> Next is the issue of subjects. Last year the final list came out a few
> days before the summit started, making it impossible for people who were
> not attending the summit to prepare material for those attending to
> present/include on their behalf.
I think you completely miss the point of KS here. It is not the venue for someone
presenting long material on something and "selling" it, but for free
discussion on specific topics. That is why the invitee list is so
closely controlled to make sure these productions are productive.
If there are any presentations they should be very short and merely be a
quick intro -- anything elaborate is not really welcome. That's quite different
from other conferences.
-Andi
Andi Kleen wrote:
>> Next is the issue of subjects. Last year the final list came out a few
>> days before the summit started, making it impossible for people who were
>> not attending the summit to prepare material for those attending to
>> present/include on their behalf.
>
> I think you completely miss the point of KS here. It is not the venue for someone
> presenting long material on something and "selling" it, but for free
> discussion on specific topics. That is why the invitee list is so
> closely controlled to make sure these productions are productive.
Last year the subject of DMA engines was put up, however most of the
people interested in the subject weren't even invited. In that case
there's really little concrete that can come out of the discussion.
Expecting everyone to sit down and get the full picture of a subject
in 5 minutes is unrealistic and will not lead to very useful outcome.
> If there are any presentations they should be very short and merely be a
> quick intro -- anything elaborate is not really welcome. That's quite different
> from other conferences.
Of course, thats why I suggested more like an abstract published in
advance, like why is this subject worthy and what would the person
like to achieve by having it discussed. Hopefully with some technical
ideas included.
Cheers,
Jes
> Last year the subject of DMA engines was put up, however most of the
> people interested in the subject weren't even invited. In that case
> there's really little concrete that can come out of the discussion.
Nobody claimed the committee was perfect. Shit happens.
There were also plenty of productive discussions.
> Expecting everyone to sit down and get the full picture of a subject
> in 5 minutes is unrealistic and will not lead to very useful outcome.
If they never heard of the issue before they are unlikely to
be useful in the discussion even after an hour of talk. That is not
how it works again.
> Of course, thats why I suggested more like an abstract published in
> advance, like why is this subject worthy and what would the person
> like to achieve by having it discussed. Hopefully with some technical
> ideas included.
Abstract of a discussion? Interesting concept. Maybe.
If you mean abstract of a talk then I think you're wrong.
Not sure that abstract of a discussion thing would really work though.
It seems a bit contradicting in itself.
-Andi
Andi Kleen wrote:
> Abstract of a discussion? Interesting concept. Maybe.
>
> If you mean abstract of a talk then I think you're wrong.
>
> Not sure that abstract of a discussion thing would really work though.
> It seems a bit contradicting in itself.
I was thinking more an abstract as in something that should provide a
short summary of the problem and why it should be discussed at KS.
I don't think papers etc. would do any good at this level. Something
specifically targetted at KS.
Cheers,
Jes
On Tue, Jan 30, 2007 at 06:51:51AM +0100, Jes Sorensen wrote:
> Last year the subject of DMA engines was put up, however most of the
> people interested in the subject weren't even invited. In that case
> there's really little concrete that can come out of the discussion.
Likewise IOMMUs.
I think Andrew's suggestion of adding a CFP phase to KS is excellent -
get some new blood in the room and spice up the discussion.
Cheers,
Muli
On Tue, Jan 30, 2007 at 08:43:12AM +0200, Muli Ben-Yehuda wrote:
> On Tue, Jan 30, 2007 at 06:51:51AM +0100, Jes Sorensen wrote:
>
> > Last year the subject of DMA engines was put up, however most of the
> > people interested in the subject weren't even invited. In that case
> > there's really little concrete that can come out of the discussion.
>
> Likewise IOMMUs.
There were a number of people there last year who understood IOMMUs
and could easily talk at length about them if able to do so.
iirc, you were also invited, but were unable to travel due to
bad things falling from the sky in Israel at the time ?
> I think Andrew's suggestion of adding a CFP phase to KS is excellent -
> get some new blood in the room and spice up the discussion.
I don't see anything that really precludes the idea. Or even the
notion of both that and some of the old format for some of the
sessions.
Dave
--
http://www.codemonkey.org.uk
On Tue, Jan 30, 2007 at 02:18:16AM -0500, Dave Jones wrote:
> > Likewise IOMMUs.
>
> There were a number of people there last year who understood IOMMUs
> and could easily talk at length about them if able to do so. iirc,
> you were also invited, but were unable to travel due to bad things
> falling from the sky in Israel at the time ?
That was OLS, not KS. Also, that's not the impression I got from
reading the lwn.net summary (and I know several IOMMU people such as
Olof were not invited... I think it was just jejb and ak?).
Cheers,
Muli
On Tue, 2007-01-30 at 09:29 +0200, Muli Ben-Yehuda wrote:
> On Tue, Jan 30, 2007 at 02:18:16AM -0500, Dave Jones wrote:
> > > Likewise IOMMUs.
> >
> > There were a number of people there last year who understood IOMMUs
> > and could easily talk at length about them if able to do so. iirc,
> > you were also invited, but were unable to travel due to bad things
> > falling from the sky in Israel at the time ?
>
> That was OLS, not KS. Also, that's not the impression I got from
> reading the lwn.net summary (and I know several IOMMU people such as
> Olof were not invited... I think it was just jejb and ak?).
Most of the useful IOMMU discussion happened at OLS anyway in the
various virtualisation talks. I would expect that if there's a
virtualisation mini summit going on that these would again be discussed
there.
I suppose on a side note, in spite of the fact that the virtualisation
summit has identified IO as the major discussion area, no virtualisation
people actually signed up for the Filesystem and IO summit ...
James
On Tue, Jan 30, 2007 at 07:11:34AM +0100, Jes Sorensen wrote:
> > Not sure that abstract of a discussion thing would really work though.
> > It seems a bit contradicting in itself.
>
> I was thinking more an abstract as in something that should provide a
> short summary of the problem and why it should be discussed at KS.
> I don't think papers etc. would do any good at this level. Something
> specifically targetted at KS.
If you really want to get people up to speed on a topic, then you need
a bibliography - which in practice may just be a few links to articles on
the h/w involved, research papers on the s/w techniques, or just to
previous mailing list discussions (especially if these were on a
special topic list rather then on LKML). That would give people whose
interest is piqued by the abstract a starting point to learn a bit
more before the summit so there can be an informed discussion.
-Tony