>> events ... it's already been a huge struggle in Lisbon and Halifax
>> trying to find a Restaurant big enough for the closing party.
>>
>> The other reason for struggling to keep Plumbers around 500 is that the
>> value of simply running into people and having an accidental hallway
>> track, which is seen as a huge benefit of plumbers, starts diminishing.
>>   In fact, having a working hallway starts to become a problem as well
>> as we go up in numbers (plus in that survey we keep sending out those
>> who reply don't want plumbers to grow too much in size).
>>
>> The other problem is content: you're a 3 day 4 track event and we're a
>> 3 day 6 track event.  We get enough schedule angst from 6 tracks ... 10
>> would likely become hugely difficult.  If we move to 5 days, we'd have
>> to shove the Maintainer Summit on the Weekend (you can explain that one
>> to Linus) but we'd still be in danger of the day 4 burn out people used
>> to complain about when OLS and KS were co-located.
>>
>> So, before you suggest Plumbers as the magic answer consider that the
>> problems you cite below don't magically go away, they just become
>> someone else's headache.
>>
>> That's not to say this isn't a good idea, it's just to execute it we'd
>> have to transform Plumbers and we should have a community conversation
>> about that involving the current Plumbers PC before deciding it's the
>> best option.
> 
> It's unlikely that this could still be done given that we're also facing
> a little uncertainty for Plumbers. It seems like a lot of additional
> syncing would be needed.
> But the main concern I have is that co-locating both is probably quite
> challenging for anyone attending both especially when organizing
> something like a microconference.
> 

Yeah I want to be clear I'm not talking about this years conference, I'm talking 
about future conferences and if/how we want to make changes.

I picked plumbers because by-in-large the overlap between plumbers attendance 
and LSFMMBPF attendance is pretty large, but obviously it doesn't have to be 
just that.  Ted and others have suggested having a larger more inclusive 
conference opposite of plumbers, which I think is a really cool idea.  Thanks,

Josef


^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [LSFMMBPF TOPIC] Killing LSFMMBPF
       [not found] ` <20200306160548.GB25710@bombadil.infradead.org>
@ 2020-03-06 17:04   ` Al Viro
  2020-03-06 17:37   ` James Bottomley
  1 sibling, 0 replies; 30+ messages in thread
From: Al Viro @ 2020-03-06 17:04 UTC (permalink / raw)
  To: Matthew Wilcox
  Cc: Josef Bacik, lsf-pc, Linux FS Devel, linux-mm, linux-xfs,
	Btrfs BTRFS, bpf, linux-ext4, linux-block

On Fri, Mar 06, 2020 at 08:05:48AM -0800, Matthew Wilcox wrote:

> > 4) Presentations.  90% of the conference is 1-2 people standing at the front
> > of the room, talking to a room of 20-100 people, with only a few people in
> > the audience who cares.  We do our best to curate the presentations so we're
> > not wasting peoples time, but in the end I don't care what David Howells is
> > doing with mount, I trust him to do the right thing and he really just needs
> > to trap Viro in a room to work it out, he doesn't need all of us.
> 
> ... and allow the other 3-5 people who're interested or affected the
> opportunity to sit in.  Like a mailing list, but higher bandwidth.

Latency can be more unpleasant, actually - you try to discuss something between
3 people, when one is in .uk, another - in .us (east coast) and the third one -
in ,au (also east coast).  Timezone deltas - 5 hours and 8 hours, in opposite
directions...  Incidentally, that was about mount, with me and David being
two of participants; I somewhat hoped to get that sorted out at LSF, but Ian
won't be there anyway.  OTOH, that's just an 8 hours delta...


^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [LSFMMBPF TOPIC] Killing LSFMMBPF
       [not found] ` <20200306160548.GB25710@bombadil.infradead.org>
  2020-03-06 17:04   ` Al Viro
@ 2020-03-06 17:37   ` James Bottomley
  2020-03-06 18:06     ` Jason Gunthorpe
  2020-03-06 18:23     ` Matthew Wilcox
  1 sibling, 2 replies; 30+ messages in thread
From: James Bottomley @ 2020-03-06 17:37 UTC (permalink / raw)
  To: Matthew Wilcox, Josef Bacik
  Cc: lsf-pc, Linux FS Devel, linux-mm, linux-xfs, Btrfs BTRFS, bpf,
	linux-ext4, linux-block

On Fri, 2020-03-06 at 08:05 -0800, Matthew Wilcox wrote:
[...]
> 2. Charge attendees $300 for a 3-day conference.  This seems to be
> the going rate (eg BSDCan, PGCon).  This allows the conference to be
> self-funding without sponsors, and any sponsorship can go towards
> evening events, food, travel bursaries, etc.

Can I just inject a dose of reality here:  The most costly thing is
Venue rental (which comes with a F&B minimum) and the continuous Tea
and Coffee.  Last year for Plumbers, the venue cost us $37k and the
breaks $132k (including a lunch buffet, which was a requirement of the
venue rental).  Given we had 500 attendees, that, alone is $340 per
head already.  Now we could cut out the continuous tea and coffee ...
and the espresso machines you all raved about last year cost us about
$7 per shot.  But it's not just this, it's also AV (microphones and
projectors) and recording, and fast internet access.  That all came to
about $100k last year (or an extra $200 per head).  So you can see,
running at the level Plumbers does you're already looking at $540 a
head, which, co-incidentally is close to our attendee fee.  To get to
$300 per head, you lot will have to give up something in addition to
the espresso machines, what is it to be?

James



^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [LSFMMBPF TOPIC] Killing LSFMMBPF
  2020-03-06 17:37   ` James Bottomley
@ 2020-03-06 18:06     ` Jason Gunthorpe
  2020-03-06 19:07       ` Martin K. Petersen
  2020-03-06 18:23     ` Matthew Wilcox
  1 sibling, 1 reply; 30+ messages in thread
From: Jason Gunthorpe @ 2020-03-06 18:06 UTC (permalink / raw)
  To: James Bottomley
  Cc: Matthew Wilcox, Josef Bacik, lsf-pc, Linux FS Devel, linux-mm,
	linux-xfs, Btrfs BTRFS, bpf, linux-ext4, linux-block

On Fri, Mar 06, 2020 at 09:37:59AM -0800, James Bottomley wrote:
> On Fri, 2020-03-06 at 08:05 -0800, Matthew Wilcox wrote:
> [...]
> > 2. Charge attendees $300 for a 3-day conference.  This seems to be
> > the going rate (eg BSDCan, PGCon).  This allows the conference to be
> > self-funding without sponsors, and any sponsorship can go towards
> > evening events, food, travel bursaries, etc.
> 
> Can I just inject a dose of reality here:  The most costly thing is
> Venue rental (which comes with a F&B minimum) and the continuous Tea
> and Coffee.  Last year for Plumbers, the venue cost us $37k and the
> breaks $132k (including a lunch buffet, which was a requirement of the
> venue rental).  Given we had 500 attendees, that, alone is $340 per
> head already.  Now we could cut out the continuous tea and coffee ...
> and the espresso machines you all raved about last year cost us about
> $7 per shot.  But it's not just this, it's also AV (microphones and
> projectors) and recording, and fast internet access.  That all came to
> about $100k last year (or an extra $200 per head).  So you can see,
> running at the level Plumbers does you're already looking at $540 a
> head, which, co-incidentally is close to our attendee fee.  To get to
> $300 per head, you lot will have to give up something in addition to
> the espresso machines, what is it to be?

Yes, I can confirm this from another smaller hotel-style conference
I've been involved organizing on occasion. $600-$800 is required to
break even without major sponsorship $$ for the ~100 people mark, and
that is without the usual food and venue perks we see at
plumbers/lsfmm.

Jason


^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [LSFMMBPF TOPIC] Killing LSFMMBPF
  2020-03-06 17:37   ` James Bottomley
  2020-03-06 18:06     ` Jason Gunthorpe
@ 2020-03-06 18:23     ` Matthew Wilcox
  2020-03-06 19:25       ` James Bottomley
  1 sibling, 1 reply; 30+ messages in thread
From: Matthew Wilcox @ 2020-03-06 18:23 UTC (permalink / raw)
  To: James Bottomley
  Cc: Josef Bacik, lsf-pc, Linux FS Devel, linux-mm, linux-xfs,
	Btrfs BTRFS, bpf, linux-ext4, linux-block

On Fri, Mar 06, 2020 at 09:37:59AM -0800, James Bottomley wrote:
> Can I just inject a dose of reality here:  The most costly thing is
> Venue rental (which comes with a F&B minimum) and the continuous Tea
> and Coffee.  Last year for Plumbers, the venue cost us $37k and the
> breaks $132k (including a lunch buffet, which was a requirement of the
> venue rental).  Given we had 500 attendees, that, alone is $340 per
> head already.  Now we could cut out the continuous tea and coffee ...
> and the espresso machines you all raved about last year cost us about
> $7 per shot.  But it's not just this, it's also AV (microphones and
> projectors) and recording, and fast internet access.  That all came to
> about $100k last year (or an extra $200 per head).  So you can see,
> running at the level Plumbers does you're already looking at $540 a
> head, which, co-incidentally is close to our attendee fee.  To get to
> $300 per head, you lot will have to give up something in addition to
> the espresso machines, what is it to be?

I was basing that on https://www.bsdcan.org/2020/registration.php
which is a ~200 person conference, charging $200 for 2 days.  They
provide morning & afternoon snacks as well as lunch and coffee.


^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [LSFMMBPF TOPIC] Killing LSFMMBPF
  2020-03-06 15:56 ` Theodore Y. Ts'o
  2020-03-06 16:08   ` Josef Bacik
@ 2020-03-06 18:30   ` Rik van Riel
  2020-03-07 18:54   ` [LSFMMBPF TOPIC] LSFMMBPF 2020 COVID-19 status update Luis Chamberlain
  2 siblings, 0 replies; 30+ messages in thread
From: Rik van Riel @ 2020-03-06 18:30 UTC (permalink / raw)
  To: Theodore Y. Ts'o, Josef Bacik
  Cc: lsf-pc, Linux FS Devel, linux-mm, linux-xfs, Btrfs BTRFS, bpf,
	linux-ext4, linux-block

[-- Attachment #1: Type: text/plain, Size: 1818 bytes --]

On Fri, 2020-03-06 at 10:56 -0500, Theodore Y. Ts'o wrote:

> The problem with Plumbers is that it's really, really full.  Not
> having invitations doesn't magically go away; Plumbers last year had
> to deal with long waitlist, and strugglinig to make sure that all of
> the critical people who need be present so that the various Miniconfs
> could be successful.
> 
> This is why I've been pushing so hard for a second Linux systems
> focused event in the first half of the year.  I think if we colocate
> the set of topics which are currently in LSF/MM, the more file system
> specific presentations, the ext4/xfs/btrfs mini-summits/working
> sessions, and the maintainer's summit / kernel summit, we would have
> critical mass.  And I am sure there will be *plenty* of topics left
> over for Plumbers.

That sounds like a good idea to me, as well.

Instead of trying to invite "all the useful people" on
a few fixed topics, which LSF/MM did very well, but which
the Linux community has long outgrown, resulting in us
being unable to invite some very good people, we can turn
things on its head a little organization wise.

We can host a number of (half day?) mini conferences on
various Linux kernel topics, maybe still with some focus
around LSF/MM/IO, and have a small number of general track
discussion sessions.

Once the topics of the mini conferences have been chosen,
people can figure out whether or not they want to attend
that year.

This could help us avoid the "oops, we couldn't invite two
of the people who really should have been here for this 
discussion" issue that has been difficult to avoid with 
LSF/MM having more interest, and more topics, than available
slots for people to attend.

TL;DR: decide on topics, not invitees.

-- 
All Rights Reversed.

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [LSFMMBPF TOPIC] Killing LSFMMBPF
  2020-03-06 18:06     ` Jason Gunthorpe
@ 2020-03-06 19:07       ` Martin K. Petersen
  2020-03-06 19:15         ` James Bottomley
  0 siblings, 1 reply; 30+ messages in thread
From: Martin K. Petersen @ 2020-03-06 19:07 UTC (permalink / raw)
  To: Jason Gunthorpe
  Cc: James Bottomley, Matthew Wilcox, Josef Bacik, lsf-pc,
	Linux FS Devel, linux-mm, linux-xfs, Btrfs BTRFS, bpf,
	linux-ext4, linux-block


Jason,

> Yes, I can confirm this from another smaller hotel-style conference
> I've been involved organizing on occasion. $600-$800 is required to
> break even without major sponsorship $$ for the ~100 people mark, and
> that is without the usual food and venue perks we see at
> plumbers/lsfmm.

Yep. Our actual per-person cost for LSF/MM/BPF is in excess of $1K. That
limits who we can invite. Personally I absolutely hate the invitation
aspect and process. But we are very constrained wrt. how many we can
actually accommodate by the amount of funding we get. Things appear to
be better this year, but sponsor mergers and acquisitions have been a
major concern the past few years.

The premise of LSF/MM/BPF is to provide a venue where the right people
can talk low-latency, face to face. Without the distractions of a 1000
person event setting. The reason LSF/MM/BPF has been free to attend has
been to ensure that attendance fees wouldn't be a deterrent for the
people who should be there. The downside is that the invitation process
has been a deterrent for other, likely valuable, contributors.

I would love for LSF/MM/BPF/BBQ to be an umbrella event like LPC where
we could have miniconfs with all the relevant contributors for each
topic area to be present. The addition of the 3rd day was done to
facilitate that so that XFS folks, btrfs folks, etc. could congregate in
a room to discuss things only they cared about. But the current
attendance headcount cap means that not all topics can be covered due to
crucial people missing.

Also, there are several areas where I do think that the present LSF/MM
format still has merit. First of all, not all topics are large enough to
justify an entire miniconf or topic-specific workshop. We have many
topics that can be covered in an hour or less and that's the end of
that. The other aspect is that key people straddle multiple filesystems,
subsystems, etc. If we *only* had XFS/btrfs/BPF miniconfs, scheduling
would be near impossible. Hence the current division between scheduled
days and workshop day. Also, we do have cross-track topics that need
involvement across the board. I would personally be happy with 1 track
day and 2 workshop days if we could get critical mass for the workshop
topics.

In the old days, when LSF tracks were 10-12 people each, I felt we got
stuff done. Since then we have more than doubled the headcount for each
track in an attempt to get more people involved. But I feel that the
discussions are much less useful. Despite enforcing the no-slides rules,
etc.

If we combine sponsor funding with per-attendee fees to facilitate a
larger event, the question becomes: What should the headcount limit be?
200? 500? The reason I ask is that I think funding can be worked
out. But I also think it is important enough that we don't exceed the
"productive group size" too much for a given topic. And we usually put
that somewhere between 10 and 15. It is very rare to see more than this
many attendees actively participate in a discussion. This means for an
attendee cap of 200, we should aim to have ~20 concurrent topics
happening for it to be productive. Maybe slash that number in half to
compensate for the people in the hallway tracks?

One thing a few of us discussed a year or two ago was to have actual
per-session headcount limits. And make people bid on the sessions they
wanted to participate in and then cap each session at 15. That would
obviously be very hard to schedule and enforce. But I still think we
need to think about how we can bring N hundred people together and make
sure they congregate in productive groups of 10-15. That's really the
key as far as I'm concerned. We have tried the pure unconference
approach and that wasn't very productive either. So we need to land
somewhere in the middle...

-- 
Martin K. Petersen	Oracle Linux Engineering


^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [LSFMMBPF TOPIC] Killing LSFMMBPF
  2020-03-06 19:07       ` Martin K. Petersen
@ 2020-03-06 19:15         ` James Bottomley
  2020-03-06 19:20           ` Martin K. Petersen
  0 siblings, 1 reply; 30+ messages in thread
From: James Bottomley @ 2020-03-06 19:15 UTC (permalink / raw)
  To: Martin K. Petersen, Jason Gunthorpe
  Cc: Matthew Wilcox, Josef Bacik, lsf-pc, Linux FS Devel, linux-mm,
	linux-xfs, Btrfs BTRFS, bpf, linux-ext4, linux-block

On Fri, 2020-03-06 at 14:07 -0500, Martin K. Petersen wrote:
> Jason,
> 
> > Yes, I can confirm this from another smaller hotel-style conference
> > I've been involved organizing on occasion. $600-$800 is required to
> > break even without major sponsorship $$ for the ~100 people mark,
> > and that is without the usual food and venue perks we see at
> > plumbers/lsfmm.
> 
> Yep. Our actual per-person cost for LSF/MM/BPF is in excess of $1K.
> That limits who we can invite. Personally I absolutely hate the
> invitation aspect and process. But we are very constrained wrt. how
> many we can actually accommodate by the amount of funding we get.
> Things appear to be better this year, but sponsor mergers and
> acquisitions have been a major concern the past few years.

To be a bit mercenary (hey, it's my job, I'm Plumbers treasurer this
year) our sponsors are mostly the same companies.  If we combine LSF/MM
and Plumbers, I can't see too many of them stepping up to sponsor us
twice, so we'll have a net loss of sponsor funding for the combined
event as well.  This is likely another argument for doing two
separately sponsored events.

James



^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [LSFMMBPF TOPIC] Killing LSFMMBPF
  2020-03-06 19:15         ` James Bottomley
@ 2020-03-06 19:20           ` Martin K. Petersen
  0 siblings, 0 replies; 30+ messages in thread
From: Martin K. Petersen @ 2020-03-06 19:20 UTC (permalink / raw)
  To: James Bottomley
  Cc: Martin K. Petersen, Jason Gunthorpe, Matthew Wilcox, Josef Bacik,
	lsf-pc, Linux FS Devel, linux-mm, linux-xfs, Btrfs BTRFS, bpf,
	linux-ext4, linux-block


James,

> To be a bit mercenary (hey, it's my job, I'm Plumbers treasurer this
> year) our sponsors are mostly the same companies.  If we combine
> LSF/MM and Plumbers, I can't see too many of them stepping up to
> sponsor us twice, so we'll have a net loss of sponsor funding for the
> combined event as well.  This is likely another argument for doing two
> separately sponsored events.

Yep. And I do think it's beneficial to have two developer-focused events
per year.

-- 
Martin K. Petersen	Oracle Linux Engineering


^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [LSFMMBPF TOPIC] Killing LSFMMBPF
  2020-03-06 18:23     ` Matthew Wilcox
@ 2020-03-06 19:25       ` James Bottomley
  0 siblings, 0 replies; 30+ messages in thread
From: James Bottomley @ 2020-03-06 19:25 UTC (permalink / raw)
  To: Matthew Wilcox
  Cc: Josef Bacik, lsf-pc, Linux FS Devel, linux-mm, linux-xfs,
	Btrfs BTRFS, bpf, linux-ext4, linux-block

On Fri, 2020-03-06 at 10:23 -0800, Matthew Wilcox wrote:
> On Fri, Mar 06, 2020 at 09:37:59AM -0800, James Bottomley wrote:
> > Can I just inject a dose of reality here:  The most costly thing is
> > Venue rental (which comes with a F&B minimum) and the continuous
> > Tea and Coffee.  Last year for Plumbers, the venue cost us $37k and
> > the breaks $132k (including a lunch buffet, which was a requirement
> > of the venue rental).  Given we had 500 attendees, that, alone is
> > $340 per head already.  Now we could cut out the continuous tea and
> > coffee ... and the espresso machines you all raved about last year
> > cost us about $7 per shot.  But it's not just this, it's also AV
> > (microphones and projectors) and recording, and fast internet
> > access.  That all came to about $100k last year (or an extra $200
> > per head).  So you can see, running at the level Plumbers does
> > you're already looking at $540 a head, which, co-incidentally is
> > close to our attendee fee.  To get to $300 per head, you lot will
> > have to give up something in addition to the espresso machines,
> > what is it to be?
> 
> I was basing that on https://www.bsdcan.org/2020/registration.php
> which is a ~200 person conference, charging $200 for 2 days.  They
> provide morning & afternoon snacks as well as lunch and coffee.

I didn't say you couldn't run a conference for this low, I was just
point out what Plumbers currently costs.  FOSDEM clearly manages in
Europe for free, but what they have to give up is huge: No
tea/coffee/lunch/breakfast at all.  Doing it in a University helps with
A/V, and venue rental, but you're severely constrained by their
timetable (Ethan Miller did offer us UC Santa Cruz one year for LSF/MM
but we eventually concluded that date constraints and logistics would
just be too difficult).

James



^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [LSFMMBPF TOPIC] long live LFSMMBPF
  2020-03-06 14:35 [LSFMMBPF TOPIC] Killing LSFMMBPF Josef Bacik
                   ` (6 preceding siblings ...)
       [not found] ` <20200306160548.GB25710@bombadil.infradead.org>
@ 2020-03-06 19:27 ` Chris Mason
  2020-03-06 19:41   ` James Bottomley
  2020-03-07  3:14 ` [LSFMMBPF TOPIC] Killing LSFMMBPF Steve French
  2020-03-10 13:13 ` Michal Hocko
  9 siblings, 1 reply; 30+ messages in thread
From: Chris Mason @ 2020-03-06 19:27 UTC (permalink / raw)
  To: Josef Bacik
  Cc: lsf-pc, Linux FS Devel, linux-mm, linux-xfs, Btrfs BTRFS, bpf,
	linux-ext4, linux-block

On 6 Mar 2020, at 9:35, Josef Bacik wrote:
>
> Many people have suggested this elsewhere, but I think we really need 
> to seriously consider it.  Most of us all go to the Linux Plumbers 
> conference.  We could accomplish our main goals with Plumbers without 
> having to deal with all of the above problems.

I think James and Ted have covered pretty well why Plumbers isn’t a 
great fit, but I agree with the overall idea.

>
> 1) The invitation process.  This goes away.  The people/companies that 
> want to discuss things with the rest of us can all get to plumbers the 
> normal way.  We get new blood that we may miss through the invitation 
> process because they can simply register for Plumbers on their own.
>

Lsfmmmbop has always been most useful when focused on smaller and 
tighter sessions that aren’t well suited to open audiences.  I think 
the BPF and MM sessions are generally really happy with their size and 
level of discussion, while the FS one would benefit from a larger crowd 
split up by project.  This is much easier to do if we’re attached to a 
bigger conference, where the plenary sessions are available to the whole 
conf and the breakout sessions are smaller and completely project 
focused.

I think we’ve outgrown the original name, but I’d still call it 
something, we’ll need rooms and t-shirts and maybe a group event that 
we need to fund.

> 2) Presentations.  We can have track miniconfs where we still curate 
> talks, but there could be much less of them and we could just use the 
> time to do what LSFMMBPF was meant to do, put us all in a room so we 
> can hack on things together.

Agree here, although kernel recipes is a great example of a conf people 
visit for the presentations.

>
> 3) BOFs.  Now all of the xfs/btrfs/ext4 guys can show up, because 
> again they don't have to worry about some invitation process, and now 
> real meetings can happen between people that really want to talk to 
> each other face to face.
>
> 4) Planning becomes much simpler.  I've organized miniconf's at 
> plumbers before, it is far simpler than LSFMMBPF.  You only have to 
> worry about one thing, is this presentation useful.  I no longer have 
> to worry about am I inviting the right people, do we have enough money 
> to cover the space.  Is there enough space for everybody?  Etc.

We’ve talked about working closely with KS, Plumbers and the 
Linuxfoundation to make a big picture map of the content and frequency 
for these confs.  I’m sure Angela is having a busy few weeks, but lets 
work with her to schedule this and talk it through.  OSS is a good fit 
in terms of being flexible enough to fit us in, hopefully we can make 
that work.

-chris



^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [LSFMMBPF TOPIC] long live LFSMMBPF
  2020-03-06 19:27 ` [LSFMMBPF TOPIC] long live LFSMMBPF Chris Mason
@ 2020-03-06 19:41   ` James Bottomley
  2020-03-06 19:56     ` Chris Mason
  2020-03-06 20:25     ` Theodore Y. Ts'o
  0 siblings, 2 replies; 30+ messages in thread
From: James Bottomley @ 2020-03-06 19:41 UTC (permalink / raw)
  To: Chris Mason, Josef Bacik
  Cc: lsf-pc, Linux FS Devel, linux-mm, linux-xfs, Btrfs BTRFS, bpf,
	linux-ext4, linux-block

On Fri, 2020-03-06 at 14:27 -0500, Chris Mason wrote:
> On 6 Mar 2020, at 9:35, Josef Bacik wrote:
[...]
> > 4) Planning becomes much simpler.  I've organized miniconf's at 
> > plumbers before, it is far simpler than LSFMMBPF.  You only have
> > to worry about one thing, is this presentation useful.  I no longer
> > have to worry about am I inviting the right people, do we have
> > enough money to cover the space.  Is there enough space for
> > everybody?  Etc.
> 
> We’ve talked about working closely with KS, Plumbers and the 
> Linuxfoundation to make a big picture map of the content and
> frequency  for these confs.

And, lest anyone think we all operate in isolation, we do get together
periodically to discuss venues, selection and combination.  The last
big in-person meeting on this topic was at Plumbers in Vancouver in
2019, where we had Plumbers, KS/MS, LSF/MM and the LF conference people
all represented.

>   I’m sure Angela is having a busy few weeks, but lets work with her
> to schedule this and talk it through.  OSS is a good fit  in terms of
> being flexible enough to fit us in, hopefully we can make  that work.

And, for everyone who gave us feedback in the Plumbers surveys that co-
locating with a big conference is *not* what you want because of
various problems like hallway track disruptions due to other conference
traffic and simply the difficulty of finding people, the current model
under consideration is one conference organization (the LF) but two
separate venues, sort of like OpenStack used to do for their big
conference and design summit to minimize disruption and increase
developer focus.

James



^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [LSFMMBPF TOPIC] Killing LSFMMBPF
  2020-03-06 16:08   ` Josef Bacik
@ 2020-03-06 19:48     ` Theodore Y. Ts'o
  0 siblings, 0 replies; 30+ messages in thread
From: Theodore Y. Ts'o @ 2020-03-06 19:48 UTC (permalink / raw)
  To: Josef Bacik
  Cc: lsf-pc, Linux FS Devel, linux-mm, linux-xfs, Btrfs BTRFS, bpf,
	linux-ext4, linux-block

On Fri, Mar 06, 2020 at 11:08:36AM -0500, Josef Bacik wrote:
> 
> I'd be down for this.  Would you leave the thing open so anybody can
> register, or would you still have an invitation system?  I really, really
> despise the invitation system just because it's inherently self limiting.
> However I do want to make sure we are getting relevant people in the room,
> and not making it this "oh shit, I forgot to register, and now the
> conference is full" sort of situations.  Thanks,

There are lots of different ways it can be done.  The Maintainer's
Summit is an invite-only half-day event.  That's mainly because it's
about development processes, and there are lots of people who have a
strong interest in that, but we want to keep it done to small number
of people so we can have real conversations.

At Plumbers, the miniconfs leads can give a list of (six?) people they
really want to be present.  A few get free registration; the others
get guaranteed registrations thus bypassing the waitlist.  One of the
problems is that the miniconf leads don't always get the list of
people to the planning committee until late in the process, which made
the waitlist management problem even more painful.  At the miniconf,
there is social pressure so that the key attendees are seated near the
front of the room, and there might be audience of a few hundred that
are in listen-mostly mode, but for most technical topics, that isn't
that much of a problem.

I've also seen other cases where the room is small, and there is a
list of people who have guaranteed access to the room, and everyone
else (up to the fire limit) might have to sit or stand against the
wall, etc.

If we have a conference with many tracks, the different tracks can
have different admittance policies, such is as the case with the
Maintainer's Summit, Kernel Summit, Miniconfs, etc.  So that's
something which I think can be negotiated.

I suspect that for most of the LSF/MM contential topics, I doubt we
would have hundreds of people clamoring to get in on a discussion
about to handle, say, clearing DAX flag on files that might still be
in use by some RDMA drive.  That is *such* a fascinating topic, but I
doubt there really will be a need to limit attendance.  :-)

      	    	   	     	     - Ted

P.S.  I do need to note that there is one big advantage to invite-only
summts such as the LSF/MM and the old-style Kernel Summit.  Companies
who really want to present about, say, dual-actuator HDD's, or the
latest NVMe / Open Channel interface, are much more likely to pay $$$
to get access to an invite-only event.  When we moved to the
process-only Maintainer's Summit, and the Kernel Summit for the
technical tracks, it most definitely hurt the amount of sponsorship
dollars that we got for the Maintainer's Summit.

That's not a bad thing, but it might mean that we will need to cut
costs by drafting behind the LF, and maybe not having as nice evening
receptions, or as nice attendee gifts like we used to do in the early
years of the Kernel Summit.  Personally, I think that's *fine*; it's
the collaboration with fellow developers which is highest on my list
of priorities, and not the opportunities for fine dining or going to
fun cities.  And if giving up on some of these amenities means that I
can bring some of the more junior engineers on my team so they can
meet other file system developers, I'm **all** for it.  But for some
folks, they may view this tradeoff as a loss.


^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [LSFMMBPF TOPIC] long live LFSMMBPF
  2020-03-06 19:41   ` James Bottomley
@ 2020-03-06 19:56     ` Chris Mason
  2020-03-06 20:25     ` Theodore Y. Ts'o
  1 sibling, 0 replies; 30+ messages in thread
From: Chris Mason @ 2020-03-06 19:56 UTC (permalink / raw)
  To: James Bottomley
  Cc: Josef Bacik, lsf-pc, Linux FS Devel, linux-mm, linux-xfs,
	Btrfs BTRFS, bpf, linux-ext4, linux-block

On 6 Mar 2020, at 14:41, James Bottomley wrote:

> On Fri, 2020-03-06 at 14:27 -0500, Chris Mason wrote:
>> On 6 Mar 2020, at 9:35, Josef Bacik wrote:
> [...]
>>> 4) Planning becomes much simpler.  I've organized miniconf's at
>>> plumbers before, it is far simpler than LSFMMBPF.  You only have
>>> to worry about one thing, is this presentation useful.  I no longer
>>> have to worry about am I inviting the right people, do we have
>>> enough money to cover the space.  Is there enough space for
>>> everybody?  Etc.
>>
>> We’ve talked about working closely with KS, Plumbers and the
>> Linuxfoundation to make a big picture map of the content and
>> frequency  for these confs.
>
> And, lest anyone think we all operate in isolation, we do get together
> periodically to discuss venues, selection and combination.  The last
> big in-person meeting on this topic was at Plumbers in Vancouver in
> 2019, where we had Plumbers, KS/MS, LSF/MM and the LF conference 
> people
> all represented.

Yeah, there’s a lot of cross-over between all the PCs, so we have lots 
of chances to talk it through.

>
>>   I’m sure Angela is having a busy few weeks, but lets work with 
>> her
>> to schedule this and talk it through.  OSS is a good fit  in terms of
>> being flexible enough to fit us in, hopefully we can make  that work.
>
> And, for everyone who gave us feedback in the Plumbers surveys that 
> co-
> locating with a big conference is *not* what you want because of
> various problems like hallway track disruptions due to other 
> conference
> traffic and simply the difficulty of finding people, the current model
> under consideration is one conference organization (the LF) but two
> separate venues, sort of like OpenStack used to do for their big
> conference and design summit to minimize disruption and increase
> developer focus.
>

Agreed, but I do like the idea of doing the plenary in the bigger 
conference sessions.

-chris


^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [LSFMMBPF TOPIC] long live LFSMMBPF
  2020-03-06 19:41   ` James Bottomley
  2020-03-06 19:56     ` Chris Mason
@ 2020-03-06 20:25     ` Theodore Y. Ts'o
  1 sibling, 0 replies; 30+ messages in thread
From: Theodore Y. Ts'o @ 2020-03-06 20:25 UTC (permalink / raw)
  To: James Bottomley
  Cc: Chris Mason, Josef Bacik, lsf-pc, Linux FS Devel, linux-mm,
	linux-xfs, Btrfs BTRFS, bpf, linux-ext4, linux-block

On Fri, Mar 06, 2020 at 11:41:45AM -0800, James Bottomley wrote:
> And, for everyone who gave us feedback in the Plumbers surveys that co-
> locating with a big conference is *not* what you want because of
> various problems like hallway track disruptions due to other conference
> traffic and simply the difficulty of finding people, the current model
> under consideration is one conference organization (the LF) but two
> separate venues, sort of like OpenStack used to do for their big
> conference and design summit to minimize disruption and increase
> developer focus.

Ths is what I tried to push last year, which was to colocate LSF/MM
and KS/MS in Austin, at the same time as OSS 2020, but in a separate
hotel so we didn't have to deal with the cast of thousands which go to
OSS.  I also liked it because OSS 2020 is in June, so it would have
been from a spacing perspective it would have been an easy way to
start moving MS/KS from the second half of the year into first half of
the year.

But some folks pointed out (not without reason), that Palm Springs was
a lot more fun than Austin, and OSS still has a somewhat bad
reputation of having some really trashy talks, and so even in separate
venue, there were people who really didn't like the idea.

Because of this, when the LF (in December 2019) suggested moving the
MS/KS to Austin as part of OSS, I didn't think we would have critical
mass to overcome the reputation of talks like "#OSSummit: Seven
Properties of Highly Secure IoT." and so I told Angela, "No, we really
can't do this without something like LSF/MM to make sure we have
critical mass for a second Linux systems conference."

						- Ted


^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [LSFMMBPF TOPIC] Killing LSFMMBPF
  2020-03-06 14:35 [LSFMMBPF TOPIC] Killing LSFMMBPF Josef Bacik
                   ` (7 preceding siblings ...)
  2020-03-06 19:27 ` [LSFMMBPF TOPIC] long live LFSMMBPF Chris Mason
@ 2020-03-07  3:14 ` Steve French
  2020-03-10 13:13 ` Michal Hocko
  9 siblings, 0 replies; 30+ messages in thread
From: Steve French @ 2020-03-07  3:14 UTC (permalink / raw)
  To: Josef Bacik
  Cc: lsf-pc, Linux FS Devel, linux-mm, linux-xfs, Btrfs BTRFS, bpf,
	linux-ext4, linux-block, CIFS

Don't forget about Vault - there were some very useful hallway
discussions at Vault this year as well ... even if a bit smaller than
it should be ...

On Fri, Mar 6, 2020 at 8:36 AM Josef Bacik <josef@toxicpanda.com> wrote:
>
> Hello,
>
> This has been a topic that I've been thinking about a lot recently, mostly
> because of the giant amount of work that has been organizing LSFMMBPF.  I was
> going to wait until afterwards to bring it up, hoping that maybe it was just me
> being done with the whole process and that time would give me a different
> perspective, but recent discussions has made it clear I'm not the only one.
>
> LSFMMBPF is not useful to me personally, and not an optimal use of the
> communities time.  The things that we want to get out of LSFMMBPF are (generally)
>
> 1) Reach consensus on any multi-subsystem contentious changes that have come up
> over the past year.
>
> 2) Inform our fellow developers of new things that we are working on that we
> would like help with, or need to think about for the upcoming year.
>
> 3) "Hallway track".  We are after all a community, and I for one like spending
> time with developers that I don't get to interact with on a daily basis.
>
> 4) Provide a way to help integrate new developers into the community with face
> time.  It is far easier to work with people once you can put a face to a name,
> and this is especially valuable for new developers.
>
> These are all really good goals, and why we love the idea of LSFMMBPF.  But
> having attended these things every year for the last 13 years, it has become
> less and less of these things, at least from my perspective.  A few problems (as
> I see them) are
>
> 1) The invitation process.  We've tried many different things, and I think we
> generally do a good job here, but the fact is if I don't know somebody I'm not
> going to give them a very high rating, making it difficult to actually bring in
> new people.
>
> 2) There are so many of us.  Especially with the addition of the BPF crowd we
> are now larger than ever.  This makes problem #1 even more apparent, even if I
> weighted some of the new people higher who's slot should they take instead?  I
> have 0 problems finding 20 people in the FS community who should absolutely be
> in the room.  But now I'm trying to squeeze in 1-5 extra people.  Propagate that
> across all the tracks and now we're at an extra 20ish people.
>
> 3) Half the people I want to talk to aren't even in the room.  This may be a
> uniquely file system track problem, but most of my work is in btrfs, and I want
> to talk to my fellow btrfs developers.  But again, we're trying to invite an
> entire community, so many of them simply don't request invitations, or just
> don't get invited.
>
> 3) Sponsorships.  This is still the best way to get to all of the core
> developers, so we're getting more and more sponsors in order to buy their slots
> to get access to people.  This is working as intended, and I'm not putting down
> our awesome sponsors, but this again adds to the amount of people that are
> showing up at what is supposed to be a working conference.
>
> 4) Presentations.  90% of the conference is 1-2 people standing at the front of
> the room, talking to a room of 20-100 people, with only a few people in the
> audience who cares.  We do our best to curate the presentations so we're not
> wasting peoples time, but in the end I don't care what David Howells is doing
> with mount, I trust him to do the right thing and he really just needs to trap
> Viro in a room to work it out, he doesn't need all of us.
>
> 5) Actually planning this thing.  I have been on the PC for at least the last 5
> years, and this year I'm running the whole thing.  We specifically laid out
> plans to rotate in new blood so this sort of thing stopped happening, and this
> year we've done a good job of that.  However it is a giant amount of work for
> anybody involved, especially for the whole conference chair.  Add in something
> like COVID-19 to the mix and now I just want to burn the whole thing to the
> ground.  Planning this thing is not free, it does require work and effort.
>
> So what do I propose?  I propose we kill LSFMMBPF.
>
> Many people have suggested this elsewhere, but I think we really need to
> seriously consider it.  Most of us all go to the Linux Plumbers conference.  We
> could accomplish our main goals with Plumbers without having to deal with all of
> the above problems.
>
> 1) The invitation process.  This goes away.  The people/companies that want to
> discuss things with the rest of us can all get to plumbers the normal way.  We
> get new blood that we may miss through the invitation process because they can
> simply register for Plumbers on their own.
>
> 2) Presentations.  We can have track miniconfs where we still curate talks, but
> there could be much less of them and we could just use the time to do what
> LSFMMBPF was meant to do, put us all in a room so we can hack on things together.
>
> 3) BOFs.  Now all of the xfs/btrfs/ext4 guys can show up, because again they
> don't have to worry about some invitation process, and now real meetings can
> happen between people that really want to talk to each other face to face.
>
> 4) Planning becomes much simpler.  I've organized miniconf's at plumbers before,
> it is far simpler than LSFMMBPF.  You only have to worry about one thing, is
> this presentation useful.  I no longer have to worry about am I inviting the
> right people, do we have enough money to cover the space.  Is there enough space
> for everybody?  Etc.
>
> I think this is worth a discussion at the very least.  Maybe killing LSFMMBPF is
> too drastic, maybe there are some other ideas that would address the same
> problems.  I'd love to hear the whole communities thoughts on this, because
> after all this is supposed to be a community event, and we should all be heard.
> Thanks,
>


-- 
Thanks,

Steve


^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [LSFMMBPF TOPIC] LSFMMBPF 2020 COVID-19 status update
  2020-03-06 15:56 ` Theodore Y. Ts'o
  2020-03-06 16:08   ` Josef Bacik
  2020-03-06 18:30   ` Rik van Riel
@ 2020-03-07 18:54   ` Luis Chamberlain
  2020-03-07 19:00     ` Josef Bacik
  2020-03-07 19:12     ` James Bottomley
  2 siblings, 2 replies; 30+ messages in thread
From: Luis Chamberlain @ 2020-03-07 18:54 UTC (permalink / raw)
  To: Theodore Y. Ts'o
  Cc: Josef Bacik, lsf-pc, Linux FS Devel, linux-mm, linux-xfs,
	Btrfs BTRFS, bpf, linux-ext4, linux-block

On Fri, Mar 06, 2020 at 10:56:11AM -0500, Theodore Y. Ts'o wrote:
> Should we have LSF/MM/BPF in 2020 and COVID-19?

I'll try to take a proactive approach by doing my own digging, where
is what I have found, and few other proactive thoughts which might help:

The latest update posted on the LSFMM page from March 2 states things
are moving along as planned. After that on March 4th officials from the
county made a trip to Coachella Valley (22 minutes away from the LSFFMM
venue hotel) "to quell public fears about the spread of the novel
coronavirus", and announced that "there are no plans to cancel any of
the upcoming large events like Coachella, Stagecoach and the BNP" [0].

So, hippies are still getting together.

How about our brethren?

If we have to learn from efforts required to continue on with the in
light of the risks, we can look at what SCALE 18 is doing, taking place
right now in Pasadena [1], their page lists a list of proactive measures
required on their part to help alleviate fears and just good best
practices at this point in time.

The landscape seems positive, if we want, to move forward in Palm Springs then.

When are attendees supposed to get notifications if they are invited?

Since the nature of the conference however is unique in that it is
world-wide and invite-only it makes me wonder if the value is reduced
because of this and if we should cancel.

Does the latency involved on the confirmation of attending decrease
the value due to the current haphazard situation with COVID-19?

I am involved in other conferences and am seeing personal driven
cancelations for general concerns. For folks in the US it would be
easier / less risky to travel, so my concerns would be less than others.
But -- would we have higher personal cancelations from EU folks? What
are folks thoughts on this right now? Is anyone in the EU not coming
at all due to concerns who wouldn't mind voicing their concerns even
if LSFMM continues?

And then there is the other question: can we cancel? Or is that
economically just  too late at this point?

[0] https://kesq.com/news/2020/03/04/palm-springs-officials-to-hold-coronavirus-news-conference-on-thursday/
[1] https://www.socallinuxexpo.org/scale/18x

  Luis


^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [LSFMMBPF TOPIC] LSFMMBPF 2020 COVID-19 status update
  2020-03-07 18:54   ` [LSFMMBPF TOPIC] LSFMMBPF 2020 COVID-19 status update Luis Chamberlain
@ 2020-03-07 19:00     ` Josef Bacik
  2020-03-07 19:12     ` James Bottomley
  1 sibling, 0 replies; 30+ messages in thread
From: Josef Bacik @ 2020-03-07 19:00 UTC (permalink / raw)
  To: Luis Chamberlain, Theodore Y. Ts'o
  Cc: lsf-pc, Linux FS Devel, linux-mm, linux-xfs, Btrfs BTRFS, bpf,
	linux-ext4, linux-block

On 3/7/20 1:54 PM, Luis Chamberlain wrote:
> On Fri, Mar 06, 2020 at 10:56:11AM -0500, Theodore Y. Ts'o wrote:
>> Should we have LSF/MM/BPF in 2020 and COVID-19?
> 
> I'll try to take a proactive approach by doing my own digging, where
> is what I have found, and few other proactive thoughts which might help:
> 
> The latest update posted on the LSFMM page from March 2 states things
> are moving along as planned. After that on March 4th officials from the
> county made a trip to Coachella Valley (22 minutes away from the LSFFMM
> venue hotel) "to quell public fears about the spread of the novel
> coronavirus", and announced that "there are no plans to cancel any of
> the upcoming large events like Coachella, Stagecoach and the BNP" [0].
> 
> So, hippies are still getting together.
> 
> How about our brethren?
> 
> If we have to learn from efforts required to continue on with the in
> light of the risks, we can look at what SCALE 18 is doing, taking place
> right now in Pasadena [1], their page lists a list of proactive measures
> required on their part to help alleviate fears and just good best
> practices at this point in time.
> 
> The landscape seems positive, if we want, to move forward in Palm Springs then.
> 
> When are attendees supposed to get notifications if they are invited?
> 
> Since the nature of the conference however is unique in that it is
> world-wide and invite-only it makes me wonder if the value is reduced
> because of this and if we should cancel.
> 
> Does the latency involved on the confirmation of attending decrease
> the value due to the current haphazard situation with COVID-19?
> 
> I am involved in other conferences and am seeing personal driven
> cancelations for general concerns. For folks in the US it would be
> easier / less risky to travel, so my concerns would be less than others.
> But -- would we have higher personal cancelations from EU folks? What
> are folks thoughts on this right now? Is anyone in the EU not coming
> at all due to concerns who wouldn't mind voicing their concerns even
> if LSFMM continues?
> 
> And then there is the other question: can we cancel? Or is that
> economically just  too late at this point?

We on the PC are working very closely with the Linux Foundation to make sure we 
have many different options to choose from.  As you can see on the event website 
we already have policies in place to protect ourselves if we do decide to have 
the conference as planned.

If we decide to make changes then everybody who has been invited will be 
informed as soon as possible.  As it stands we are going ahead as planned, but 
preparing for alternatives if that becomes unfeasible.  Thanks,

Josef


^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [LSFMMBPF TOPIC] LSFMMBPF 2020 COVID-19 status update
  2020-03-07 18:54   ` [LSFMMBPF TOPIC] LSFMMBPF 2020 COVID-19 status update Luis Chamberlain
  2020-03-07 19:00     ` Josef Bacik
@ 2020-03-07 19:12     ` James Bottomley
  1 sibling, 0 replies; 30+ messages in thread
From: James Bottomley @ 2020-03-07 19:12 UTC (permalink / raw)
  To: Luis Chamberlain, Theodore Y. Ts'o
  Cc: Josef Bacik, lsf-pc, Linux FS Devel, linux-mm, linux-xfs,
	Btrfs BTRFS, bpf, linux-ext4, linux-block

On Sat, 2020-03-07 at 18:54 +0000, Luis Chamberlain wrote:
> On Fri, Mar 06, 2020 at 10:56:11AM -0500, Theodore Y. Ts'o wrote:
> > Should we have LSF/MM/BPF in 2020 and COVID-19?
[...]
> If we have to learn from efforts required to continue on with the in
> light of the risks, we can look at what SCALE 18 is doing, taking
> place right now in Pasadena [1], their page lists a list of proactive
> measures required on their part to help alleviate fears and just good
> best practices at this point in time.

I agree Scale18x is the poster child for following WHO advice to the
letter, but there are crucial differences:

   1. Scale18x has a lot of local attendees, so the conference can go
      ahead somewhat easily with local content and local attendees.  We
      have no-one for LSF/MM/BPF in Palm Springs.
   2. Scale18x did have some issues with non-local content because of
      corporate travel bans.  The whole of LSF/MM/BPF is non-local content
      and would thus be significantly disrupted.

The big problem with 2. is that a lot of corporate policies at the
moment are unconsidered blanket bans.  Even corporations who do
consider better might still be stricter than the WHO advice.  So my
company, IBM, is saying events >1000 cancel and events <1000 use your
own discretion provided they're promising to obey all the health
guidelines.  If I'd been presenting at Scale18x I'd have had to cancel,
even though under our guidelines I can still go to LSF/MM/BPF

> The landscape seems positive, if we want, to move forward in Palm
> Springs then.

For a counter example, just look at the LF Member summit which was due
to happen just after Scale18x:

https://events.linuxfoundation.org/lf-member-summit/

and that's a smaller event than Scale18x.  Remember too that the LF
runs LSF/MM so if they decide to cancel, there's not much the
organizing committee can do about it.

James



^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [LSFMMBPF TOPIC] Killing LSFMMBPF
  2020-03-06 14:35 [LSFMMBPF TOPIC] Killing LSFMMBPF Josef Bacik
                   ` (8 preceding siblings ...)
  2020-03-07  3:14 ` [LSFMMBPF TOPIC] Killing LSFMMBPF Steve French
@ 2020-03-10 13:13 ` Michal Hocko
  2020-03-10 13:40   ` Josef Bacik
  9 siblings, 1 reply; 30+ messages in thread
From: Michal Hocko @ 2020-03-10 13:13 UTC (permalink / raw)
  To: Josef Bacik
  Cc: lsf-pc, Linux FS Devel, linux-mm, linux-xfs, Btrfs BTRFS, bpf,
	linux-ext4, linux-block

On Fri 06-03-20 09:35:41, Josef Bacik wrote:
> Hello,
> 
> This has been a topic that I've been thinking about a lot recently, mostly
> because of the giant amount of work that has been organizing LSFMMBPF.

There is undoubtedly a lot of work to make a great conference. I have hard
time imagine this could be ever done without a lot of time and effort on
the organizing side. I do not believe we can simply outsource a highly
technical conference to somebody outside of the community. LF is doing a
lot of great work to help with the venue and related stuff but content
wise it is still on the community IMHO.

[...]
> These are all really good goals, and why we love the idea of LSFMMBPF.  But
> having attended these things every year for the last 13 years, it has become
> less and less of these things, at least from my perspective.  A few problems
> (as I see them) are
> 
> 1) The invitation process.  We've tried many different things, and I think
> we generally do a good job here, but the fact is if I don't know somebody
> I'm not going to give them a very high rating, making it difficult to
> actually bring in new people.

My experience from the MM track involvement last few years is slightly
different. We have always had a higher demand than seats available
for the track. We have tried really hard to bring people who could
contribute the most requested topic into the room. We have also tried to
bring new contributors in. There are always compromises to be made but
my recollection is that discussions were usually very useful and moved
topics forward. The room size played an important role in that regard.

> 2) There are so many of us.  Especially with the addition of the BPF crowd
> we are now larger than ever.  This makes problem #1 even more apparent, even
> if I weighted some of the new people higher who's slot should they take
> instead?  I have 0 problems finding 20 people in the FS community who should
> absolutely be in the room.  But now I'm trying to squeeze in 1-5 extra
> people.  Propagate that across all the tracks and now we're at an extra
> 20ish people.

Yes, BPF track made the conference larger indeed. This might be problem
for funding but it didn't really cause much more work for tracks
organization (well for MM at least).

> 3) Half the people I want to talk to aren't even in the room.  This may be a
> uniquely file system track problem, but most of my work is in btrfs, and I
> want to talk to my fellow btrfs developers.  But again, we're trying to
> invite an entire community, so many of them simply don't request
> invitations, or just don't get invited.

I do not have the same experience on the MM track. Even though the whole
community is hard to fit into the room, there tends to be a sufficient
mass to move a topic forward usually. Even if we cannot conclude many
topics there are usually many action items as an outcome.

[...]

> So what do I propose?  I propose we kill LSFMMBPF.

This would be really unfortunate. LSFMMBPF has been the most attractive
conference for me exactly because of the size and cost/benefit. I do
realize we are growing and that should be somehow reflected in the
future. I do not have good answers how to do that yet unfortunately.
Maybe we really need to split the core agenda and topics which could be
discussed/presented on other conferences. Or collocate with another
conference but I have a feeling that we could cover more since LSFMMBPF
-- 
Michal Hocko
SUSE Labs


^ permalink raw reply	[flat|nested] 30+ messages in thread

* Re: [LSFMMBPF TOPIC] Killing LSFMMBPF
  2020-03-10 13:13 ` Michal Hocko
@ 2020-03-10 13:40   ` Josef Bacik
  0 siblings, 0 replies; 30+ messages in thread
From: Josef Bacik @ 2020-03-10 13:40 UTC (permalink / raw)
  To: Michal Hocko
  Cc: lsf-pc, Linux FS Devel, linux-mm, linux-xfs, Btrfs BTRFS, bpf,
	linux-ext4, linux-block

On 3/10/20 9:13 AM, Michal Hocko wrote:
> On Fri 06-03-20 09:35:41, Josef Bacik wrote:
>> Hello,
>>
>> This has been a topic that I've been thinking about a lot recently, mostly
>> because of the giant amount of work that has been organizing LSFMMBPF.
> 
> There is undoubtedly a lot of work to make a great conference. I have hard
> time imagine this could be ever done without a lot of time and effort on
> the organizing side. I do not believe we can simply outsource a highly
> technical conference to somebody outside of the community. LF is doing a
> lot of great work to help with the venue and related stuff but content
> wise it is still on the community IMHO.
> 
> [...]
>> These are all really good goals, and why we love the idea of LSFMMBPF.  But
>> having attended these things every year for the last 13 years, it has become
>> less and less of these things, at least from my perspective.  A few problems
>> (as I see them) are
>>
>> 1) The invitation process.  We've tried many different things, and I think
>> we generally do a good job here, but the fact is if I don't know somebody
>> I'm not going to give them a very high rating, making it difficult to
>> actually bring in new people.
> 
> My experience from the MM track involvement last few years is slightly
> different. We have always had a higher demand than seats available
> for the track. We have tried really hard to bring people who could
> contribute the most requested topic into the room. We have also tried to
> bring new contributors in. There are always compromises to be made but
> my recollection is that discussions were usually very useful and moved
> topics forward. The room size played an important role in that regard.
> 
>> 2) There are so many of us.  Especially with the addition of the BPF crowd
>> we are now larger than ever.  This makes problem #1 even more apparent, even
>> if I weighted some of the new people higher who's slot should they take
>> instead?  I have 0 problems finding 20 people in the FS community who should
>> absolutely be in the room.  But now I'm trying to squeeze in 1-5 extra
>> people.  Propagate that across all the tracks and now we're at an extra
>> 20ish people.
> 
> Yes, BPF track made the conference larger indeed. This might be problem
> for funding but it didn't really cause much more work for tracks
> organization (well for MM at least).
> 
>> 3) Half the people I want to talk to aren't even in the room.  This may be a
>> uniquely file system track problem, but most of my work is in btrfs, and I
>> want to talk to my fellow btrfs developers.  But again, we're trying to
>> invite an entire community, so many of them simply don't request
>> invitations, or just don't get invited.
> 
> I do not have the same experience on the MM track. Even though the whole
> community is hard to fit into the room, there tends to be a sufficient
> mass to move a topic forward usually. Even if we cannot conclude many
> topics there are usually many action items as an outcome.
> 
> [...]
> 
>> So what do I propose?  I propose we kill LSFMMBPF.
> 
> This would be really unfortunate. LSFMMBPF has been the most attractive
> conference for me exactly because of the size and cost/benefit. I do
> realize we are growing and that should be somehow reflected in the
> future. I do not have good answers how to do that yet unfortunately.
> Maybe we really need to split the core agenda and topics which could be
> discussed/presented on other conferences. Or collocate with another
> conference but I have a feeling that we could cover more since LSFMMBPF
> 

LSFMMBPF is still by far the most useful conference I attend, so much so that 
it's basically the only thing I attend anymore.

My point is less about no longer having a conference at all, and more about 
changing what we currently have to be more useful to more people.  For MM, and I 
assume BPF, it's much different as you guys are all on the same codebase.  You 
get 25 people in the room chances are a much larger percentage of you are 
interested in each individual topic.

File systems and storage?  Way less so.  We've expanded to 3 days of conference, 
which has only exacerbated this issue for me.  Now I have a full day that I'm 
trying to fill with interesting topics that we're all interested in, and it's a 
struggle.  If instead we had everybody from the file system community there then 
I could just say "OK day 3 is BoF day, have your FS specific meetups!" and be 
done with it.  But as it stands I know XFS is missing probably 1/3 of their main 
contributors, and Btrfs is missing 1/2 to 2/3 of our developers.

In order to accomplish that we need to radically change the structure of the 
conference, hence my hyperbolic suggestion.  I think what Ted suggested is 
probably my ideal solution, we have a kernel focused spring conference where the 
whole community gets together, and then we have tracks that we carve up.

But is it a problem worth solving?  I'm not sure.  I know how I feel, but maybe 
I'm the crazy one.  I think its worth discussing.  If more people like how we 
currently do it then we can just keep trucking along.  It's not like I'll stop 
showing up, this is still a tremendously useful conference.  I just think we can 
do better.  Thanks,

Josef


^ permalink raw reply	[flat|nested] 30+ messages in thread

end of thread, other threads:[~2020-03-10 13:40 UTC | newest]

Thread overview: 30+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-03-06 14:35 [LSFMMBPF TOPIC] Killing LSFMMBPF Josef Bacik
2020-03-06 15:29 ` Jason Gunthorpe
2020-03-06 15:30 ` [Lsf-pc] " Amir Goldstein
2020-03-06 15:55 ` Josef Bacik
2020-03-06 15:56 ` Theodore Y. Ts'o
2020-03-06 16:08   ` Josef Bacik
2020-03-06 19:48     ` Theodore Y. Ts'o
2020-03-06 18:30   ` Rik van Riel
2020-03-07 18:54   ` [LSFMMBPF TOPIC] LSFMMBPF 2020 COVID-19 status update Luis Chamberlain
2020-03-07 19:00     ` Josef Bacik
2020-03-07 19:12     ` James Bottomley
2020-03-06 16:04 ` [LSFMMBPF TOPIC] Killing LSFMMBPF Nikolay Borisov
2020-03-06 16:15 ` James Bottomley
2020-03-06 16:28   ` Christian Brauner
2020-03-06 16:31     ` Josef Bacik
     [not found] ` <20200306160548.GB25710@bombadil.infradead.org>
2020-03-06 17:04   ` Al Viro
2020-03-06 17:37   ` James Bottomley
2020-03-06 18:06     ` Jason Gunthorpe
2020-03-06 19:07       ` Martin K. Petersen
2020-03-06 19:15         ` James Bottomley
2020-03-06 19:20           ` Martin K. Petersen
2020-03-06 18:23     ` Matthew Wilcox
2020-03-06 19:25       ` James Bottomley
2020-03-06 19:27 ` [LSFMMBPF TOPIC] long live LFSMMBPF Chris Mason
2020-03-06 19:41   ` James Bottomley
2020-03-06 19:56     ` Chris Mason
2020-03-06 20:25     ` Theodore Y. Ts'o
2020-03-07  3:14 ` [LSFMMBPF TOPIC] Killing LSFMMBPF Steve French
2020-03-10 13:13 ` Michal Hocko
2020-03-10 13:40   ` Josef Bacik

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).