[comp.org.usenix] W91 USENIX in retrospect

brian@ucsd.Edu (Brian Kantor) (01/28/91)

Not a bad conference, with several interesting papers and a couple of
lively panel discussions, as well as the usual highly-informative
hallway discussions and BOFs, but important things first: the
hospitality suite ratings.  I only went to the suites in the Kempinski
(home of the concrete-slab mattress) hotel.  Of those, I remember
these:

1. IBM research.  Fantastic; the munchies were the same as all the other
suites, but there were lots more of them, and they were always hot.  A
good selection of beers, but it wasn't until the third day that someone
finally realized that Corona is supposed to be served with LIME, not
lemon.  Unfortunately, there were no dark beers.  The toys were nice and
varied nightly.  Good machines with interesting demos; technical people
were always there and seemed to know what they were talking about, and
there were no suits to jump into your wallet.  Fantastic desserts after
the Usenix conference reception (which didn't have any) and a good place
to get free beers instead of buying them at the reception.  The dessert
inspection tool (pen flashlight) was quite handy to avoid dairy-based
confections.  It's amazing what you can put through a luggage-tag
laminating machine.

2. Pencom.  Nice ice cream and beers, build your own sundaes.  I guess they
aren't really selling technical things so there weren't any questions of
that kind to ask, but at least they didn't try to snuggle up into your
pockets as soon as you walked in.  The screwdriver with the auto-retracting
phillips end is still their mainstay trinket, and they had a few
laughing golf-balls.  This time I didn't have the urge to count my
fingers after shaking hands with the reps; they're learning to mellow
out a bit.  Maybe they'd be ok to work for after all.

3. O'Reilly and Associates.  Had a book signing for the new awk and perl
books; crowded as hell but good munchies and beer.  You could buy books
from them there; I heard they weren't selling on the Uniforum showfloor.

4. Cray research.  Minor munchies; ok if you're into fruit, grease and
salt.  Beers were ok, they ran out of "desk-top crays" [apparently a
ceramic coaster] early on, but had lots of nifty balsa airplanes that
were superb for bombing sorties against the sales-droids at the
infomart.  Again, not much in the way of technical people to ask
questions of, but it seemed like a nice place to apply for a job, which
might well have been the reason why they had the suite.  Whoever
changed the sign outside their suite to read "Control Data" wins the
golden cudgel award for subtlety.

I regret to say that the trade show was pretty much of a zero for me
this time - didn't see anything of great interest.  I heard it called
"uniborum" several times.  There didn't seem to be any really
revolutionary new things there, but there were a lot of "me too"
products being demonstrated by booth-boyz and booth-bimbos who really
didn't know much about the product or company.  The sales-droids all
seemed to emit an aura of desperation; times must be harder than I
thought.  Whoever it was with the rap-singer booth-show seemed to be
scaring more people away than attracting them.  Besides the sales
literature on some interesting products and a few neat demos, the major
trinkets were good chocolate from Amdahl and some nifty stash-bottles
from CITA, perfect for hanging your carkeys in when you're swimming or
jogging.  There were some more substantial toys if you went to a bunch
of booths and got stamps in some order - then you got a nerf football
or the like.  Someone was giving away a Miata; I wonder who got it?

Summary: technoids 3, salesdroids 0.  The technical conference was worth
it and the trade show wasn't.  Your mileage may vary.  Especially if you
won the Miata.
	- Brian

david@indetech.com (David Kuder x2003) (01/28/91)

In article <26879@ucsd.Edu> brian@ucsd.Edu (Brian Kantor) writes:
>or the like.  Someone was giving away a Miata; I wonder who got it?

The someone was Uniplex trying to entice the masses to buy an
integrated office thingy suffering from all the proper TLAs.
The winner was NCR's marketing director who spent the whole show
just across the aisle in the NCR booth.

P.S.  I resent being called "booth boyz".  I'd wear a t-shirt
and jeans over the suit anyday. 
-- 
David A. Kuder              Looking for enough time to get past patchlevel 1
415 438-2003  david@indetech.com  {uunet,sun,sharkey,pacbell}!indetech!david

de5@ornl.gov (Dave Sill) (01/29/91)

In article <26879@ucsd.Edu>, brian@ucsd.Edu (Brian Kantor) writes:
>Not a bad conference, with several interesting papers and a couple of
>lively panel discussions, as well as the usual highly-informative
>hallway discussions and BOFs, but important things first: the
>hospitality suite ratings.

So how does one find out about hospitality suites and get invited to
them? 

--
Dave Sill (de5@ornl.gov)	  It will be a great day when our schools have
Martin Marietta Energy Systems    all the money they need and the Air Force
Workstation Support               has to hold a bake sale to buy a new bomber.

amanda@visix.com (Amanda Walker) (01/29/91)

In article <1991Jan28.005533.2951@iwarp.intel.com>, merlyn@iwarp.intel.com
(Randal L. Schwartz) writes:
> My take on Uniforum: I've never heard so many buzzwords in one room.
> Wow.  (Doesn't anyone sell anything that isn't a "transparently
> multiscalable object-oriented SysV-compatible architecture" any more? :-)

Hey, you forgot "Open," which must have been an official prerequisite for
getting booth space.  I epsecially liked the vendors that claimed exclusives
on whatever type of "open system" they were selling.

Kind of missed the point, I'd say...

--
Amanda Walker
Visix Software Inc.

jeffs@soul.esd.sgi.com (Jeff Smith) (01/29/91)

In article <26879@ucsd.Edu>, brian@ucsd.Edu (Brian Kantor) writes:
|> 1. IBM research.  Fantastic; ....

It was actually commercial IBM (IBM Austin, home of the RS/6000).
In the past (like W90) Reasearch had done it.

Personally I didn't make it, but I did manage to get most of
the goodies 2nd hand :-).

jeffs

shaver@convex.com (Dave Shaver) (01/29/91)

Randal L. Schwartz writes:
>My take on Uniforum: I've never heard so many buzzwords in one room.

My favorite buzz-o-rama was the IBM booth where their "GUI allows for
customized ICONS."  WOW!  You mean I get to *pick* my OWN icons!  :-)

/\  Dave Shaver
\\  CONVEX Computer Corporation, Richardson, TX
\/  Internet: shaver@convex.com    UUCP:  uunet!convex!shaver

steve@ibmpa.awdpa.ibm.com (Steve DeJarnett) (01/29/91)

In article <26879@ucsd.Edu> brian@ucsd.Edu (Brian Kantor) writes:
>1. IBM research.  Fantastic;

	Well, thanks.  Glad you enjoyed it.  We enjoyed putting it on (even if
we are just catching up on our sleep again).  The only thing to note is that 
this one was by IBM AWD (the RS/6000 people).  Last winter's was IBM Research.  
Now, the suite did seem to attract a lot of the research folk, but, then, it 
seemed to attract a lot of people, in general.

>Unfortunately, there were no dark beers.

	So noted.  We only specified "imported beers only" to the hotel, which
they took to mean the Heineken, Corona, and the other one (whose name escapes
me at the moment).  I know we've already mentioned that to the suite organizer,
but we'll make sure we're more explicit when we specify "imported beers" next
time.

	Steve

Steve DeJarnett			Internet: steve@ibmpa.awdpa.ibm.com
IBM AWD Palo Alto		UUCP:	  uunet!ibmpa.awdpa.ibm.com!steve
(415) 855-3510			IBM VNET: dejarnet at ausvmq
These opinions are my own.  I doubt IBM wants them.......

jjhnsn@slcs.slb.com (James Lee Johnson) (01/29/91)

In article <1991Jan28.214313.29284@ibmpa.awdpa.ibm.com> steve@ibmpa.awdpa.ibm.com (Steve DeJarnett) writes:
>In article <26879@ucsd.Edu> brian@ucsd.Edu (Brian Kantor) writes:
>>1. IBM research.  Fantastic;
> ...
>>Unfortunately, there were no dark beers.
>
>	So noted.  We only specified "imported beers only" to the hotel, which
>they took to mean the Heineken, Corona, and the other one (whose name escapes
>me at the moment).  I know we've already mentioned that to the suite organizer,
>but we'll make sure we're more explicit when we specify "imported beers" next
>time.

I suspect people would have enjoyed trying Shiner Bock, a good Texas beer.
Unfortunately, I'm not sure you can get Shiner beer in Dallas :-)

    jj
--
"I like black boxes; I hate black magic."
James Lee Johnson   Schlumberger Lab for Computer Science    Austin, Texas
Internet:  jjhnsn@slcs.slb.com                       SINet:  slcs::johnson
UUCP:  cs.utexas.edu!slcs.slb.com!jjhnsn

scs@lokkur.dexter.mi.us (Steve Simmons) (01/29/91)

steve@ibmpa.awdpa.ibm.com (Steve DeJarnett) writes:

>In article <26879@ucsd.Edu> brian@ucsd.Edu (Brian Kantor) writes:

>>Unfortunately, there were no dark beers.

>	So noted.  We only specified "imported beers only" to the hotel, which
>they took to mean the Heineken, Corona, and the other one (whose name escapes
>me at the moment).

It was `BitBerger'.  Who could ask for a more appropriate beer for USENIX?
-- 
"SO be it!  The fate of the UNIVERSE is in your hands!"
"Talk about job-related stress."

henry@garp.mit.edu (Henry Mensch) (01/30/91)

steve@ibmpa.awdpa.ibm.com (Steve DeJarnett) writes:
|> >Unfortunately, there were no dark beers.
|> 
|> 	So noted.  We only specified "imported beers only" to the hotel ...

the whole thing is the hotel's fault ... for an outfit owned by lufthansa
german airlines, it is a crime that they only had one beer you might want to
drink (this was also the case at the bar and in the restaurants).

-- 
# Henry Mensch    /   <henry@garp.mit.edu>   /   E40-379 MIT,  Cambridge, MA
# <hmensch@uk.ac.nsfnet-relay> / <henry@tts.lth.se> / <mensch@munnari.oz.au>
#     via X.400: S=mensch; OU=informatik; P=tu-muenchen; A=dbp; C=de

rob@hanalei.berkeley.edu (Rob Robertson) (01/31/91)

In article <5051@media-lab.MEDIA.MIT.EDU> henry@garp.mit.edu (Henry Mensch) writes:

   the whole thing is the hotel's fault ... for an outfit owned by lufthansa
   german airlines, it is a crime that they only had one beer you might want to
   drink (this was also the case at the bar and in the restaurants).

hey man, what do you expect from a fly by night outfit...

			       * goan *
rob

pat@grebyn.com (Pat Bahn) (01/31/91)

In article <1991Jan28.171959.9679@cs.utk.edu> de5@ornl.gov writes:
>
>So how does one find out about hospitality suites and get invited to
>them? 


If you have to ask, YOU ARE ABSOLUTELY NOT ALLOWED TO GO!!!!!!!!

:-)

BTW
Spider systems at the 89 interop had a invite only hospitality suite
with
 a sales droid barring the way.  So yours truly and two guys from martin
BS'd him into letting us in.  I told him we were here to survey products
for the SDINET and that absolutely spider wouldn't be on it.

Worked like a champ.  Unfortunately I had to spend all the rest of the
night getting their president of my case.  He wanted to talk saales and
I wanted to drink........

-- 
=============================================================================
Pat @ grebyn.com  | There is a fine line between art and insanity. 
301-948-8142      |  					Pat Bahn
=============================================================================

dyker@cs.colorado.edu (Barbara Dyker) (02/01/91)

>| 3. O'Reilly and Associates.  Had a book signing for the new awk and perl
>| books; crowded as hell but good munchies and beer.  You could buy books
>| from them there; I heard they weren't selling on the Uniforum showfloor.

> No selling allowed in Uniforum.

Huh?  I know Prentice Hall sells books at InterOp and I saw them at
Uniforum assuming they would be happy to sell me a book (and bill me
for 1 cent rounding error later as they did before).

I can understand a rule not allowing selling of computer products,
but books?  Noone takes money, but you can't say there was no "selling".


...Barb
dyker@cs.colorado.edu 

merlyn@iwarp.intel.com (Randal L. Schwartz) (02/01/91)

In article <1991Jan31.220216.5438@csn.org>, dyker@cs (Barbara Dyker) writes:
| Huh?  I know Prentice Hall sells books at InterOp and I saw them at
| Uniforum assuming they would be happy to sell me a book (and bill me
| for 1 cent rounding error later as they did before).
| 
| I can understand a rule not allowing selling of computer products,
| but books?  Noone takes money, but you can't say there was no "selling".

Just passin' on what they told me.  *I'd* have been perfectly pleased
if they could have sold some Perl books at the O'R&A Uniforum booth,
but I was told "no selling".  In fact, they were sending people over
to the little bookstore just down the hallway if they expressed
interest, or taking orders via an order sheet.  But no money exchanged
hands.

Maybe P-H thinks they're big enough that they can get away with it?

Just another Perl book hacker,
-- 
/=Randal L. Schwartz, Stonehenge Consulting Services (503)777-0095 ==========\
| on contract to Intel's iWarp project, Beaverton, Oregon, USA, Sol III      |
| merlyn@iwarp.intel.com ...!any-MX-mailer-like-uunet!iwarp.intel.com!merlyn |
\=Cute Quote: "Intel: putting the 'backward' in 'backward compatible'..."====/

njs@scifi.uucp (Nicholas J. Simicich;?) (02/01/91)

For IBM Research, we specified beer lists in advance, and went out to
get the ones that the hotel couldn't get for us after negotiating
corkage.  We went for particular brands.

I was sorry to miss this Usenix.

-- 
Nick Simicich - uunet!bywater!scifi!njs - njs@ibm.com
SSI #AOWI 3958, HSA 318

brad@looking.on.ca (Brad Templeton) (02/04/91)

Well, for unknown reasons the Marriott was willing to let me cater my
own suite during USENIX.   Thus I kept everybody happy for a pittance
compared to hotel catering charges with a keg and bottles of different
beers, a bathtub of ice and various munchies.

The problem was the 9 minute walk to the alternate hotel, of course.
It occurs to me that it might be worth encouraging USENIX to look for
hotels that are more amenable to this sort of thing.   Almost *any*
hotel is amenable if it makes the difference in getting a convention.

(Admittedly some hotels have sworn off SF conventions, which are major
consumers of non-catered suite space.)

Based on what IBM pays to host their suite, it could offer USENIX a
multi-thousand dollar bribe to pick a hotel that allows self-catered
suites, and still save money!

Usenix folk are interested in neat new technology, nice toys, and
interesting and tasty things to eat and drink.   They have little
concern for fancy service, bartenders in tuxedos and armies of waiters.
-- 
Brad Templeton, ClariNet Communications Corp. -- Waterloo, Ontario 519/884-7473

libes@cme.nist.gov (Don Libes) (02/05/91)

In article <26879@ucsd.Edu> brian@ucsd.Edu (Brian Kantor) writes:
>Not a bad conference, with several interesting papers and a couple of
>lively panel discussions, as well as the usual highly-informative
>hallway discussions and BOFs, but important things first: the
>hospitality suite ratings.  

Now that we've seen 20 messages describing the window contest, book
selling, lack of dark beer, and hospitality suite analysis comparable
to the best of CNN Gulf War reporting, I'd be interested to read about
the technical aspects.

Would anyone care to summarize some of the best talks and papers?
Bear in mind, most readers don't have the proceedings yet.  Someone
already asked if they are on-line, but I saw no answer.  Perhaps
individuals could post where their papers are.

Perhaps we could start pushing the technical aspects of these
conferences a little harder.  How about naming a "Best Paper" in each
conference?  Maybe in 3 months (to give everyone time to receive and
read the proceedings) we can survey for this honor?

Don Libes          libes@cme.nist.gov      ...!uunet!cme-durer!libes

ed@mtxinu.COM (Ed Gould) (02/05/91)

Brad Templeton:
>Well, for unknown reasons the Marriott was willing to let me cater my
>own suite during USENIX. ...

>Usenix folk are interested in neat new technology, nice toys, and
>interesting and tasty things to eat and drink.   They have little
>concern for fancy service, bartenders in tuxedos and armies of waiters.

This is all true, but just doesn't make it onto the list of
important criteria for choosing a site for a USENIX Conference.

The primary considerations are space availability and cost.  The
multi-thousand dollar "bribe" that Brad suggested IBM might be
willing to pay to self-cater their suite just doesn't go very far
towards defraying the $10,000 - $25,000 it takes to rent a meting
room large enough to hold our sessions for a week.  Or the further
expense for rooms for tutorials, BOFs, the terminal room, parallel
sessions, etc.  If the hotel has sufficient facilities on site,
then the meeting rooms are typically included in the package:  If we
book enough sleeping rooms, the rest of the space comes for free.
If the hotel doesn't have the space, then we have to rent it elsewhere.

On the whole, Judy DesHarnais, our Conference Coordinator, does a
wonderful job of balancing our needs and desires against the not
always sympathetic positions of the people who manage the facilities.

(speaking for myself)

-- 
Ed Gould                    mt Xinu, 2560 Ninth St., Berkeley, CA  94710  USA
ed@mtxinu.COM		    +1 415 644 0146

"I'll fight them as a woman, not a lady.  I'll fight them as an engineer."

terri_watson@cis.ohio-state.edu (02/05/91)

Well, I won't really summarize any papers, but I might encourage some
discussion on the kernel panel session.  I found it interesting and
slightly entertaining.

The main theme was micro vs. monolithic kernel design <or the lack
thereof :>.

what should be in the kernel?  is it better to have "everything you
ever wanted + the kitchen sink" in there to minimize context switching
etc, or would you rather move it into user space for "clean structure"
and "modularity" but pay a penalty for messages etc, or what about
separate components that can be tested in user space, but moved into
kernel space for increased efficiency?

obviously, not all the pros & cons are listed above, but this is a
sample of the talk(/discussion/argument/<meta-del> -- argument?  did
_I_ say that?).

other phrases included importance of "glue" between the boxes -- how
do you connect the components together?  

large kernels:
are "large" kernels inherently bad?  or is it just that they weren't
designed properly?  could they be designed properly?  can they be made
to work on new architectures with parallelism & distributed environs
effectively?

micro kernels:
do you have to pay a severe penalty for message passing?  are they
easier to debug?  more flexible?  one argument is that such kernels
are more effective in providing mechanism, not policy

how many primitives should be in the kernel?  

One point that was made near the end of the discussion was that some
of the conflict seemed to arise from the style of computing that the
user preferred, single machine/board based, or more distributed.
Should a user want to share "his" machines resources with others?
(Personally, I would say emphatically YES in the general case, but
apparently some there were not of the same opinion.  For those of us
with no money to buy many machines of our own, the thought of stealing
idle cycles from hundreds of other machines is immensely appealing.
<grin>)

Terri Watson
The Ohio State University
elf@cis.ohio-state.edu

henry@zoo.toronto.edu (Henry Spencer) (02/05/91)

In article <ELF.91Feb4151419@giza.cis.ohio-state.edu> terri_watson@cis.ohio-state.edu writes:
>micro kernels:
>do you have to pay a severe penalty for message passing?  are they
>easier to debug?  more flexible?  one argument is that such kernels
>are more effective in providing mechanism, not policy

And is it really appropriate to call 60,000 lines of source a "micro" kernel?
(Mach.)
-- 
"Maybe we should tell the truth?"      | Henry Spencer at U of Toronto Zoology
"Surely we aren't that desperate yet." |  henry@zoo.toronto.edu   utzoo!henry

brad@looking.on.ca (Brad Templeton) (02/05/91)

Oh, I agree, clearly the conference's needs for space and facilities are
paramount.

What I really wanted to say was that hotels can be far more flexible than
they pretend to be, and that USENIX might consider that.   USENIX is the
technical meeting place for the unix community, but it is/has become a social
activity as well.
-- 
Brad Templeton, ClariNet Communications Corp. -- Waterloo, Ontario 519/884-7473

pst@jessica.Stanford.EDU (Paul Traina) (02/05/91)

In article <1991Feb4.223144.14566@zoo.toronto.edu> henry@zoo.toronto.edu (Henry Spencer) writes:


   And is it really appropriate to call 60,000 lines of source a "micro" kernel?
   (Mach.)

You're right Henry,  UNIX kernels are too slow and bloated.

I could have done the whole thing in...oh, I dunno, maybe...AWK!?!

...yeah...AWK, that's it, that's the ticket!

		/vmunix_awk

						Paul

p.s. :-)

sonya@canstar.UUCP (Sonya D. Neufer) (02/06/91)

In article <1991Feb05.075920.7712@looking.on.ca> brad@looking.on.ca (Brad Templeton) writes:
>What I really wanted to say was that hotels can be far more flexible than
>they pretend to be, and that USENIX might consider that.

What specific ways might USENIX better consider how hotels are willing
to be more flexible??  Don't forget to include the unseen factors in your
suggestion.  For Example, I have heard the suggestion that the USENIX
"get-together" should be held outside the conference hotel.  If this is
done, then the hotel sees their income decrease and they will charge more
for other items, such as meeting room space.  As well, getting over 1,100
people to another location is not my idea of scheduling fun, much less cost
reduction.

I am sure USENIX would give full consideration to well thought out specific
suggestions.

Sonya

rjg@umnstat.uucp (Robert J. Granvin) (02/06/91)

>Usenix folk are interested in neat new technology, nice toys, and
>interesting and tasty things to eat and drink.   They have little
>concern for fancy service, bartenders in tuxedos and armies of waiters.

Just as a personal comment, for anyone who is taking score, I like neat
new technology, but I also appreciate good service at good facilities,
and am willing to pay for it.  If I'm passing through a city on my way
to somewhere else, and just a place to crash, a Motel 6 ot a cardboard
box will do just fine.  If I'm going to exist someplace for a while, I'm 
happy to pay extra (although sometimes grudgingly :-) for better quality 
and superior service.  But that's just me (comfort alert! :-).

Robert J. Granvin                           E/Mail: rjg@umnstat.stat.umn.edu
User Services Specialist                    AT&T:   +1 612 625 9224
School of Statistics
University of Minnesota