USENIX Conference Format
Gregory M. Paris
and by-and-large enjoyed it. In particular, I really liked the city
itself and the Marriott hotel staff were outstandingly friendly and
helpful. (Of course, I've had a great time every time I've visited
Texas; its people are uncommonly upbeat and friendly.)
Now to the raising of knife to the sacred cow....
Is the twice-yearly USENIX conference gradually becoming irrelevant?
Irrelevant in the sense that it hasn't kept pace with the changes in
the "UNIX world" (if there is such a thing).
Before you hit the flame key, please let me try to expand upon my
thinking.
First, let me strike from discussion the conference tutorial sessions.
To me it seems that there's still much that can be learned at the
tutorials and, my guess is, there will always be room to learn new
things from tutorials as long as UNIX-related development continues.
Similarly, the BoFs (the well-attended ones anyway) are, by their
very nature, relevant. Let's omit them too.
No, my major concern is with the two-track technical sessions and
invited talks. Just watching the other attendees during these sessions
(some who were incredibly rude and talked throughout the presentations,
paying no heed whatsoever to the presenters), I've come to the
conclusion that most presentations and talks were relevant to only a
minority of those in the room (and of course, at any one time, many
attendees would be at the other track or out in the foyer, making the
actual percentage of interested attendees even lower). I don't fault
the presenters or their materials for this low interest level, rather,
I think the following reasons may hold:
1. The depth and breadth of the "UNIX world" is now such that
the field is too large for any single paper to apply to the
work of more than a small percentage of attendees. Indeed,
many of the presentations probably seemed arcane to most
attendees. (I think this fits in with Stuart Feldman's
system coherence curve.)
2. Many people don't have access to UNIX source, so neat hacks,
ala USENIX's early years, don't carry much interest. I'll
go so far as to say that USENIX is a hacker's organization
but many of its members have little chance to hack.
3. Much of the "interesting" work is being done in UNIX-related
areas, for example, X, that just plain aren't UNIX. Relatedly,
system admins, such as myself, often are more concerned with
interoperation of the many commercial applications they're
required to support than they are with UNIX evolution.
4. UNIX, in its many flavors and forms, is a commercial product!
That means that more and more of the attendees are working in
the Real World and have no use for any presentations other
than those that will help solve problems *now*. Although many
interesting presentations were made, too many of them were of
merely academic interest. Conversely, the paper I found most
pragmatic, "Mainframe Services from Gigabit-Networked
Workstations," drew almost hostile questioning from some who
I perceived as being "academic-minded" members of the audience.
(Disclaimer: I have no association with the writers of this
paper; please don't associate them with my views.)
5. Software bloat, though controversial, I think is a real issue
with many attendees. Are people as interested today in the
latest new whizbang piece of software as they were, say, ten
years ago? Doesn't UNIX already have a kitchen sink?
I could go on, but I'll leave further expansion to any of you who might
agree with me. In the meanwhile, I'll be rummaging up that asbestos
suit....
--
Greg Paris <pa...@merlin.dev.cdx.mot.com>
Motorola Codex, 20 Cabot Blvd C1-30, Mansfield, MA 02048-1193
Brad Templeton
presented to the broad audience.
There is, however, much that goes on at USENIX that is valuable.
BOFs, Tutorials, good keynotes and certain good sessions, and most of all
the networking and meeting in person with people in the field.
But how do you get those good things without a focus? Corporations are
not going to pay to send their people to a technical party, are they?
With that in mind I will point out that you don't want a conference where
every paper is important to you. (Or where there are multi-tracks and
so you must be in one session or another at all times.) You want to spend
some time outside of sessions in the halls and bar where the real conference
goes on.
If I were to design an "ideal" conference I might consider the following
points:
a) It should be at an airport hub city, so people can get to it on
nonstop and direct flights.
b) In particular, the large groups from the SF Bay and Bosnywash
should have easy access so we can get large numbers of attendees.
c) It should run Thursday-Saturday, or even to Sunday morning, so
that people can get there on cheap fares. (And indeed, so that
cheap hotel rates can be had.)
d) In-the-halls networking should be considered as important as
presentations.
e) Presenters should be good speakers. Yes, it goes against the grain
of technical people to discriminate on anything but technical
excellence, but if it's a presentation, you're going to *see* it.
You want a clear and captivating summary of what is new, with
pointers into the paper for the gritty details. Attending a
talk in person should always give me the feeling that I got
a lot more than I would have gotten from watching a video of the
talk or reading the paper. Otherwise, why did I go thousands of
miles?
In addition, papers should be aimed at a general, but highly
Unix-educated audience. The number of papers aimed at very small
segments of the audience should be kept to a minimum.
f) BOFs, in the daytime
g) Real BOFs that are not just unannounced talks with a little more
Q&A after. Working sessions, not dissemination.
h) Nice facilities, of course, but no need to go overboard. $3
Danishes and Cokes are not important. Good rooms and A/V are.
Nice big bar(s) with non-smoking areas would be nice.
None of this is particular criticism for the most recent conference.
S.A. was nice, the hotel was one of the best, and aside from it not being
an airport hub city everything was pretty good.
--
Brad Templeton, ClariNet Communications Corp. -- Sunnyvale, CA 408/296-0366
Steve McDowell
In message <1992Jun16....@merlin.dev.cdx.mot.com> pa...@zygon.dev.cdx.mot.com
> The depth and breadth of the "UNIX world" is now such that
> the field is too large for any single paper to apply to the
> work of more than a small percentage of attendees. Indeed,
> many of the presentations probably seemed arcane to most
> attendees.
But the USENIX conferences are still the *only* place that one can go and
get that variety. Focused workshops are fine, but Conferences are good to
see what's going on in many different areas. That, to me, is what makes the
thing so attractive. If you don't want the variety, then go to the
workshops.
> Many people don't have access to UNIX source, so neat hacks,
> ala USENIX's early years, don't carry much interest. I'll
> go so far as to say that USENIX is a hacker's organization
> but many of its members have little chance to hack.
What I get out of these "neat hacks" is not so much how can I make my
UNIX do what their's is doing, but more "whoa, that's a great way to
do that type of thing". It adds to my knowledge base, and next time
I need to do something similar (even at the applications level) I will
remember back to that neat hack that I saw at USENIX.
> Much of the "interesting" work is being done in UNIX-related
> areas, for example, X, that just plain aren't UNIX. Relatedly,
> system admins, such as myself, often are more concerned with
> interoperation of the many commercial applications they're
> required to support than they are with UNIX evolution.
USENIX has always been very technically oriented, and often caters to the
fringe-UNIX crowd. . I'm not sure that I want them to focus too much on the
"using" aspect; I *like* the "doing" mentality that USENIX promotes. What
you're asking for is already being done at Uniforum (at least that's what I
get from their brouchures).
> Software bloat, though controversial, I think is a real issue
> with many attendees. Are people as interested today in the
> latest new whizbang piece of software as they were, say, ten
> years ago? Doesn't UNIX already have a kitchen sink?
Yes, UNIX has a kitchen sink. But UNIX can also be much better. The focus
of these conferences has always been to talk about making it better; how
to make it more efficient; how to make the world rotate a little faster.
Your points are well taken, and there may be room for some change. I think
that's what the board had in mind when they invented the alternate technical
tracks.
But let me tell you, a USENIX conference is the absolute best place to go
and satisfy your hunger for a variety of cravings. It tells me what's happening
and how it's happening and it lets me meet people that I might not meet
otherwise. I wouldn't get that out of, say, the micro-kernel workshop or the
LISA conference. I want everything. And, I want it in one place. USENIX has
kindly given me that for less money than any other conference I've ever
attended. Kudos.
--
Steve McDowell . . . . o o o o o Opinions are
Exlog, Inc. _____ o mine, not my
mcdo...@exlog.com _____==== ]OO|_n_n__][. employers..
[_________]_|__|________)<
Steve Simmons
>Is the twice-yearly USENIX conference gradually becoming irrelevant?
There were a group of doing an informal brainstorming session by the
bulletin board who came up with pretty much the same issues. We
decided to post the discussion here and see what came of it. Greg both
beat us to the punch and did a good job in his post; there's a few more
at the end.
After the discussion and decision to post, I was informed that the
topic has apparently come up several times in the past and may even now
be under informal discussion by the board. So if folks have opinions,
this is probably a good time to express them.
Aside from issues relating to the quality of the conference, there are
two points which we felt were contributing to the loss of attendance at
San Antonio and Nashville.
The first is the "choose one" problem. If your employer or pocketbook
can afford only one USENIX per year, which one will you go to?
Obviously the one in the better vacation spot. San Francisco in
January is usually more attractive to travelers than Nashville in
June.
This applies equally well to paper submissions. If you can afford only
one conference, you will submit to the conference at the more
attractive site.
The second is the "choose two" problem. Even if your employer can
afford to send you out twice, you might well choose one of the
specialty conferences. LISA is now drawing over 500 people, and it's
quite possible that those 500 are *not* going to summer USENIX.
Note that the second of these is somewhat independent of the
recession. "Choose two" was simply not an issue when there were only
two.
--
Simmons' Law Of Diskless Workstations: "For every dollar you spend on
a diskless workstation, expect to spend a dollar on a file server."
Corrollary: "If you spend $.75 instead of $1.00, expect to spend another
$.75 fixing it when it doesn't perform to spec."
Albert Cheng
In article <1992Jun16.0...@clarinet.com>, br...@clarinet.com (Brad Templeton) writes:
>...
>If I were to design an "ideal" conference I might consider the following
>points:
>
> a) It should be at an airport hub city, so people can get to it on
> nonstop and direct flights.
I don't know how important this is. For people like me who work in the
middle of no where, every destination takes a connected flight.
> c) It should run Thursday-Saturday, or even to Sunday morning, so
> that people can get there on cheap fares. (And indeed, so that
> cheap hotel rates can be had.)
Please, no. It is bad enough to miss the family for several days. If
the weekend is gone too, my kids would be attending USENIX before I
know it. And some people like me would like to attend church--it helps
to keep me saint, oopps, I mean sane. :-)
Rich Salz
> If your employer or pocketbook
>can afford only one USENIX per year, which one will you go to?
>Obviously the one in the better vacation spot.
In the past it was very easy to decide which one to go to: June.
Judging from the open board meetings I've attended, people are concerned
that the choice is no longer as clear.
/r$
William LeFebvre
|> Is the twice-yearly USENIX conference gradually becoming irrelevant?
|> First, let me strike from discussion the conference tutorial sessions.
...
|> Similarly, the BoFs (the well-attended ones anyway) are, by their
|> very nature, relevant. Let's omit them too.
Ummm....that's half the conference right there. Maybe you should also
omit the keynote presentation from your criticism since you did seem to
get something valuable out of it. Also, you should not group the "tool
off" in with the sessions you are criticizing (since it really wasn't a
presentation of a paper). And the WIPs are just like BOFs, so those
don't count either. Oh yeah, the vendor floor isn't a presentation,
either, so it is immune to your "overall criticisms" either. What's left?
When you get right down to it, your ONLY objection is to the technical
paper presentations and the invited talks. That's a small percentage
of the conference. If you think that that's all there is to the
conference, then you should either (1) stop going or (2) look closer
at all the other activities (including just sitting around the lobby
and striking up conversations with folks).
I only attended about half the paper sessions (and even at that I usually
went for the invited talks). But it was still worth going. Why? Because
I meet people. I talk with others about problems and solutions. I strike
up relationships. I make contacts. That's half the reason for going.
If you want "right now solutions" for management folks, then you are better
off at Uniforum.
|> Just watching the other attendees during these sessions
|> (some who were incredibly rude and talked throughout the presentations,
|> paying no heed whatsoever to the presenters)
Yes that is rude. SESSION ATTENDEES SHOULD BE QUIET! If you must carry
on a conversation, then step out in the hallway.
|> I've come to the
|> conclusion that most presentations and talks were relevant to only a
|> minority of those in the room
That may be true, but that minority kept changing. It would be a
different crowd for each session (perhaps even each paper). But ONE
person attending just ONE presentation is valulable to that person.
I doubt that it was the same small group attending all sessions. This
is called "diversity" and is necessary for the same reasons you cite:
not everyone has the same interests or the same needs.
|> (and of course, at any one time, many
|> attendees would be at the other track or out in the foyer, making the
|> actual percentage of interested attendees even lower)
As I said earlier, interaction with other attendees is half the reason
for going to the conference. No this is not "partying", this is
interaction on a professional level. If you want a term that you
computer-illiterate manager understands, then call it "networking"
(although I personally hate using that term in that context, that is
what it means).
Now that you have stated what you perceive to be a problem, do you
have any solutions to propose?
William LeFebvre
Computing Facilities Manager and Analyst
Department of Electrical Engineering and Computer Science
Northwestern University
<ph...@eecs.nwu.edu>
W. Richard Stevens
In the two books I've written, I'm amazed at how many Usenix Conference
papers I actually reference for state of the art ideas. The papers are
relevant and good (at least to me).
I've never submitted a paper to a Usenix conference, as I have the
feeling that if I put all that work into a paper I'd rather submit
it to "Computing Systems" and get a refereed publication. Conference
proceedings used to have the advantage that the turnaround was much
faster than refereed publications, but Computing Systems is just as
fast. (Many thanks to Mike O'Dell and Peter Salus.)
I almost never attend a paper presentation at the conference because
I'd much rather read the paper. Most of the talks I have attended are
awful due to the speaker (not the topic). I have found the invited
talks to be very interesting and usually better presented than the
technical papers.
I also wish the BOFs were during the day. I got tied up each night
at San Antonio and missed all the BOFs that I wanted to attend. I'd
rather attend a BOF than a presentation of a paper that's in the
Proceedings.
The vendor exhibit at San Antonio was awful. As I understand it,
vendors were not allowed to attend even if they wanted to, as someone
(Usenix Board?) wanted to reduce the size of the exhibit to make it
"more friendly to chat with the vendors in a relaxed environment."
Supposedly only one vendor of each type was allowed to attend.
OK, I too dislike the zoo atmosphere that a Uniforum presents, but I
think it would be much better to restrict the actions of the vendors
(no magicians, no animals, etc.) than to cut it down to what San Antonio
had. I used to justify attendance at Usenix to my management based on
attendance at a good vendor exhibit where I could talk to many vendors
at a single location. Just what does Usenix plan for the vendor exhibit
at San Diego and Cincinatti?
In a field that's growing like Unix I am amazed at the lack of growth
of Usenix and the dwindling attendance at Usenix conferences. What's
wrong here? Other conferences are growing like crazy (Interop,
eXhibition), why not Usenix? I just don't buy the idea that all the
other conferences are taking everyone away. I also don't buy the idea
that we have to keep Usenix small to keep it focused.
Rich Stevens (rste...@noao.edu)
Scott Hazen Mueller
In article <1992Jun16.1...@noao.edu>, rste...@noao.edu (W. Richard Stevens) writes:
>I also wish the BOFs were during the day. I got tied up each night at
>San Antonio and missed all the BOFs that I wanted to attend. I'd rather
>attend a BOF than a presentation of a paper that's in the Proceedings.
Real good point. I'm still pretty new to the conference scene since my first
two jobs didn't send Tech Support types to conferences, but sysadmins live
better... Anyway, I've been to a Usenix and a LISA, and both time my wife
came with, though she stayed only for a long weekend for the latter. To some
extent in both cases, the conference pretty much had to be 9-5 so that I could
spend time with Nancy, which pretty well ruled out the reception in both
cases, and the BOFs at Usenix. Daylight BOFs would definitely have helped in
my case.
--
Scott Hazen Mueller +1 408 285 5762 sc...@dsg.tandem.com
Tandem Computers, Unix System Administrator, Email Postmaster and Usenet Admin
David W. Parter
ideas for future conferences. As I told him, this is a one-man
brainstorming session, not necessarily things I'd argue strongly for,
because I am not in a position to know the details and background to
really judge what will work and what won't. (Until San Anotonio, I
hadn't been to a Usenix since '87, but that's another story).
If someone wants me to elaborate on any of these ideas, send me mail
(or post your comments) and I'll see what I can do .... of course,
anyone else is free to jump on (or all over) these ideas. As was
pointed out to me, there are several ways we each can contribute to the
quality of future Usenix conferences. One way is to write papers (or
get someone else to write them!), another way is to come up with ideas,
and possibly do the work to follow up on those ideas that seem to have
merit....
--david
--------
david parter university of wisconsin -- madison
dpa...@cs.wisc.edu computer sciences department
==========================================================================
Comments, ideas, thoughts about Usenix: a one-man brainstorm
1. San Antonio Usenix:
hotels - very nice. The "other" hotel had doubles for the price
of singles, which was very nice for those on a somewhat
limited budget. Try and get the conference rate to last
through the weekend (since it is cheaper to stay over a
saturday)
coffee breaks - not enough coffee in the morning (before the
first session), reasonable variety of sweet rolls.
Perhaps there could be snaks of some kind for the
afternoon break as well... (nothing elaborate).
Does the hotel recycle? All those cans and bottles
going in the trash was annoying to see. I think USENIX
people would not mind seeing separate trash and
recycling bins at the coffee breaks.
terminal room - very nice, I had no problems with it.
2. Schedule
- what percentage of the participants attend the tutorials? What
percentage arrive in time for the Tuesday night BOFs? Bruce and
I came in late Tuesday night, and missed the bofs that night.
Maybe something could be done about that (like at least list
the conference as starting tuesday night, instead of wednesday
morning, so those who have to justify everything to the boss
can say they have to be there by 6pm.... or put bofs on the
schedule some other time....)
- weekend scheduling: with airfares cheaper if you stay over a
Saturday, maybe it is time to schedule the conference over a
weekend. Academics will probably like it (miss less classes),
industry people hate it (people who work plan other things for
the weekend, and comp time doesn't make it up).
2. Reception/Hospitality:
Contrary to what I said that night (when I said it was just
fine), the reception was actually a bit more than necessary.
Maybe the top 25% could have been shaved off... but then again,
I guess the only extravagant thing was shrimp... tacos are
cheap, and I consider an open bar a necessity...
[I'll elaborate here... I think the reception is important. It
is important that the conference participants as a group socialize
together. Without the reception, or with a sub-par reception,
people won't all be in the same room, and it will be much more
difficult to meet new people see old people and keep the personal
side of the "unix community" which is very important]
Someone was talking about having pizza at the bofs. I guess his
idea was to provide pizza and soda (or beer?) in the area where
the coffee was, and let people take it into whatever bof they
go to... I don't know that this is necessary or a good idea.
Some of the bofs are really marketing events, they should buy
their own...
I think it would be a good idea for the board to host a
reception (pizza and beer would be just fine, don't need hard
drinks all the time) the first night. This way, people could
socialize, and meet the board, and other people. An open Board
meeting serves a purpose (you aren't trying to get business
done, are you?), but talking in smaller groups in a more social
setting would also be good. To do this, you'd have to keep the
registration desk open later that evening, so (resonably) late
arrivals could get their name badge and then be recieved. (Oh,
I see from the program that there was a reception Sunday night.
Hmmm.... I think it should be Tuesday night, and specifically
hosted by the board, with all or most of the board there,
wearing something identifiable, in order to foster
communication between the masses and the board).
3. Papers/Technical Sessions:
submission deadlines - it is a problem that the deadline is so
far in advance, or to put it another way, so close
after the previous conference.
"sub-contract" the sessions: for some of the sessions, have a
member of the program committee take responsibility:
assign a topic, and have that person solicit papers ON
THAT TOPIC (which is different than you all soliciting
papers from whoever you can). Also, give the chair
flexibility to structure the session (or a double
session, both morning or both afternoon sessions) as he
or she sees fit.
invitied speakers: instead of only 90 minute invited talks,
have 45 minute invited talks, (mini keynotes) to kick
off regular paper sessions. 1 45 minute invited talk
followed by 2 20 minute submitted papers, for some of
the sessions (this can work with the "sub-contracted"
session chairs idea. If the sessin chair wants an
invited speaker, one can be invited).
various length talks: perhaps some 15 minute talks instead of
25 minute talks would make for better presentations.
Plus, if the submission deadline were later, (for
shorter talks) you might get more submissions that are
acceptable for shorter talks. Similarly, some topic
(and some speakers) deserve 40 or even 45 minutes.
mini symposia: combining the above ideas, a half-day
mini-symposium could focus on a topic, and by being
advertised and organized that way, achieve critical
mass (in terms of submissions, papers, participation).
"circus" session - Schedule a session (and make sure to
publicize it in advance, with *GOOD* publicity) that is
a "circus": the day before, ANYONE who wants to speak
shows up for a quick organizational meeting. The time
is divided evenly amongst all speakers (if there are 90
minutes and 9 speakers, everyone gets 10 minutes. 5
speakers and they get 20 minutes each. 15 speakers,
they get 6 minutes). Lots are drawn to determine the
order, and thats it. If handled right, (not quite like
the gong show), it can be dynamic, and give people an
idea of what others are working on. The advantage is the
short lead time -- you don't have to submit something 6
months in advance. You could do something similar with
the WIPS sessions, but somehow they are lower status
and people don't go (especially when scheduled against
the tool-off!). This can be used at workshops and other
conferences as well. Also very useful for people who
only get to go to conferences if they are presenting a
paper at the conference...
instructions to speakers: it is sad that it is necessary, but
some of the speakers need better instructions on what
size to make their slides, how to make an effective
presentation, etc.
question and answer: the session chair should run the question
and answer portion of each talk, making sure that one
participant does not monopolize the time, help the
speaker avoid being rude to a rude questioner, etc.
(number each mic with a reasonably big sign, so it is
easy to call on people).
panel discussions: on the right topic, with the right panel, a
panel discussion would be good for larger topics that
need discussing, rather then reporting.
use Computing Systems: Assuming that publication in Computing
Systems is considered somewhat prestigous (is it
refereed?), allow and encourage the best papers from
Usenix conferences (including the specialized
conferences) to be published in Computing Systems as
well as in the proceedings of the conference.
4. Other ideas/comments:
Introduction/open remarks: Every conference, by every
organization, has "opening remarks" on the program. I
don't remember what was said this time. I think that
one thing that was/is lacking from a Usenix conference
is a feeling of participation in USENIX THE
ORGANIZATION. For many, it is just a conference. I
think the board should be introduced in the opening
remarks (from the podium, not from the floor). The
Usenix president should make some brief remarks. Not
enough to bore everyone, just enough to let people know
that they are part of something, and can get involved.
Demo room: perhaps (as the trade show gets smaller and more
boring) there should be a demo room, where people can
show/play with new software. I have some ideas on how
to make the technical arrangements (people could bring
a system, or someone could provide some common
platforms for other to use), but I suspect other do too.
==========================================================================
Steven Bellovin
> other conferences are taking everyone away.
Last time I was on a Usenix program committee, there was a fair amount
of discussion about whether or not the average quality of the papers was
going down. Our tentative conclusion was that it was, and that the
specialized Usenix conferences (LISA, C++, Security, etc.) were at least
partly to blame.
Gregory M. Paris
>If you want "right now solutions" for management folks, then you are better
>off at Uniforum.
I haven't been to a Uniforum for several years, but maybe you're right.
Will I have to buy a new suit? :-)
>Now that you have stated what you perceive to be a problem, do you
>have any solutions to propose?
Not that I claim to be any kind of expert on crafting conferences, but
rather than let myself be painted as just a critic, I'll offer some
ideas (not all original, I admit):
1. Expand the tutorials from two days to three. Make more of the
tutorials half day instead of full day. For the people who
think that networking (of the human variety) is a vital part
of the conference, allow paying per tutorial so that attendees
don't have to be in tutorials for three days straight. Bag
the box lunches.
2. Move the BoFs to the day, as suggested here by at least a few
others. I too missed BoFs I was interested in due to
"other commitments." In particular, scheduling BoFs for
immediately after the reception seems a losing proposition
to me. Really though, after attending the conference from
9 to 5, who truly wants to go back at 7, 9 or 11?
3. Keep the Keynote.
4. Dump the Technical Presentations except for one or two papers
that should be be presented on the basis of "Best Submitted."
Publish the rest, but don't present them. Maybe it's not
appropriate, but a cash award for best paper doesn't seem out of
the question to me. Enlarge the time for the presentations from
the current 30 minutes to a period long enough to allow the
presentations to expand on the papers. As others have mentioned,
it's somewhat pointless to listen to the presentations when they
contain only a fraction of what's in the corresponding papers.
5. What to do with the Invited Talks... hmmmm. Well, if the
BoFs are during the day, I'm not sure what the point of the
Invited Talks is at all. I do like those panel discussions
they have at other conferences, why doesn't USENIX have any
of those (or did I somehow miss them)?
6. As others have mentioned, expand the Vendor Exhibits. I don't
see the point of the puny vendor area of the last USENIX. I
do think it helps "prove the value" of a conference trip if
one can bring back lots of vendor literature and a bunch of
business cards. Maybe some of you think this is silly, but
it's just part of the Real World.
7. Many have stated that the human networking that goes on during
the conference is one of its most important aspects. Taking
that to be true, why not help out the process? For instance,
get a room dedicated to the function. Have it supplied all
the time with coffee, soft drinks, etc., instead of the limited
supplies that appeared only during breaks. Possibly put the
terminal room function in this same area so that people can
do both kinds of networking at once. (I noticed this was
happening in the Marriott terminal room, but it was pretty
cramped in there; not much room for more than a few extra
bodies.) Also, I think more could be done to reach out to
people who are too shy to get involved in this networking.
Stereotypes aside, not every attendee is great at socializing,
expecially in the company of some of the more flamboyant and/or
well known USENIX attendees.
8. More fun stuff. Most people miss the facesaver; OK, we've
all heard the reasons why it's gone. That doesn't mean that
other fun techno things can't be set up. Hey, charge a fee
and help out the revenue situation. Some ideas: virtual
reality demo, beat the machine (at whatever), electronic
music competition, anything to spark some new interest.
9. Find some ways for people to be able to involve their spouses
(or other significant persons) in the conference. You have to
admit that the $40 to buy a ticket for the reception was just
a bit steep.
10. Do a better job of promoting the conference. For example,
do mailings to people identified as managers with money spending
authority. (You know, the kind of list that Uniforum uses. I'm
sure any of the trade rags would be happy to sell USENIX such a
list.) Try to convince these management types that they're really
missing out if they don't send their technical people to USENIX.
One other thing, I'd do away with the ribbons on the badges. In my
humble opinion, the ribbons are off-putting; they help make the
conference seem cliquish.
That's about all I have. Any further and I'd feel guilty about
stealing too many ideas from other posters.
--
Greg Paris <pa...@merlin.dev.cdx.mot.com>
Motorola Codex, 20 Cabot Blvd C1-30, Mansfield, MA 02048-1193
"I don't want to get, you know, here we are close to the election --
Andrew Partan
> vendors were not allowed to attend even if they wanted to, as someone
> (Usenix Board?) wanted to reduce the size of the exhibit to make it
> "more friendly to chat with the vendors in a relaxed environment."
> Supposedly only one vendor of each type was allowed to attend.
I think that you are rather wrong here - my understanding is that the
vendor show was small becuase not many vendors were interested in
showing. The general comment seems to be that there are too many other
shows to go to these days, and some of the other ones (like Interop &
Uniforum) are a lot larger.
As far as 'only one of each type' - did you count the number of book
places that were showing? There were at least 3 of these.
--a...@uunet.uu.net (Andrew Partan)
Christopher Rath
conferences, Winter'92 in San Francisco, and a Summer conference a couple
of years ago in Baltimore. Both conferences are excellent.
I would like to attend more often, but my employer will only pay for one
conference/course per year. In addition, if my work schedule (projects)
conflicts then you can guess what happens...I don't go.
The next conference I will be able to attend is the Summer'93 conference
in Ohio.
I say all that because there has been some discussion about why attendance
at USENIX conferences is not growing. My own lack of attendance has
nothing to do with the conference format or content, and everything to do
with work schedules and limited dollars.
All that said, I will add one remark. It would be nice to see the occasional
conference scheduled outside of the USA. It's no more expensive for me to
fly to Europe than San Francisco (cheaper, in fact).
---
Christopher Rath | cr...@bnr.ca
BNR Woodline |
Ottawa, ON, Canada | "Hydrogen is a colourless, odourless gas which, given
(613) 765-3141 | enough time, turns into people." -- Henry Hiebert
David Trueman
>br...@clarinet.com (Brad Templeton) writes:
>> c) It should run Thursday-Saturday, or even to Sunday morning, so
>> that people can get there on cheap fares. (And indeed, so that
>> cheap hotel rates can be had.)
>Please, no. It is bad enough to miss the family for several days. If
>the weekend is gone too, my kids would be attending USENIX before I
>know it.
I was going to suggest c) myself, but didn't have the courage :-)
As someone else with a family, I would like the conference to run on Saturday
at least, since I *have* to stay over Saturday night anyway (can't afford
higher airfares) and this would allow me to arrive at the conference
one (or two) days later, thus abandoning my family for less time.
--
{uunet watmath}!dalcs!david or da...@cs.dal.ca
Albert Cheng
In article <Bpzqp...@cs.dal.ca>, da...@cs.dal.ca (David Trueman) writes:
>I was going to suggest c) myself, but didn't have the courage :-)
>As someone else with a family, I would like the conference to run on Saturday
>at least, since I *have* to stay over Saturday night anyway (can't afford
>higher airfares) and this would allow me to arrive at the conference
>one (or two) days later, thus abandoning my family for less time.
But then you have to pay two more nights (fri. & sat.) of hotel cost
too. I once found out a way to go around this sat. stay over (SSO).
It was cheaper to buy two round trips SSO tickets than one regular
round trip. Say, I am flying from Chicago to L.A. and back. Use
first ticket departing Chicago late tuesday night, booking the return
flight 364 days later coming back. Use the second ticket departing
L.A. late friday night, booking the return flight 364 days later
coming back. If I have to fly between Chicago and L.A. again, I
could pay some penalty ($25 each, total $50) to change the dates
of both return flights. If I never got around to L.A. in a year,
I am still ahead, or I may sell/give/barter the two return flights
to someone else.
Rich Salz
> that should be be presented on the basis of "Best Submitted."
of saying "your talk should make people interested in reading your paper,"
and I thought most of the talks I attended did that. The Q&A afterwards
is invaluable, as is the opportunity to go up to the podium afterwards.
> Really though, after attending the conference from
> 9 to 5, who truly wants to go back at 7, 9 or 11?
I do. It's also a good way to seque to the bar or other socializing.
/r$
Brian Kantor
>It's also a good way to segue to the bar or other socializing.
Especially if you have some very thirsty long-hair burnt-out hippy dude
standing behind you chanting "bar. drink. bar. drink." and shooing you
towards that particular oasis....
- Brian
Barry Shein
Just to throw 2c in here on the vendor show:
A sizeable, professional vendor show is an enormous effort a large
cash exposure, it's easy to stick your neck out US$100K or more for
even a relatively modest show. Someone has to guarantee the price of
the space, security, setup (electricity etc) even if no one shows up,
by the time you could reel it in it's far too late, you're committed
to most of the expenses. On the plus side, they can be profitable, of
course.
The Nashville vendor show was a little bit of a loser, though nothing
serious, but we were fairly disappointed about the lack of interest.
The impression was that Nashville was just not where the vendors
wanted to show, and many of them are trying to stretch
recession-drained marketing budgets that may only allow them to show
at two to four shows in a year, so it's easy to see how they found it
easy to trim this one and opt for the flashy, big-city shows.
One trepidation was that San Antonio didn't have a whole lot going for
it (for vendor exhibits) over Nashville ON THIS SCORE (we're talking
vendor perceptions here, not facts, I like San Antonio just fine.)
See, a lot of vendors generally don't want to see JUST the technical
conference attendees, they also rely on a lot of walk-in traffic. In a
major location it's not unusual to have three times as many people do
a show-only pass as total technical attendee count. Agreed that the
technical attendees constitute a higher "quality" to show to, but
sheer numbers are important: How many brochures got handed out, how
many business cards were collected and what was the turnstile count
are often the only objective deliverables a marketeer has to bring
back to the office and let the boss know whether the $20K or more to
exhibit was well spent or not.
Consequently (ahem, wonder if anyone has read this far), it was
proposed that we do at least a friendly little show rather than none
at all and we went with that. I think it worked out just fine, but I
wouldn't extrapolate it to any future decisions, we'll take those as
they come and consider interest and location and market conditions for
each future conference.
--
-Barry Shein
Software Tool & Die | b...@world.std.com | uunet!world!bzs
Purveyors to the Trade | Voice: 617-739-0202 | Login: 617-739-WRLD
Barry Shein
From: cr...@BNR.CA (Christopher Rath)
>All that said, I will add one remark. It would be nice to see the occasional
>conference scheduled outside of the USA. It's no more expensive for me to
>fly to Europe than San Francisco (cheaper, in fact).
You might want to check out Europen's Utrecht, NL conference in
November, get yourself a gehaktbalt and some erdsoep (I probably
spelled those wrong, so shoot me.)
Terence P. Ma
>4. Dump the Technical Presentations except for one or two papers
> that should be be presented on the basis of "Best Submitted."
> Publish the rest, but don't present them.
> Enlarge the time for the presentations from
> the current 30 minutes to a period long enough to allow the
> presentations to expand on the papers. As others have mentioned,
> it's somewhat pointless to listen to the presentations when they
> contain only a fraction of what's in the corresponding papers.
Personally, I strongly disagree with this philosophy. I attend
quite a few meetings a year, most of which are at professional
scientific societies. At the Society for Neuroscience, for
instance, we have well over 12,000 papers presented in the 4.5 days
of the meeting (obviously in multiple simultaneous sessions).
This was my first USENIX and I, too, found problems with the
presentations. The main reason was that there were a few papers of
interest to me, that were scattered here and there. By and large,
the Invited Talks were outstanding.
What I suggest USENIX consider is a slightly different track.
1) Have shorter (15 minutes) presentations where the essential
kernel (so to speak) of NEW information is presented. Instead of
publishing full length papers (that don't count as refereed pubs
which are more important for promotion and job purposes at
academic institutions), publish the abstracts. Limit the number
of these "platform sessions."
2) Have "poster sessions" where presenters submit abstracts (some of
the organizations I attend have refereed [for acceptance or
rejection] abstracts) which are published. A poster board
(typical size being 4'x6') is available for each person to put
their results on the board. Then, schedule an hour or two that
the person has to be at their poster to talk to people. This
way, there is increased "networking," people can talk to other
people, people can choose those things of interest they want to
attend and see, this gives the presenter as much or as little
time as they want to present their topic.
Having presented many "platform" talks and many "posters," it is my
experience that platforms are easier to prepare. However, posters
are more useful, allow for better interaction, is less stressful on
me, and lets me talk in the way I talk best (as opposed to trying to
lecture from the platform).
Anyhow ... my ideas for the day.
Tere
--
Terence P. Ma, Ph.D. Department of Anatomy
VOICE: 601-984-1654 University of Mississippi Med. Ctr.
FAX: 601-984-1655 2500 North State Street
INTERNET: t...@anat.UMSMED.EDU Jackson, MS 39216-4505
Brian Kantor
rather disappointed in some of the papers this time that were but
thinly disguised advertisements for commercial products.
Oh, there have been a few of these in the past, but my perception is
that there were more of them this time than before.
Used to be that people would write up a nice paper after they'd come up
with something really spiffy, and be proud to present that paper at
Usenix so we'd all get the benefit thereby.
Given the current economy, it's not too surprising that the hope of
making some money off of it leads to the commercial content of some of
the papers. If it's not too blatant, few will mind. When there is
enough technical content to show the cleverness of the technique, or
inspire people to investigate it further, a little marketing isn't a
problem.
But some of the presentations ARE rather blatant marketing, and rather
devoid of technical value. Anything that references a 'proprietary
technique' shouldn't see the light of day at a Usenix technical forum.
What to do? Well, I think just discouraging it will work - a simple
statement by the Usenix board that the conference technical sessions
aren't really meant to be marketing presentations may well do the trick.
And then we issue compressed-air foghorns to key people in the audience.
When the commercial content goes over the threshold, they sound a
warning blast. Ought to do the trick, yes indeed.
Where is Chuck Barris when we need him most?
- Brian
Tom Christiansen
:From: cr...@BNR.CA (Christopher Rath)
:>All that said, I will add one remark. It would be nice to see the occasional
:>conference scheduled outside of the USA. It's no more expensive for me to
:>fly to Europe than San Francisco (cheaper, in fact).
:
:You might want to check out Europen's Utrecht, NL conference in
:November, get yourself a gehaktbalt and some erdsoep (I probably
:spelled those wrong, so shoot me.)
Don't forget Europen's Sevilla show in sunny Spain next Spring, either.
--tom
Terence P. Ma
>
>Just to throw 2c in here on the vendor show:
>
>it (for vendor exhibits) over Nashville ON THIS SCORE (we're talking
>vendor perceptions here, not facts, I like San Antonio just fine.)
Good point!
>Consequently (ahem, wonder if anyone has read this far)
I did!
>it was
>proposed that we do at least a friendly little show rather than none
>at all and we went with that.
If two USENIX conferences are going to be kept, then why not one
with a show and a real down-and-gritty tech conference?
Terence P. Ma
>1. San Antonio Usenix:
>
> coffee breaks - not enough coffee in the morning (before the
> first session), reasonable variety of sweet rolls.
> Perhaps there could be snaks of some kind for the
> afternoon break as well... (nothing elaborate).
Personally, I think that having the hotel catering service provide
all this is a waste of money for the organization. There are
restaurants, cafeterias, and other services available at any major
convention site, why should the organization pay for the food/drink
needs?
> - weekend scheduling: with airfares cheaper if you stay over a
> Saturday, maybe it is time to schedule the conference over a
> weekend. Academics will probably like it (miss less classes),
> industry people hate it (people who work plan other things for
> the weekend, and comp time doesn't make it up).
Most conferences I attend start on Sunday with specialty workshops
with "technical" sessions starting on Monday. For USENIX, a
reasonable alternative may be tutorials on Sunday and Monday
followed by the technical sessions.
>2. Reception/Hospitality:
[ ... ]
> [I'll elaborate here... I think the reception is important. It
> is important that the conference participants as a group socialize
> together. Without the reception, or with a sub-par reception,
> people won't all be in the same room, and it will be much more
> difficult to meet new people see old people and keep the personal
> side of the "unix community" which is very important]
Agree with this point. The organization should throw an opening
bash and a "reception" near the middle/end. But not all that extra
morning and break foods/drinks!
>3. Papers/Technical Sessions:
>
> invitied speakers: instead of only 90 minute invited talks,
> have 45 minute invited talks, (mini keynotes) to kick
> off regular paper sessions. 1 45 minute invited talk
> followed by 2 20 minute submitted papers, for some of
> the sessions (this can work with the "sub-contracted"
> session chairs idea. If the sessin chair wants an
> invited speaker, one can be invited).
I think that 1 hour special invited talks are VERY nice to have.
Having some "big name" person starting off a symposium, workshop,
platform session is fine, but I don't think it should be tied to the
invited talks.
> various length talks:
I have experience trying this, and it is a scheduling and political
nightmare.
> mini symposia:
Great idea!
Tom Christiansen
:> coffee breaks - not enough coffee in the morning (before the
:> first session), reasonable variety of sweet rolls.
:> Perhaps there could be snaks of some kind for the
:> afternoon break as well... (nothing elaborate).
:
:Personally, I think that having the hotel catering service provide
:all this is a waste of money for the organization. There are
:restaurants, cafeterias, and other services available at any major
:convention site, why should the organization pay for the food/drink
:needs?
Because a hotel intends to extract N dollars from us one way or the other,
where N is relatively constant. If we don't use their exorbitant catering
services for breaks and box lunches, then they will just find another way
of getting their money, such as charging for meeting room rental, having a
hire conference rate for the sleeping rooms, etc.
--tom
Steve McDowell
> What I suggest USENIX consider is a slightly different track.
>
> 1) Have shorter (15 minutes) presentations where the essential
> kernel (so to speak) of NEW information is presented. Instead of
> publishing full length papers (that don't count as refereed pubs
> which are more important for promotion and job purposes at
> academic institutions), publish the abstracts.
As long as the shorter presentations doesn't impact the Q&A period, then
I'm all for it.
Whether the papers are considered refereed or not is irrelevant. Having the
full text of the papers published is, I think, essential. The bound papers
provide an invaluable, and handy, reference.
Along these lines, I think that the papers should screened a bit more
stringently. There should be no papers that are blatant commercial
promotions -- the Winter conference had several. Although, on the
whole, I've been fairly pleased with the technical level of USENIX
submissions.
> 2) Have "poster sessions" where presenters submit abstracts (some of
> the organizations I attend have refereed [for acceptance or
> rejection] abstracts) which are published. A poster board
> (typical size being 4'x6') is available for each person to put
> their results on the board. Then, schedule an hour or two that
> the person has to be at their poster to talk to people. This
Aren't poster sessions more useful when you have a really *huge* number
of interesting submissions? Is this really the case with Usenix?
Personally, I like the presentation forum as it currently exists.
Barry Shein
From: t...@anat.UMSMED.EDU (Terence P. Ma)
>If two USENIX conferences are going to be kept, then why not one
>with a show and a real down-and-gritty tech conference?
I think we'll probably make the decisions based on looking at each
site and what seems reasonable rather than use the rhythm method.
(did I just say that? yes you did Barry, and in public too.)
Some have appropriate space, some don't, some are places vendors
express an interest in, some aren't, etc etc etc.
But I think we're getting the message that people like vendor shows
(put another way, that many like them seems clear, and I haven't
really seen anyone speak out against them.) Now we just have to figure
out how to keep them roughly cost-neutral.
Barry Shein
>There should be no papers that are blatant commercial
>promotions -- the Winter conference had several. Although, on the
>whole, I've been fairly pleased with the technical level of USENIX
>submissions.
This is the second posting that said essentially this. Could people
who feel this way be more specific so we could do a reality check
(i.e. which papers?)
Send to me directly (b...@world.std.com or whatever, replying to this
note is fine) if you wish to be discreet and I'll collect but I won't
repost any list (but we might want to go over this at a board meeting
and see if something needs to be thought about.)
Thanks.
Brad Templeton
certain appeal to the concept of a limited show. By this I don't mean
a small show, which is not that great, but a low-key show.
Limiting all vendors to a 10x10 booth and a basic setup provides a show
that is more about products than flash. And for smaller vendors it
can be a *lot* cheaper to exhibit at. You ship literature and what you
need to demo and you talk to prospects.
However, such a show of small booths would still be better if it had 50
vendors instead of 20. Not counting staff salaries, it cost me only
approximately $3,000 to show at this USENIX, which is not too bad.
(Although I admit to doing Uniforum for similar money because it was
a local show.)
--
Brad Templeton, ClariNet Communications Corp. -- Sunnyvale, CA 408/296-0366
Dick Dunn
How can you do this and make the conference fit in a week? Shortening the
main conference from three days to two seems unlikely to work, in
particular because it takes away an evening...
>2. Move the BoFs to the day, as suggested here by at least a few
> others. I too missed BoFs I was interested in due to
> "other commitments."...
Well, don't make "other commitments" then! (1/2:-)
>... In particular, scheduling BoFs for
> immediately after the reception seems a losing proposition
> to me...
It's not all that bad...as long as the BoF takes into account the condition
of the folks arriving from the reception!
>... Really though, after attending the conference from
> 9 to 5, who truly wants to go back at 7, 9 or 11?
I certainly do. Actually, they tend to start at 6 sometimes, which makes
dinner a bit chancy if the tech sessions are over at 5:30. But aside from
that, the reason I say "don't make other commitments" half-seriously, and
say that I'm glad for evening BoFs, is that when I go to the conference,
I'm there full-time. It costs a lot of money to be there--both money for
food/lodging, and income I can't make because I'm out of town. So the
best way to optimize that is to make the conference as concentrated as
possible, and that means using the evenings. I figure a good conference
day starts at 8:30 stumbling around looking for coffee and probably ends
after midnight in the bar after the last BoF.
As far as making other commitments...yes, I understand that as long as
you're traveling, it makes sense to take time to meet with people for
business, get together with friends, or just go see the sights. I do
that--I get together with people over lunch; I'll skip a session or a BoF
I'm not interested in to go do other things. But I really don't think the
whole conference should slow down just because of my diversions. Finding
time to do things within a busy conference is easier if you look over the
whole program schedule well in advance. If you really want extended time
for other things, make that Saturday, stay over, and get the cheaper
airfare.
--
Dick Dunn r...@raven.eklektix.com -or- raven!rcd Boulder, Colorado
...The way to meet an impossible circumstance is with voluntary craziness.
Marc Wiz
>Here are my random thoughts on Usenix conferences.
>
Stuff deleted.
>The vendor exhibit at San Antonio was awful. As I understand it,
>vendors were not allowed to attend even if they wanted to, as someone
>(Usenix Board?) wanted to reduce the size of the exhibit to make it
>"more friendly to chat with the vendors in a relaxed environment."
I asked someone about this and it was said that for vendor exhibits
you go to Uniforum. (I don't remember who it was and don't remember the
exact words but that was the gist of it)
Not wishing to sound cheap but some of us do pay our own way to go to Usenix
and that's all we get to go to. While my primary reason for going was for
technical reasons I was looking forward to a nice vendor exhibit area to see
what was new and hot in person. And while I have no data to back me up,
I would imagine the people in the booths would be there to answer questions
of a more technical nature than one would get at Uniforum.
This was my second Usenix and I thoroughly enjoyed it. I thought the choice
of San Antonio was wonderful. (No not because I'm in Austin :-)
Marc
--
Marc Wiz Yes that really is my last name.
MaBell: (512)244-8780
Internet: mw...@mpd.tandem.com
The views expressed are my own. Mine all mine!
W. Richard Stevens
>
>How can you do this and make the conference fit in a week? Shortening the
>main conference from three days to two seems unlikely to work, in
>particular because it takes away an evening...
I believe Interop this fall will be having tutorials Monday/Tuesday
and Thursday/Friday. Some of the more popular ones will be given
twice, giving the attendees the choice of attending the technical
sessions or a tutorial.
Rich Stevens (rste...@noao.edu)
Pat Wilson
>>There should be no papers that are blatant commercial
>>promotions -- the Winter conference had several. Although, on the
>>whole, I've been fairly pleased with the technical level of USENIX
>>submissions.
>This is the second posting that said essentially this. Could people
>who feel this way be more specific so we could do a reality check
>(i.e. which papers?)
I wonder if the program committee(s) weren't bitten by the same thing
that got us at LISA V. We accepted a set of extended abstracts that
looked great and "non-productlike" (even though they were describing
a vendor product), but by the time the papers came, the product had
been given an official name(TM), which was then used in the paper.
I don't think any of us (on the PC) realized how much difference this
would make to the overall "feel" of the papers, even though the (high)
technical content hadn't changed at all...
--
Pat Wilson
Systems Manager, Project NORTHSTAR
p...@northstar.dartmouth.edu
Michel Fingerhut
respect the Sabbath observers among us. It is the case that Usenix avoided
being scheduled on major Jewish holidays (which was not the case, unfortunately,
for the last one, whose first day started on the jewish Pentecost).
Steve Simmons
>Personally, I think that having the hotel catering service provide
>all this is a waste of money for the organization. There are
>restaurants, cafeterias, and other services available at any major
>convention site, why should the organization pay for the food/drink
>needs?
Most hotel services are not set up for 1000 people getting coffee
and donuts in 15 minutes, then vanishing. One might argue that they
should set it up in the lobby and charge for it.
--
Simmons' Law Of Diskless Workstations: "For every dollar you spend on
a diskless workstation, expect to spend a dollar on a file server."
Corrollary: "If you spend $.75 instead of $1.00, expect to spend another
$.75 fixing it when it doesn't perform to spec."
Peter Salus
[160 lines deleted]
> use Computing Systems: Assuming that publication in Computing
> Systems is considered somewhat prestigous (is it
> refereed?), allow and encourage the best papers from
> Usenix conferences (including the specialized
> conferences) to be published in Computing Systems as
> well as in the proceedings of the conference.
[more deleted]
Personally, as the managing editor of Computing Systems, I
wish Parter knew things prior to posting about them. Yes,
Computing Systems is refereed, Virginia. Not only that, but it has
in the past published revised and extended papers that had
previously been offered at Conferences and Workshops (a version
of the Synthesis paper, for example, was delivered at the IEEE
Real Time workshop; "Watchdogs," was a USENIX conference
paper). However, such papers are handled just as any other
submission: a minimum of three readers, frequently a lot of
revision requested, etc. Reprinting already published work
is not the way to run things. I would put the reputation of
the journal at great risk. We have achieved a level of recognition
in 4 years which many journals would envy. This has been achieved
through the efforts of Mike O'Dell (Ed. in Chief) and our volunteers
(the Board is listed in every issue of the journal). Only
extensively revised and extended conference papers should be
submitted to the journal: we encourage submissions, but we
don't need padding.
Peter H. Salus
Managing Editor
COMPUTING SYSTEMS
--
---------------------------------------------------------------------
Sun User Group, Inc; Suite 315; 1330 Beacon St.; Brookline, MA 02146
voice +1 617 232-0514 fax +1 617 232-1347
Steve Simmons
I've always thought of BOFs as semi-spontaneous workshops or meetings of
equals where informality rules. Invited talks are (or should be) more
focused and more one-to-many in orientation and attitude. Very different
beasts.
Howard C Berkowitz
>Although I found this conference quite enjoyable and informative, I was
>rather disappointed in some of the papers this time that were but
>thinly disguised advertisements for commercial products.
>Oh, there have been a few of these in the past, but my perception is
>that there were more of them this time than before.
>Used to be that people would write up a nice paper after they'd come up
>with something really spiffy, and be proud to present that paper at
>Usenix so we'd all get the benefit thereby.
>Given the current economy, it's not too surprising that the hope of
>making some money off of it leads to the commercial content of some of
>the papers. If it's not too blatant, few will mind. When there is
>enough technical content to show the cleverness of the technique, or
>inspire people to investigate it further, a little marketing isn't a
>problem.
>But some of the presentations ARE rather blatant marketing, and rather
>devoid of technical value. Anything that references a 'proprietary
>technique' shouldn't see the light of day at a Usenix technical forum.
>What to do? Well, I think just discouraging it will work - a simple
>statement by the Usenix board that the conference technical sessions
>aren't really meant to be marketing presentations may well do the trick.
Just remember to keep this in perspective. Some of the
user groups, especially in the IBM mainframe world, have gotten
to what I felt was a radical political correctness set of restrictions.
>And then we issue compressed-air foghorns to key people in the audience.
>When the commercial content goes over the threshold, they sound a
>warning blast. Ought to do the trick, yes indeed.
As an example of what is too much (and I will follow with
some positive suggestions), I stopped attending Computer Measurement
Group (which is primarily IBM mainframe performance) after receiving
a reprimand for the "commercialism" of my presentations.
I was shocked to receive that, as I had bent over backwards to
avoid marketing. I worked for a (now defunct) hardware performance
monitor company as a product architect, and was their representative
to ANSI and CCITT performance standards group. I presented two
papers on what I believed to be creative _algorithms_ for distributed
performance measurement, and then talked about the problems of
implementation on our box. Whenever I mentioned our product (as
a case study), I explicitly identified directly competing products.
Most of my design discussion related to formal standards work.
On asking some of the program committee what people felt was overly
commercial, I was informed that they were offended that the _borders_
on each of my transparencies had the company name in the bottom left
corner.
"General policies," in my experience, don't work. Specific guidelines
can. Vendor presenters of technical papers should at least be told:
1. You can/cannot refer to your products at all
a. You can refer to them only in the context of case studies
b. You can explicitly identify them only as one way of
doing something.
2. You can/cannot identify your employer
a. In your introduction
b. In your graphics "style"
c. In the body of the presentation
3. You must/do not need to identify competitors
4. You can/cannot discuss the _technical_ approaches used
by competitors, and contrast them _technically_ with yours.
Howard Berkowitz
PSC International
(703)998-5017
Henry Spencer
as a representative of Usenix in any way, shape, or form.)
In article <paris.7...@zygon.dev.cdx.mot.com> pa...@zygon.dev.cdx.mot.com (Gregory M. Paris) writes:
> ...Bag the box lunches.
Switch to bag lunches, you mean? :-)
Seriously, though... The box lunches are there for a very good reason:
it is very difficult for 1000 people to get fed reasonably well in the
time available unless you make special arrangements to feed them. Even
if the local area is thick with restaurants -- not always the case, as
witness Nashville -- that's still too many people for everybody to get
fed immediately. So no box lunches means a substantially longer lunch
hour, which cuts into the tutorial time available.
The problem is considerably reduced for the technical sessions, since
not everybody leaves for lunch at the same time, and in many cases the
lunch is a social event that is expected to run into the afternoon
sessions somewhat. There isn't the same synchronized human wave of
hungry people all stampeding out in search of food simultaneously.
The box lunches were invented to solve a very real problem; please don't
propose scrapping them without proposing an alternative.
> ... Really though, after attending the conference from
> 9 to 5, who truly wants to go back at 7, 9 or 11?
Lots of us, actually. ("Back"? You mean you leave? :-)) Usenix is
not a 9-to-5 kind of conference.
We *could* use more BOF slots, and putting some during the day would
be a reasonable approach. Especially the more marketeering-oriented
"BOFs", which are really unrefereed talks, not Birds Of a Feather sessions.
>5. What to do with the Invited Talks... hmmmm. Well, if the
> BoFs are during the day, I'm not sure what the point of the
> Invited Talks is at all.
BOFs -- real BOFs -- are not one person standing up at the front talking.
Fact is, the Invited Talks are clearly one of the bigger success stories
of recent Usenixes, and getting rid of them would be foolish. Their role
is to present interesting material that is *not* cutting-edge research
results and the like. I probably spent more time in them than in the
technical sessions this time, and the one program item I really regret
missing was an invited talk.
> ...I do like those panel discussions
> they have at other conferences, why doesn't USENIX have any
> of those (or did I somehow miss them)?
Usenix has had panel discussions at times; if you want more, volunteering
to organize them would help. :-) Did you miss the Tooloff?
>6. As others have mentioned, expand the Vendor Exhibits...
> ... it helps "prove the value" of a conference trip if
> one can bring back lots of vendor literature...
As others have mentioned, the big problem with this is attracting vendors.
(There are secondary issues of expense -- nothing is free at conferences.)
The ideal solution to this one is co-locating with Uniforum, which is the
big vendor show, but that's gotten increasingly difficult in recent years.
>7. Many have stated that the human networking that goes on during
> the conference is one of its most important aspects. Taking
> that to be true, why not help out the process? For instance,
> get a room dedicated to the function...
There have been some attempts at this in the past, as witness the Usenix
Lounge area (although it's turned into the Smoker's Ghetto, which renders
it useless to many of us...). They generally see little use. The hotel
bar and the nearby restaurants tend to serve this function instead.
> ...Have it supplied all
> the time with coffee, soft drinks, etc...
Would you pay an extra, say, $10 on your registration for this? Having
hotels do this is **EXPENSIVE**. These people make their money on things
like drinks and food; the actual guest rooms are expected to break even
at best.
Note that the hotel bar is already supplied this way, and *is* the room
you are looking for. Sure, a few non-Usenix people wander in, but
they're mostly harmless.
> ...Possibly put the
> terminal room function in this same area so that people can
> do both kinds of networking at once. (I noticed this was
> happening in the Marriott terminal room, but it was pretty
> cramped in there...
The terminal room tends to be limited by space available. On the few
occasions when the room actually has been bigger than really needed,
we've sometimes tried to provide a lounge-and-conversation area, which
has typically seen little or no use.
Bear in mind that the location of the terminal room is often heavily
constrained by (a) communications, (b) power, (c) lockability, and
(d) availability during the tutorials (which are *very* space-intensive).
Our problem with it is usually too little space, rather than too much.
We really can't reduce the number of terminals much, not with the level
of demand.
Any suggestion that we move the terminal room after the tutorials end
will be met by a suggestion that you volunteer for terminal-room setup
work a couple of times before making such a silly suggestion.
> ... Also, I think more could be done to reach out to
> people who are too shy to get involved in this networking.
> Stereotypes aside, not every attendee is great at socializing,
> expecially in the company of some of the more flamboyant and/or
> well known USENIX attendees.
Specific suggestions would be most welcome. There is actually quite a
bit of concern about this very issue. *What*, in particular, could
be done?
>9. Find some ways for people to be able to involve their spouses
> (or other significant persons) in the conference. You have to
> admit that the $40 to buy a ticket for the reception was just
> a bit steep.
It was almost certainly less than the per-head cost. Social events like
that are *EXPENSIVE* to mount, which is one reason for debate about
their value.
>One other thing, I'd do away with the ribbons on the badges. In my
>humble opinion, the ribbons are off-putting; they help make the
>conference seem cliquish.
Some of those ribbons are just about the only recognition that volunteer
helpers get. Volunteers typically are not in overwhelming supply; any
cheap way of encouraging them is worth it. Volunteered lately?
--
There is nothing wrong with making | Henry Spencer @ U of Toronto Zoology
mistakes, but... make *new* ones. -D.Sim| he...@zoo.toronto.edu utzoo!henry
Steve Simmons
>In article <paris.7...@zygon.dev.cdx.mot.com> pa...@zygon.dev.cdx.mot.com (Gregory M. Paris) writes:
>> ... Also, I think more could be done to reach out to
>> people who are too shy to get involved in this networking.
>> Stereotypes aside, not every attendee is great at socializing,
>> expecially in the company of some of the more flamboyant and/or
>> well known USENIX attendees.
>Specific suggestions would be most welcome. There is actually quite a
>bit of concern about this very issue. *What*, in particular, could
>be done?
One simple method to do this at a BOF is to urge people sitting in
the back to move closer to the front. A stock line like "Please sit
up front so I/we don't have to burn my throat out and we can hear your
comments; I promise not to be offended if you get up and leave later"
usually works. There's always a few stick-in-the muds, but it's
better than nothing.
Lee `nomaD` Damon
>(Standard disclaimer: I speak only as a fairly experienced attendee, not
>as a representative of Usenix in any way, shape, or form.)
>
>In article <paris.7...@zygon.dev.cdx.mot.com> pa...@zygon.dev.cdx.mot.com (Gregory M. Paris) writes:
>> ...Bag the box lunches.
>
>Switch to bag lunches, you mean? :-)
>
>Seriously, though... The box lunches are there for a very good reason:
>it is very difficult for 1000 people to get fed reasonably well in the
>time available unless you make special arrangements to feed them.
I would like to take this chance to thank the Mariott for doing such a
good job on the box lunches this time! The food was fresh, the
sandwiches had actual content, and putting the lettuce and tomatoes on
a seperate platform so we could help ourselves made a world of
difference in terms of sog.
Now if only all the other hotels would do it this way!
(Ya, I know about the vegitarian chicken salad on Tuesday, it was an
interesting error for a chef to make, but after all... it was Texas.
What do Texans know about vegans? :)
nomad
--
work: no...@watson.ibm.com - Lee "nomad" Damon - \
play: castle!nomad or no...@castle.org \
Systems Admin, The Farm, IBM TJ Watson Research Center, Yorktown NY / \
"Today's Oxymoron: America the Free" / \
Terence P. Ma
>> ... Also, I think more could be done to reach out to
>> people who are too shy to get involved in this networking.
>> Stereotypes aside, not every attendee is great at socializing,
>> expecially in the company of some of the more flamboyant and/or
>> well known USENIX attendees.
>
>Specific suggestions would be most welcome. There is actually quite a
>bit of concern about this very issue. *What*, in particular, could
>be done?
I was a first-time attendee in San Antonio. I tend to be quite shy
and quiet (though some would debate that point). I was thoroughly
distressed and stressed by the idea I should wear a First Timer
ribbon -- and I never did.
On the other hand, I found a lot of people who were very willing to
be friendly and helpful. I ended up visiting and talking and
listening in the hot tub and pool. I ended up having someone to eat
with almost every meal. People, old-timers and first timers, would
come up and start a conversation. I found the environment so
congenial that I felt comfortable going up to some "old-timers" and
asking really dumb newbie questions. (Perhaps in part this is
because Lee Damon took me under his wing?)
Hell, if I knew more about computers, I would seriously consider
switching my job to go into one where people are as congenial and
considerate as they were at USENIX. I attend a lot of conferences
every year, and USENIX was the best where environment is concerned.
Andrew Hume
~ Although I found this conference quite enjoyable and informative, I was
~ rather disappointed in some of the papers this time that were but
~ thinly disguised advertisements for commercial products.
~
~ Oh, there have been a few of these in the past, but my perception is
~ that there were more of them this time than before.
~
~ Used to be that people would write up a nice paper after they'd come up
~ with something really spiffy, and be proud to present that paper at
~ Usenix so we'd all get the benefit thereby.
~
~ Given the current economy, it's not too surprising that the hope of
~ making some money off of it leads to the commercial content of some of
~ the papers. If it's not too blatant, few will mind. When there is
~ enough technical content to show the cleverness of the technique, or
~ inspire people to investigate it further, a little marketing isn't a
~ problem.
~
~ But some of the presentations ARE rather blatant marketing, and rather
~ devoid of technical value. Anything that references a 'proprietary
~ technique' shouldn't see the light of day at a Usenix technical forum.
~
~ What to do? Well, I think just discouraging it will work - a simple
~ statement by the Usenix board that the conference technical sessions
~ aren't really meant to be marketing presentations may well do the trick.
~
~ And then we issue compressed-air foghorns to key people in the audience.
~ When the commercial content goes over the threshold, they sound a
~ warning blast. Ought to do the trick, yes indeed.
~
~ Where is Chuck Barris when we need him most?
~ - Brian
can you give specifics please?
andrew
Gregory M. Paris
>The box lunches were invented to solve a very real problem; please don't
>propose scrapping them without proposing an alternative.
Having no experience in conference organizing, I presume the obvious
solution of staggering the lunch breaks must be flawed in some subtle
way, otherwise I'd propose it as an alternative to box lunches.
>> ... Really though, after attending the conference from
>> 9 to 5, who truly wants to go back at 7, 9 or 11?
>Lots of us, actually. ("Back"? You mean you leave? :-)) Usenix is
>not a 9-to-5 kind of conference.
Then I don't see the point of having USENIX in nice places. If I'm going
to travel over a thousand miles to a nice city with gorgeous weather,
chances are I'm going to want to do something in that city besides hang
around in a refrigerated hotel. I work 10, 12, 14 hour days when I'm
at work; I'm not as inclined to do so when I'm travelling.
>> ...Have it supplied all
>> the time with coffee, soft drinks, etc...
>Would you pay an extra, say, $10 on your registration for this? Having
>hotels do this is **EXPENSIVE**. These people make their money on things
>like drinks and food; the actual guest rooms are expected to break even
>at best.
I understand we paid $50 extra on the registration for the reception.
How much did the box lunches cost? In a typical week at work I spend
about $7 on coffee; $10 doesn't seem outrageous to me, especially if
my company's paying for it.
>> ... Also, I think more could be done to reach out to
>> people who are too shy to get involved in this networking.
>> Stereotypes aside, not every attendee is great at socializing,
>> especially in the company of some of the more flamboyant and/or
>> well known USENIX attendees.
>Specific suggestions would be most welcome. There is actually quite a
>bit of concern about this very issue. *What*, in particular, could
>be done?
I know of one trick that is said to work. Each person gets a number on
the back of their badge. That number matches one or two other people
at the conference. The goal is to find them. Silly yes, but results
are what count, right?
I note there's a lot of glancing at badges to see if an unfamiliar
face is someone known from the net. Unfortunately, these glances
are mostly not followed up if the name on the badge isn't recognized.
I'm guessing the SIG-DOTS idea was intended to encourage followup;
how well did it work?
>It was almost certainly less than the per-head cost. Social events like
>that are *EXPENSIVE* to mount, which is one reason for debate about
>their value.
I mentioned the $40 because the reception was the only avenue for
involving a spouse or other significant person in the conference
activities. Sure, the BoFs are open to all, but there's not much there
for non-UNIX folk. All I'm suggesting is that additional evening
activities be organized that non-UNIX types can attend and enjoy for a
low cost. (If the BoFs are scheduled in the evening, unfortunately
there's a conflict.)
The point is that I think more people would attend USENIX if they
knew that bringing a non-UNIX person with them wouldn't end up
being a choice between leaving that person alone in the hotel while
BoFs are going on, or skipping the BoFs and doing something fun
with that person. This is an issue with me; maybe it's not with
other people. (Anyone care to speak up?)
>>One other thing, I'd do away with the ribbons on the badges. In my
>>humble opinion, the ribbons are off-putting; they help make the
>>conference seem cliquish.
>Some of those ribbons are just about the only recognition that volunteer
>helpers get. Volunteers typically are not in overwhelming supply; any
>cheap way of encouraging them is worth it. Volunteered lately?
This seems a bit disingenuous. The terminal room volunteers didn't
have ribbons. I don't remember the people at the USENIX booth having
them either. On the other hand, USENIX board members did and some had
two or three. I think the ribbons have the effect of elevating the
"inner circle" above the rest of the attendees. I won't even mention
the phrase "ego stroking." (Oops! :-)
--
Greg Paris <pa...@merlin.dev.cdx.mot.com>
Motorola Codex, 20 Cabot Blvd C1-30, Mansfield, MA 02048-1193
"I don't want to get, you know, here we are close to the election --
sounding a knell of overconfidence that I don't feel." --George Bush
Christopher Davis
GMP> I note there's a lot of glancing at badges to see if an unfamiliar
GMP> face is someone known from the net. Unfortunately, these glances
GMP> are mostly not followed up if the name on the badge isn't recognized.
Name recognition, at least for me, would have been *greatly* enhanced by
putting email addresses on the badges. Then I wouldn't have to rack my
already overburdened brain trying to remember if this ``Chip'' or
``Karl'' was the one I wanted to talk to, or whatever. Sure, I could
remember some names; it still would have been a lot easier with email
addresses.
The ClariNet @-buttons were a nice stopgap, but too few people had them.
--
Christopher Davis * c...@eff.org * System Administrator, EFF * +1 617 864 0665
Samizdata isn't that different from Samizdat. -- Dan'l Danehy-Oakes
Pat Wilson
>This seems a bit disingenuous. The terminal room volunteers didn't
>have ribbons. I don't remember the people at the USENIX booth having
>them either. On the other hand, USENIX board members did and some had
>two or three. I think the ribbons have the effect of elevating the
>"inner circle" above the rest of the attendees.
The USENIX booth folk *did* have ribbons - they read "STAFF". Not sure
about the terminal room people, though when I've worked there in the
past, we had "TERMINAL ROOM STAFF" ribbons. Sorry you feel it's cliquish.
As far as things for non-tech SOs to do, I must admit that I don't
see the point - I go to these conferences to work, see old friends (and
find out what they're up to), meet new people, etc. I often don't
bring my *technical* spouse because it's too distracting for me. If
I want to see the city, I either do that on my Saturday lay-over or
duck out during the day. If I want to go on vacation, I do that at
another time.
Christopher Rath
In article <1992Jun17.2...@news.eng.convex.com>, Tom
Where does one get more info. on all these nifty conferences?
---
Christopher Rath | cr...@bnr.ca
BNR Woodline |
Ottawa, ON, Canada | "Hydrogen is a colourless, odourless gas which, given
(613) 765-3141 | enough time, turns into people." -- Henry Hiebert
Tom Christiansen
:|> Don't forget Europen's Sevilla show in sunny Spain next Spring, either.
:|> --tom
:
:Where does one get more info. on all these nifty conferences?
We post the periodically. I believe what follows was the last such posting:
--tom
Date: 21 May 92 19:51:37 GMT
From: car...@usenix.ORG (Carolyn Carr)
Subject: Calendar of UNIX-Related Events
Organization: Usenix Association Office, Berkeley
Newsgroups: comp.org.usenix
******* Calendar of UNIX-Related Events ********
This is a combined calendar of planned conferences,
workshops, and standards meetings related to the UNIX
operating system. If you have a UNIX-related event that
you wish to publicize, please contact lo...@usenix.org.
Please provide your information in the same format as below.
This calendar has been compiled with the assistance of
Alain Williams of EurOpen.
* = events sponsored by the USENIX Association.
__________________________________________________________
1992
July 3 UKUUG, London, UK
13-17 IEEE 1003, Chicago, IL
20-22 Sun Open Sys. Expo, Anaheim, CA
27-31 Siggraph, Chicago, IL
Aug 10-13 * C++, Portland, OR
18 DKUUG, Helsingor, Denmark
Sep 8-11 AUUG, Melbourne, Australia
14-17 * UNIX Security III, Baltimore, MD
22-24 UNIX Expo, New York, NY
24 DKUUG, Copenhagen, Denmark
Autumn ISO/IEC JTC1 SC22 WG15
Denmark
NUUG, Norway
SUUG, Soviet Union
Oct 5-9 NLUUG, Amsterdam,
The Netherlands
6 WG15, Denmark
18-22 OOPSLA, Vancouver, Canada
19-23 * LISA VI, Long Beach, CA
19-23 IEEE 1003, Utrecht, The Netherlands
26-30 Interop, San Francisco, CA
29 DKUUG, Odense, Denmark
Nov 25-27 EurOpen/UniForum
Utrecht, The Netherlands
26 DKUUG, Copenhagen, Denmark
Dec 7 Sun User Group, San Jose, CA
UKUUG/UKnet, Manchester, UK
______________________________________________________
1993
Jan 11-15 ISO/IEC JTC1 SC22 WG15
New Orleans, LA
25-29 * USENIX, San Diego, CA
Feb 22-24 Sun Open Sys. Expo, Chicago, IL
Mar 15-18 UniForum, San Francisco, CA
24-31 CeBIT 93, Hannover, Germany
Spring * Mach
* UNIX Applications Development
Toronto, Canada
Apr 19-23 ISO/IEC JTC1 SC22 WG15
Irvine, CA
May 3-7 EurOpen, Seville, Spain
Jun 21-25 * USENIX, Cincinnati, OH
Jul 12-16 ISO/IEC JTC1 SC22 WG15
Autumn Europen/UniForum
Utrecht, The Netherlands
* LISA VII, West Coast, USA
* SEDMS
Oct 18-22 ISO/IEC JTC1 SC22 WG15
25-29 Interop , San Francisco, CA
__________________________________________________
1994
Jan 17-21 * USENIX, San Francisco, CA
Feb 14-17 UniForum, Dallas, TX
Mar 16-23 CeBIT 94, Hannover, Germany
23-25 UniForum, San Francisco, CA
Apr 18-22 EurOpen
Jun 6-10 * USENIX, Boston, MA
Sep 12-16 Interop, San Francisco, CA
Autumn Europen/UniForum
Utrecht, The Netherlands
______________________________________________________
1995
Jan 16-20 * USENIX, New Orleans, LA
Feb 21-23 UniForum, Dallas, TX
May 1-5 EurOpen
Jun 19-22 * USENIX, San Francisco, CA
_____________________________________________________
1996
Jan 22-26 * USENIX, San Diego, CA
Mar 11-14 UniForum, San Francisco, CA
Jeff Kellem
>>>humble opinion, the ribbons are off-putting; they help make the
>>>conference seem cliquish.
>he...@zoo.toronto.edu (Henry Spencer) writes:
>>Some of those ribbons are just about the only recognition that volunteer
>>helpers get. Volunteers typically are not in overwhelming supply; any
>>cheap way of encouraging them is worth it. Volunteered lately?
pa...@zygon.dev.cdx.mot.com (Gregory M. Paris) replies:
>This seems a bit disingenuous. The terminal room volunteers didn't
>have ribbons. I don't remember the people at the USENIX booth having
>them either. On the other hand, USENIX board members did and some had
>two or three. I think the ribbons have the effect of elevating the
>"inner circle" above the rest of the attendees. I won't even mention
>the phrase "ego stroking." (Oops! :-)
The terminal room volunteers normally DO have ribbons. Sometimes, the
ribbons don't end up in the volunteers packet, or the volunteer
doesn't bother (or perhaps notice it) to put it on. But, this can be
remedied; the ribbons are usually available. People at the USENIX
booths had such ribbons as "staff" or "student slave" (for student
volunteers that work with Evi Nemeth providing much of the setup for
the conference). Yes, not all did. But, they were there.
Then, you have the various speakers with their "speaker" or "invited
talks speaker" ribbons. These seem to be useful because they point out
speakers (imagine that ;-), which does seem to cause people to bring
up some conversation.. even just "so, what are you speaking on?". For
some people, it's just what they need to start a conversation.
The USENIX board member ribbons (the ones that say "board member" :-)
are also useful -- it allows you to corner (er, I mean, find ;-) the
various board members to talk informally about the USENIX Association
and how it's run.
I believe the ribbons are valid. Okay, so some people get carried
away with them. They provide the volunteers with a little recognition
and are useful (for the most part) in striking up conversation. Even
the "terminal room staff" ribbons have started conversations, when
I've volunteered, in the past.
Just some thoughts...
-jeff
Jeff Kellem
Internet: comp...@Beyond.Dreams.ORG
Tom Christiansen
:I understand we paid $50 extra on the registration for the reception.
:How much did the box lunches cost? In a typical week at work I spend
:about $7 on coffee; $10 doesn't seem outrageous to me, especially if
:my company's paying for it.
Henry was dead-on about costs and hotels. They are counting on the
catering to make them money. Typical costs are $3 per can of pop, $50 per
pot of coffee, $15-$20 or so for a box-lunch sandwich. I kid you not.
That's not even counting the sweet rolls. I don't really think your $10
would go very far.
Quite a few people who talked to me really enjoy the reception, and
wouldn't want it to go away entirely, although chicken satay instead of
boiled jumbo shrimp might buy you can of pop or two. :-)
:I'm guessing the SIG-DOTS idea was intended to encourage followup;
:how well did it work?
The SIGDOTS appeared to be quite effective at giving people something fun
to play with and which also performed a useful service: to help locate folks
with similar interests. The only problem was that the variations became
so complex that you couldn't tell the players without a score card. In
the future, I'd favor putting little icons on the dots, like the treble
clefs for musicians, or even the words. Some people did this, and it
helped a lot.
Anyone else have any thoughts on SIGDOTS?
:I mentioned the $40 because the reception was the only avenue for
:involving a spouse or other significant person in the conference
:activities. [...] more people would attend USENIX if they
:knew that bringing a non-UNIX person with them wouldn't end up
:being a choice between leaving that person alone in the hotel while
:BoFs are going on, or skipping the BoFs and doing something fun
:with that person.
I do know a fair handful of couples who attended, but in all cases
known to me, both members were technically oriented. I have to believe
that very few people bring non-technical SOs to a technical conference,
since this isn't really a family affair. And if they do, aren't they more
likely to leave the hotel and sightsee than to try to find an officially
sponsored, family-oriented conference function other than the reception?
I'd be very surprised to learn that SO-accompaniment were a significant
determining factor in deciding whether to attend the conference for all
but a very tiny minority of our membership.
:I think the ribbons have the effect of elevating the
:"inner circle" above the rest of the attendees.
The (totally optional) yellow newcomer ribbons were an attempt to
ameliorate any such feelings. The thought was that if everyone had at
least one ribbon, then it wouldn't seem so cliquish. Some people did wear
them, and I tried to talk to such people in particular and make them
feel welcome. There were some semi-serious newcomers wearing yellow
ribbons, too, like Dennis Ritchie, Larry Wall, and Mike O'Dell. :-)
We actually considered getting rid of the ribbons entirely. But ribbons
do serve two clear purposes: they indicate the function of the wearer and
to some extent also provide a reward for achieving some state, such as
being a speaker. The terminal room volunteer ribbons serve both
functions. It's nice to know whom to ask questions of, whether the ribbon
should say terminal room volunteers, staff, conference chair, or conference
coordinator.
Now, I doubt this usage bothers people much: I imagine it's the second
aspect that makes them uncomfortable.
Consider, however, someone who's come to conferences for a while. This
year they put in a lot of work into a paper, which they present. I bet if
you ask them how they felt about wearing the red speaker ribbon, that they
felt pretty good about it, somewhat like a boy scout who finally achieves
his merit badge. If at the next conference we didn't have any ribbons, we
might put off the speakers and volunteers for whom this was an effective
and cheap little pat on the back for their contribution to the conference.
--tom
J Greely
(Gregory M. Paris) writes:
>to travel over a thousand miles to a nice city with gorgeous weather,
>chances are I'm going to want to do something in that city besides hang
>around in a refrigerated hotel.
I've been at three conferences; at all three, the weather was "nice"
by someone's definition. Unfortunately, my definition of nice is what
most people call "cold, wet, and dark". In both Nashville and San
Diego, the only time I left the hotel was to head to a restaurant; in
San Antonio, I added the mall to my rambling. When Usenix comes to
Cincinnati next year, the *only* reason I'll be willing to leave the
hotel is for food.
>I work 10, 12, 14 hour days when I'm at work; I'm not as inclined to
>do so when I'm travelling.
So don't work. Relax, enjoy, cat-nap in your room between the
interesting talks.
>I note there's a lot of glancing at badges to see if an unfamiliar
>face is someone known from the net. Unfortunately, these glances
>are mostly not followed up if the name on the badge isn't recognized.
The lack of e-mail addresses comes up frequently. Many of the people
I know simply wrote it in over the mystery dot.
>I'm guessing the SIG-DOTS idea was intended to encourage followup;
>how well did it work?
While a lot of people had fun with the idea, I can't say that I
noticed them really being used. Maybe everyone else was colorblind,
too (my one problem with the SIG-dot concept; some of the colors were
"too close" for me, especially without a reference).
>This seems a bit disingenuous. The terminal room volunteers didn't
>have ribbons. I don't remember the people at the USENIX booth having
>them either.
I know the booth people did; that's where they were putting the
SIG-dots.
>On the other hand, USENIX board members did and some had
>two or three.
...with legends such as "grand moose", "lame duck", and "hanger on".
>I think the ribbons have the effect of elevating the "inner circle"
>above the rest of the attendees.
I didn't feel any ribbon envy, even though most of the people I hung
around with had them.
"IMHO, CAPS LOCK should be somewhere more
convenient; e.g., in another building."
--
J Greely (jgr...@cis.ohio-state.edu; osu-cis!jgreely)
Henry Spencer
>>The box lunches were invented to solve a very real problem; please don't
>>propose scrapping them without proposing an alternative.
>
>Having no experience in conference organizing, I presume the obvious
>solution of staggering the lunch breaks must be flawed in some subtle
>way, otherwise I'd propose it as an alternative to box lunches.
How far can you stagger them? Half of 1000 people is still too many.
A quarter of them is probably *still* too many. Even if we accept a
quarter, and assume that most of the lunch rush comes in the first
half hour (not that good an assumption, especially for lazy hotel
restaurants suddenly placed under heavy stress), it still means that
the lunch-hour starting times get spread from 1130 to 1300 -- which
is already too early for some people and rather too late for others.
Now figure in a few tutorial speakers who run late, and maybe some bad
weather outside so people have trouble getting to non-hotel restaurants.
Feeding that many people in that short a time just demands special
arrangements. If there's something specific about the box lunches that
you don't like, I'm sure Usenix would be interested to hear about it.
(Again, I speak only as an experienced spectator, not on their behalf.)
>>Lots of us, actually. ("Back"? You mean you leave? :-)) Usenix is
>>not a 9-to-5 kind of conference.
>
>Then I don't see the point of having USENIX in nice places...
Tourism is best done before or after. I spent two weeks in San Francisco
rather than one.
>... I work 10, 12, 14 hour days when I'm
>at work; I'm not as inclined to do so when I'm travelling.
Usenix is one of those events that is lots of fun but very draining.
It isn't a suits-required conference where everything ends at 5 and
*then* you can go out and have fun. If you don't need a rest after
Usenix, you haven't been getting the full benefit of the conference. :-)
>>> ...Have it supplied all the time with coffee, soft drinks, etc...
>>Would you pay an extra, say, $10 on your registration for this? Having
>>hotels do this is **EXPENSIVE**...
>
>I understand we paid $50 extra on the registration for the reception.
>How much did the box lunches cost? In a typical week at work I spend
>about $7 on coffee; $10 doesn't seem outrageous to me...
Bear in mind that $10 is only the start. That pays for about three soft
drinks at typical hotel rates (the ones they charge Usenix, not the ones
they pay *their* suppliers). I don't know what the numbers are like these
days, but I wouldn't be at all surprised if Usenix was paying $20 each
for those box lunches. We're talking a whole different league from what
you pay at work.
>I'm guessing the SIG-DOTS idea was intended to encourage followup;
>how well did it work?
My gut reaction is "people thought it was fun, but it didn't do much to
get people talking to each other". (But then, I was the one wearing the
"ban SIGDOTS" dot... :-))
>I mentioned the $40 because the reception was the only avenue for
>involving a spouse or other significant person in the conference
>activities. Sure, the BoFs are open to all, but there's not much there
>for non-UNIX folk. All I'm suggesting is that additional evening
>activities be organized that non-UNIX types can attend and enjoy for a
>low cost. (If the BoFs are scheduled in the evening, unfortunately
>there's a conflict.)
Like I said earlier, Usenix done right is a very intense event, and
your SO just isn't going to see you much during it. Mine came along
to SF, we touristed for a week before the conference, she took a couple
of the tutorials (she's semi-techie), and then she went sightseeing and
hiking for three days while I disappeared into the conference. We were
both happy with this. (But then, she's amazingly tolerant of my quirks.)
Usenix is not a vacation, although you can plan one around it.
>>>One other thing, I'd do away with the ribbons on the badges...
>>Some of those ribbons are just about the only recognition that volunteer
>>helpers get. Volunteers typically are not in overwhelming supply...
>
>This seems a bit disingenuous. The terminal room volunteers didn't
>have ribbons. I don't remember the people at the USENIX booth having
>them either...
The terminal room volunteers normally do get them -- certainly I wore my
"Terminal Room Staff" ribbon -- although they may overlook them (you
have to check your registration packet carefully, the ribbons are easy
to miss) or decide not to wear them. The Usenix booth people likewise
definitely got them, and at least some were wearing them.
>On the other hand, USENIX board members did and some had
>two or three...
I started out wearing three, and I'm not a board member. ("Terminal
Room Staff", "Speaker", and "Panelist".)
>I think the ribbons have the effect of elevating the
>"inner circle" above the rest of the attendees...
A legitimate concern. But recognizing volunteers and making it easy
to find speakers and board members is also important and useful.
>I won't even mention the phrase "ego stroking." (Oops! :-)
As I mentioned, stroking the egos of volunteers is important. If a few
lazy good-for-nothing board members :-) get stroked at the same time,
this seems a small price to pay.
Fred Feirtag
I think that people might understand the ribbon situation better if
they looked at the ribbons as a symptom, rather than a cause, of
cliquishness.
--
Fred Feirtag |
Naval Research Lab, Code 5120 | "I carry my adornments only on my soul,
Washington, DC 20375 | decked with deeds, instead of ribbons."
fei...@wave.nrl.navy.mil | --Cyrano de Bergerac
(202) 404 - 8502 |
Noah Friedman
>It's clear the hat made me more recognizable than the ribbon. Consider
>Andy Hume, Ed Gould, or even Tom Christiansen. All of them have some
>physical trait or mode of dress which makes them fairly easy to spot.
>Speakers who fit the stereotypical programmer model -- male, brown hair,
>beard, tending towards the heavy side -- should strongly consider
>a piece of clothing or carried article which makes them distinctive.
Len Tower (who is usually the only person from the FSF to attend Usenix)
told me he always wears a red sweater wrapped around his waist. Apparently
he gets cornered a lot more often than I did. This is probably partly
because this was my first conference and few people knew me, but I also had
few distinguishing physical characteristics.
David W. Parter
> ... [the] volunteers for whom this was an effective
>and cheap little pat on the back for their contribution to the conference.
I'd like to give a public pat on the back and thank you to everyone
involved in making Usenix happen, and in making it enjoyable and hassle
free -- at least for me, except that I didn't get a T-shirt :-)
--david
Steve Simmons
>Consider, however, someone who's come to conferences for a while. This
>year they put in a lot of work into a paper, which they present. I bet if
>you ask them how they felt about wearing the red speaker ribbon, that they
>felt pretty good about it, somewhat like a boy scout who finally achieves
>his merit badge. If at the next conference we didn't have any ribbons, we
>might put off the speakers and volunteers for whom this was an effective
>and cheap little pat on the back for their contribution to the conference.
I agree with Tom on this one -- my speaker ribbon is already up on my
bulletin board. Being recognised after the talk and the many questions
and comments over the next few days added immensely to my enjoyment of
USENIX. But I don't think the ribbon did it.
Those who have seen me before this USENIX know I don't normally wear a
hat. At USENIX I wore the hat for Elizabeths and my talk, and then
made a point of wearing it the rest of the conference. It felt like
many more folks came up and talked later about our talk than after
other public things I've done at USENII. As an experiment, Friday
morning I didn't wear the hat, and the number of comments/questions
went way down. At noon I put it back on, and at least three people
who'd I'd seen earlier that day said "Oh, there you are."
It's clear the hat made me more recognizable than the ribbon. Consider
Andy Hume, Ed Gould, or even Tom Christiansen. All of them have some
physical trait or mode of dress which makes them fairly easy to spot.
Speakers who fit the stereotypical programmer model -- male, brown hair,
beard, tending towards the heavy side -- should strongly consider
a piece of clothing or carried article which makes them distinctive.
The hat did everything the ribbon should have done, and did it without
implying any class divisions. Don't get me wrong, I think speaker,
board and staff ribbons are a good idea. But for those who want to
avoid them, there are other ways of getting the same result.
--
"Ace/MIPS [i.e. the ultrix/OSF version] was another of Ken Olsen's little
jokes, mostly on MIPS."
Dave Probert in in comp.sys.sun.misc article <probert.699728622@orcas>
Scott C. Kennedy
|> he...@zoo.toronto.edu (Henry Spencer) writes:
|> >> ... Also, I think more could be done to reach out to
|> >> people who are too shy to get involved in this networking.
|> >> Stereotypes aside, not every attendee is great at socializing,
|> >> especially in the company of some of the more flamboyant and/or
|> >> well known USENIX attendees.
|>
|> >Specific suggestions would be most welcome. There is actually quite a
|> >bit of concern about this very issue. *What*, in particular, could
|> >be done?
|>
|> face is someone known from the net. Unfortunately, these glances
|> are mostly not followed up if the name on the badge isn't recognized.
|> I'm guessing the SIG-DOTS idea was intended to encourage followup;
|> how well did it work?
I got some useful things from the sig-dots before they went a little overboard,
but as with all good ideas things get taken to extremes.
Another suggestion which has been brought up year after year, is why
aren't our email addresses on our badges? We supply them every year.
Do you really know that Jeff Kellem == composer@Beyond_Dreams.org?
Or who is snark? etc..... (the list could go on)
|> The point is that I think more people would attend USENIX if they
|> knew that bringing a non-UNIX person with them wouldn't end up
|> being a choice between leaving that person alone in the hotel while
|> BoFs are going on, or skipping the BoFs and doing something fun
|> with that person. This is an issue with me; maybe it's not with
|> other people. (Anyone care to speak up?)
I agree that UseNIX is kind of hard on those SOs not involved with unix,
or with computers in general. But, it is a technical conference. I don't see
much for a non-Unix person to do apart from hang out with some really
estoteric people and try to find out what is so funny about this....
/bin/rm -r /
echo "were you sure?"
(As seen on a t-shirt at UseNIX [Hi Greg!])
|> >>One other thing, I'd do away with the ribbons on the badges. In my
|> >>humble opinion, the ribbons are off-putting; they help make the
|> >>conference seem cliquish.
|>
|> >Some of those ribbons are just about the only recognition that volunteer
|> >helpers get. Volunteers typically are not in overwhelming supply; any
|> >cheap way of encouraging them is worth it. Volunteered lately?
|>
|> This seems a bit disingenuous. The terminal room volunteers didn't
|> have ribbons. I don't remember the people at the USENIX booth having
|> them either.
Speak for yourself, most of the volunteers at the booth had "Staff" Ribbons, so
we could get coffee, and knew who to ask to help us lift this heavy box. Not
to mention the ribbons are VERY useful to newcomers when you say, ask a
staff member that. In, fact that one there... You know the one with the glasses,
long hair, wearing a t-shirt, looks kind of odd, (just described 50% of the entire
conference)
--
------------------------------------------------------------------------------
Scott C. Kennedy (s...@watson.ibm.com) | This post does not reflect the intent
Distributed High Performance Computing | or actions of my employer, and their
IBM Thomas J. Watson Research Facility | actions don't reflect mine either.
------------------------------------------------------------------------------
I had this dream the other night. I went to work one day, and nobody
remembered who I was. So, I decided to take the day off. - 2nu
Peter Whittaker
>Name recognition, at least for me, would have been *greatly* enhanced by
>putting email addresses on the badges. Then I wouldn't have to rack my
>already overburdened brain trying to remember if this ``Chip'' or
>``Karl'' was the one I wanted to talk to, or whatever. Sure, I could
>remember some names; it still would have been a lot easier with email
>addresses.
Hear, hear!
pww
Charleen Bunjiovianna
>
>>helpers get. Volunteers typically are not in overwhelming supply; any
>>cheap way of encouraging them is worth it. Volunteered lately?
>
>This seems a bit disingenuous. The terminal room volunteers didn't
>have ribbons.
Um, we certainly did. Although all the terminal room slaves :-) might not
have worn them.
BTW, I enjoyed working the terminal room; it was a good way to meet
people and get exposure to interesting problems. Thanks to Eve Podet
for giving me the opportunity.
Charleen
--
Q: How do most Montana boys end their first romance?
A: When the 4-H project goes to market.
Boyd Roberts
>
> /bin/rm -r /
> echo "were you sure?"
>
> (As seen on a t-shirt at UseNIX [Hi Greg!])
>
Let's get it right:
del()
{
rm -rf "$@"
echo 'Were you sure? \c'
read ans
}
[blue constant width font on white T-shirt]
For the particularly pedantic this requires the V8 shell with a System V
style echo(1).
One the subject of ribbons: One person's (large) ribbon collection is
another person's clique.
Boyd Roberts bo...@prl.dec.com
``When the going gets wierd, the weird turn pro...''
Dick Dunn
>Having no experience in conference organizing, I presume the obvious
>solution of staggering the lunch breaks must be flawed in some subtle
>way, otherwise I'd propose it as an alternative to box lunches.
I'm not an organizer either, but I've taught. The tutorial schedule itself
constrains the problem. You don't want to start too early (esp if you're
in any time zone other than Pacific); you don't want to end too late
(because there are other things going on). There are four parts to the
tutorial, and you'd like them to be approximately equal length. At a
total of 6 hours allotted, if you shrink one of the parts to, say, an hour,
you lengthen another to two hours. That's too long without a break or
change of scenery. It's doubly true in the afternoon--the guy who teaches
the tutorial that gets an early lunch really loses.
Hotels are geared for serving large groups of people in short periods;
might as well let them do it for the tutorials.
>>> [after] 9 to 5, who truly wants to go back at 7, 9 or 11?
>>Lots of us, actually. ("Back"? You mean you leave? :-)) Usenix is
>>not a 9-to-5 kind of conference.
>
>Then I don't see the point of having USENIX in nice places. If I'm going
>to travel over a thousand miles to a nice city with gorgeous weather,
>chances are I'm going to want to do something in that city besides hang
>around in a refrigerated hotel...
Well, I'd readily argue against saying either Nashville or San Antonio in
the summertime had "gorgeous weather"! But that's not the point. The
location of USENIX has some importance, but it's not overwhelming. When
we start seeing the main conferences on Maui or the Riviera, and workshops
in Telluride, *then* I'd say the location is significant.
If you're going to make the trip to do things other than the conference,
fine--do them outside the conference, or skip some things. The point of
the conference isn't vacation. Once again, I don't see the sense of
diluting the content of the conference in order to accommodate things that
don't have to do with the conference. Some folks have spare time they'd
like to spend doing other things; other folks have very constrained
schedules. So let the conference be as concentrated as possible.
>I note there's a lot of glancing at badges to see if an unfamiliar
>face is someone known from the net. Unfortunately, these glances
>are mostly not followed up if the name on the badge isn't recognized.
I don't look just for recognized names with unfamiliar faces; I look for
something on the badge that might make me find a way to strike up a
conversation.
But I *do* wish we could go back to putting the email addresses on the
badges. I miss that.
>I mentioned the $40 because the reception was the only avenue for
>involving a spouse or other significant person in the conference
>activities. Sure, the BoFs are open to all, but there's not much there
>for non-UNIX folk...
>...All I'm suggesting is that additional evening
>activities be organized that non-UNIX types can attend and enjoy for a
>low cost...
I'm not quite following your reasoning here. How much effort does the
conference need to make to accommodate people who aren't there for the
conference? ...especially when you're saying that you want to *remove*
evening activities (BoFs) for the people who *are* there for the
conference.
Hypothesis: If the BoFs were moved to daytime, the folks who aren't into
playing tourist in the evening would end up doing something very similar to
BoFs anyway, except without benefit of organization...the end result being
only that people would miss techie get-togethers they'd like to have joined.
There are 3 usable evenings (Tues-Wed-Thu). The reception takes care of
one of them. For the others--well, look around and see what's going on.
There's always a local-information desk that can tell you about sights to
see, restaurants, etc. Ask at the hotel. Check out the area in travel
guides before you go.
If you decide to bring your spouse/family with you, OK--but that's really
your decision and yours to deal with. The conference arrangements make
some allowances for it (I think they do fine, based on when Diane's been to
conferences with me), but that's really not the point of the conference.
As Henry said, it's not a family affair.
>The point is that I think more people would attend USENIX if they
>knew that bringing a non-UNIX person with them wouldn't end up
>being a choice between leaving that person alone in the hotel while
>BoFs are going on, or skipping the BoFs and doing something fun
>with that person...
What requires the other person to sit alone in the hotel? What does that
person do during the daytime?
It's not as if every evening time slot has a BoF that's absolutely
essential. Pick the ones you want. Skip the marginal ones and go have
fun, but let the rest of us carry on.
--
Dick Dunn r...@raven.eklektix.com -or- raven!rcd Boulder, Colorado
...The way to meet an impossible circumstance is with voluntary craziness.
Melissa Metz
Is there room for two more cents?
he...@zoo.toronto.edu (Henry Spencer) said:
> [...] pa...@zygon.dev.cdx.mot.com (Gregory M. Paris) writes:
> > people who are too shy to get involved in this networking.
> Specific suggestions would be most welcome.
Tom Christiansen <tch...@convex.COM> said:
> [...] locate folks with similar interests. The only problem was
> that the variations became so complex that you couldn't tell the
> players without a score card. Anyone else have any thoughts on
> SIGDOTS?
I thought the SIGDOTS were great for starting conversations. (And I'm
one of those shyer people.) If someone were wearing the same dots as
me, I recognized the dots and could start a conversation based on that
(I limited myself to four dots, so I could remember what they all
meant). If someone had dots I didn't recognize, "what do those dots
mean?" worked pretty well for me :-). I talked to a lot more
strangers than I did at past conferences.
As for ribbons, perhaps they could be displayed on the SIGDOT board.
I never got close enough to any luminaries (especially the
multi-ribboned ones) to read their ribbons...
I'd like to third the recommendation to put email addresses on the
badges. Every year, I'm surprised they're not there...
Melissa Metz
Unix Systems Group
Academic Information Systems
Henry Spencer
>Hypothesis: If the BoFs were moved to daytime, the folks who aren't into
>playing tourist in the evening would end up doing something very similar to
>BoFs anyway, except without benefit of organization...
The trick, I think, is to distinguish real BOFs from lectures and sales
presentations masquerading as BOFs.
A modest proposal... Any BOF organizer has to state how many slides
(overheads, whatever) he is going to present. If the number exceeds three,
he's a marketeer and he gets a daytime slot. If not, he's probably planning
to hold a real Birds Of a Feather session, not a lecture, and he gets
one of the prime-time (evening) slots.
The particularly cute aspect of this is that the marketeers think a
daytime slot is an *advantage*, so they have no incentive to lie!
Of course, this means Rob Kolstad probably ends up giving the BSDI talk
during the day, but people will come to hear him regardless...
Paul Graham
|>Some of those ribbons are just about the only recognition that volunteer
|>helpers get. Volunteers typically are not in overwhelming supply; any
|>cheap way of encouraging them is worth it. Volunteered lately?
|
|"inner circle" above the rest of the attendees. I won't even mention
|the phrase "ego stroking." (Oops! :-)
the ribbons are sposed to help you find people. they have also become
a bit of humor. i dunno if this adds, detracts or is neutral with resp. to
the "inner circle". this time newcomer ribbons were used, in the past
old-timer buttons were used. all are items intended to facilitate the
old yaccity yacc.
best of [breed|show], lame duck, hanger-on et.al. really are intended to
be funny and perhaps take the edge off the multiply be-ribboned.
just remember, unix is joke, and we want you to laugh with us twice a
year. perhaps someone should simply bring a generous supply of extras
and hand them out to the ribbon challenged.
--
p...@acsu.buffalo.edu / rutgers!ub!pjg / pjg@ubvms (Bitnet)
opinions found above are mine unless marked otherwise.
Greg Rose
>3. Much of the "interesting" work is being done in UNIX-related
> areas, for example, X, that just plain aren't UNIX. Relatedly,
> system admins, such as myself, often are more concerned with
> interoperation of the many commercial applications they're
> required to support than they are with UNIX evolution.
While I agree thoroughly with all the thoughts expressed, I want to
pick up on the one quoted above, and say "So??" Surely this means
simply that the conference should not have such a total (perceived)
focus on Unix as such, and should refocus on these other topics. I
think that many appropriate papers, such as the one which won the best
student paper ("The Recovery Box", a development in Sprite) had nothing
directly to do with Unix. Let's recognise that, and keep doing it.
--
Greg Rose Australian Computing and Communications Institute
g...@acci.com.au +61 3 282 2532
Stephane Boucher
> :I think the ribbons have the effect of elevating the
> :"inner circle" above the rest of the attendees.
> The (totally optional) yellow newcomer ribbons were an attempt to
> ameliorate any such feelings. The thought was that if everyone had at
> least one ribbon, then it wouldn't seem so cliquish. Some people did wear
> them, and I tried to talk to such people in particular and make them
> feel welcome. There were some semi-serious newcomers wearing yellow
> ribbons, too, like Dennis Ritchie, Larry Wall, and Mike O'Dell. :-)
It was my first Usenix conference. So I did wear the
newcomer ribbon. Since I'm rather shy, I tought this might
help getting in touch with people. I tought the SIGDOTS
would also have helped.
As it turned out, the SIGDOTS where amusing altough useless.
In the case of the newcomer ribbon, it never helped me get
in touch with others. There was maybe this one time at the
reception on thursday (a bit late IMHO) when I was wondering
around trying to decide what to eat next. Someone tried to
be nice, because I probably looked lonesome, and asked me if
I had a good time.
I had the perl dot and Larry did not even bother to speak to
me :-). But seriously, in the end, I only got to speak with
others by making the first move. Could it be that I looked
too straith? :-)
I think the newcomer ribbon is a nice idea, if only it could
really serve a purpose. A newcomer ribbon just to avoid
feeling out of the clique is not necessary.
But overall, I enjoyed Usenix.
--
, , ,
Stephane Boucher, ing. jr. | Ecole Polytechnique de Montreal
s...@vlsi.polymtl.ca | Tel: (514)340-5950
s...@info.polymtl.ca | Fax: (514)340-4078
Larry W. Virden
:In article <paris.7...@zygon.dev.cdx.mot.com>, pa...@zygon.dev.cdx.mot.com (Gregory M. Paris) writes:
:|> he...@zoo.toronto.edu (Henry Spencer) writes:
:|> knew that bringing a non-UNIX person with them wouldn't end up
:|> being a choice between leaving that person alone in the hotel while
:|> BoFs are going on, or skipping the BoFs and doing something fun
:|> with that person. This is an issue with me; maybe it's not with
:|> other people. (Anyone care to speak up?)
:
:I agree that UseNIX is kind of hard on those SOs not involved with unix,
:or with computers in general. But, it is a technical conference. I don't see
:much for a non-Unix person to do apart from hang out with some really
:estoteric people and try to find out what is so funny about this....
Before our family reached it's current size, my wife traveled with me
to conferences. The ones she enjoyed the most were the ones which made
some attempt to provide info on local points of interest to attendees
for during the day trips. She had a real blast at two or three of the
conferences because the organizers had a special room set up for all
the non-techies to meet and go places together. Then, when I dropped
out of sessions due to data overload, she would take me around to the
most interesting of the day's discoveries. I would decompress, come back,
and try to catch a late evening BOF, etc.
What always surprises me at some conferences (Not Usenix - I havent had
the privelege yet of attending one) is that so few interesting evening
sessions are available. When I attend a conference these days, I am
by myself, and know few if any people there. Being another of the
shy types (hey, maybe we need some special way to id each other at
these things ;->) I tend to wander around looking lost till I get
so bored that I go back to the room and do some tube surfing and then
conk out.
What I would rather have is some place where I could sit around and chew
the fat - sort of in-the-flesh usenet-ing if you would. I don't tend to
fratenize the bars or restraunts by myself - I find eating alone to be
rather boring.
This is not a conference's problem though - it's my own personality
flaws at work here. If I just ran out and started banging on doors,
I probably would find someone eventually who would have me locked up or
feel sorry and go to dinner with me...
Anyways, just a note from someone who doesn't get to travel in groups
much.
--
Larry W. Virden UUCP: osu-cis!chemabs!lvirden
Same Mbox: BITNET: lvirden@cas INET: lvi...@cas.org
Personal: 674 Falls Place, Reynoldsburg, OH 43068-1614
America Online: lvi...@aol.com
Greg Rose
people not try to limit the distribution of articles to, for example,
the USA or North America? Usenix claims to be an international
organisation, after all... (Fortunately most news distributors are
broken enough to get the articles to Australia anyway...)
In article <1992Jun22.1...@cas.org> lvi...@cas.org
(Larry W. Virden) writes about the problems of breaking the ice at
these conferences, and finding people to hang out with.
Coming all the way from Oz to Usenix conferences used to do the same
thing to me. I knew virtually no-one, and had no particular way to
meet people. So I adopted a novel strategy. I looked for people with
ribbons on, and would try to talk to them about whatever it was that
their ribbons said they did. This works well for speakers, since they
are there to talk after all, and sort-of works for the executives of
the organisation, since that is what they are paid for after all...
(:-) for those who don't recognise the sarcasm there.) I also remember
when the BOF sessions were treated as social occasions by some, and
tried to further that end by organising the first rec.aviation BOF. A
few more of these and you'd have no trouble breaking the ice.
I got off track above. The point I wanted to make is that making
connections is actually hard work, and you have to put effort into
doing it. It doesn't come free. If you can't bring yourself to talk to
complete strangers, then instead volunteer to do something at the
conference, such as organising a BOF for your own special interests,
putting up a sign about going to your favourite type of restaurant,
helping to sell T-shirts, or something. The more you put in, the more
you'll typically get out.
Anyway, a few years ago I knew no-one at Usenix conferences. Now I get
to wear ribbons, even if they are joke ones. (I volunteered a new
line of Joke Ribbons, you see...) I still don't know anyone though (;-).
Barry Shein
Just a comment on ribbons:
I'm listening, and I'll admit I'm undecided, but people really do come
up to you if, for example, you're wearing a board ribbon (as I do),
often with problems.
Anything from general suggestions and comments (gack! what is this Big
Red soda???) to immediate kinds of emergencies (umm, excuse me, but
there's 200 people in that room over there and no one up front and
things should have started 15 minutes ago and the crowd is getting
kinda ugly, maybe you can help?) to even personal kinds of problems
(hi, i have a confirmation here for a room and now the hotel is saying
they have no rooms, what should I do?)
Even if some of those things aren't an official function most of us
with staff or board ribbons are very experienced travellers and often
know, for example, how to go chat with hotel staff and get a rule bent
or some misunderstanding smoothed over or even just find the right
person to help straighten something out. Don't underestimate
experience (and appearing to represent the organization that is
bringing around $1M into the hotel right this moment.)
The worst thing that ever happened to me (tho I wasn't directly
involved in this one I was right outside, other board members were
tho), and this wasn't USENIX, was when a woman passed out (and,
unfortunately, died) during a technical session.
Sometimes it's good to be able to find someone who can take charge,
fast, even if it seems silly most of the time.
--
-Barry Shein
Software Tool & Die | b...@world.std.com | uunet!world!bzs
Purveyors to the Trade | Voice: 617-739-0202 | Login: 617-739-WRLD
Elizabeth Zwicky
>The point I wanted to make is that making
>connections is actually hard work, and you have to put effort into
>doing it. It doesn't come free. If you can't bring yourself to talk to
>complete strangers, then instead volunteer to do something at the
>conference, such as organising a BOF for your own special interests,
>putting up a sign about going to your favourite type of restaurant,
>helping to sell T-shirts, or something.
My theory was to talk to the complete strangers several hundred at a
time, by giving papers. This is not actually any scarier for me than
talking to them one at a time, and people actually *expect* you to be
nervous. Having done this I discovered two important facts: 1) Those
rings of people to be found surrounding ribbon-wearers are not usually
their closest friends. One or two of them, maybe; but a large
percentage are complete strangers. 2) People who you recognize who are
standing around by themselves in the middle of things are usually not
doing so because they have developed a sudden desire to be alone and
sent everybody away. They are doing so because there isn't anybody
they recognize to talk to and they're too shy to walk up to strangers.
I no longer feel bad about talking to people with ribbons on, whether
they're mobbed or alone.
Elizabeth D. Zwicky
zwi...@erg.sri.com
Simon Edward Spero
Just a comment on ribbons:
[...]
Red soda???) to immediate kinds of emergencies (umm, excuse me, but
there's 200 people in that room over there and no one up front and
things should have started 15 minutes ago and the crowd is getting
kinda ugly, maybe you can help?) to even personal kinds of problems
[...]
Sometimes it's good to be able to find someone who can take charge,
fast, even if it seems silly most of the time.
In Re. emergencies and board members: If you need to get in touch with a
a board member fast, ribbons aren't really enough. The best solution is to
have designated duty committee members on call via wally-phones
[walkie-talkies with one earphone]. Ok, so you get the odd committee
breaking off a conversation to stare mindlessly into space and start muttering
in a low voice, apparently to nobody, but then, doesn't that happen anyway? :-)
Re. Ribbons: providing ribbons are associated with a real job, they're a
cheap and fun reward for people who've contributed to the success of the
conference. I don't think they're particularly cliquey- if you want more
ribbons, all you have to is volunteer/submit a paper/get elected.
The first time I saw ribbons in use at a con was at Confiction, the World
Science Fiction convention in Holland in 1990; they were so popular that
they've now become a standard at UK cons.As John Sununu said "Perks, what
perks?"
Simon
--
-----
Guest account at TMC | Just Another WAIS Hacker | DOD# 0612
Leonard H. Tower Jr.
...
| Len Tower (who is usually the only person from the FSF to attend Usenix)
|told me he always wears a red sweater wrapped around his waist.
red chamois SHIRT. (Noah: when have you ever seen me with a sweater? ;-)
And then there is the towel on my shoulder...
enjoy -len
Rich Salz
>A modest proposal... Any BOF organizer has to state how many slides
>(overheads, whatever) he is going to present. If the number exceeds three,
>he's a marketeer and he gets a daytime slot.
It's definitely a presentation, not really a BOF, but also clearly not
marketing.
>Of course, this means Rob Kolstad probably ends up giving the BSDI talk
>during the day
That would be a good thing; the BSDI talk was the most blatantly commercial
BOF presentation I attended.
/r$
Leonard H. Tower Jr.
|the ribbons are sposed to help you find people. they have also become
|a bit of humor. i dunno if this adds, detracts or is neutral with resp. to
|the "inner circle". this time newcomer ribbons were used, in the past
|old-timer buttons were used. all are items intended to facilitate the
|old yaccity yacc.
|
|best of [breed|show], lame duck, hanger-on et.al. really are intended to
|be funny and perhaps take the edge off the multiply be-ribboned.
|
|just remember, unix is joke, and we want you to laugh with us twice a
|year. perhaps someone should simply bring a generous supply of extras
|and hand them out to the ribbon challenged.
Maybe instead of (or in addition to) selling conference t-shirts,
USENIX should set up a booth to sell customized ribbons? They should
be a different color then official ribbons, but the same size and cut.
Or many different colors (including a selection of colors and patterns
selected by Larry Wall).
And the "inner circle" should have to buy their "fun" ribbons like
everyone else.
At a few dollars a pop, it might even make a surplus for USENIX.
Or maybe someone else will take it on.
enjoy -len
Don Libes
>the BSDI talk was the most blatantly commercial BOF presentation I attended.
A couple comments:
1) While it was undeniably commercial, Rob was obviously trying to
parody the Ginsu-knife style commercial - as such, there was
sufficient entertainment value to keep me happy. Realistically, the
BSDInc project is more like an experiment which I suspect is of keen
interest to the type of person who attends USENIX. (Of course, Dan
Quayle's presence as an experiment with significant entertainment
value may not be equally justified.)
As an aside, I wasn't commercially offended by any presentations at
this year's USENIX, so perhaps I'm not a good judge of this issue.
2) I really don't see any point in distinguishing between day/night
sessions. I anticipate staying at talks/BOFs until 11pm each night.
I almost never do any sightseeing while the conference is in progress
because there is so much good stuff happening and the opportunity to
network is short. (Nonetheless, San Antonio was surprisingly nice and
I'd love to go back. My second favorite was Portland.)
3) I've resigned myself to wearing cold-weather clothing in the
hotels. I don't understand how you stay comfortable in shorts for an
hour and a half (listening to a technical session) without any
physical exertion.
It seems a shame to a visit a place with such a lovely warm climate,
and have to bundle up to withstand the hotel. I used to complain
about this, but I've given up. My latest hypothesis is that the hotel
staff is very comfortable in a cold room - after all, they rush around
serving food and cleaning rooms. So the hotel staff sets it cold. If
they turn the temp. up so that we're all comfortable, then the hotel
staff won't be.
Like I said, I don't expect action on this, but perhaps people should
be warned. Every year I hear people complaining about shivering in
their shorts. I no longer even both packing shorts for USENIX.
4) The paper submission cycle controversy is a red herring. There are
dozens of deadlines every month for other conferences and journals.
If a USENIX deadline is close by, we submit to it, otherwise we submit
elsewhere. In that sense, I encourage USENIX to continue providing
many publication possibilities. Note that, as Rich Stevens said,
while USENIX papers are very good, journals come first. So often,
papers that we send to USENIX are late, not because the conference
deadlines are late, but because it's languished a year in some
journal's in-box only to be rejected.
5) Craig Partridge writes:
>Just deciding to beef up the technical program isn't enough -- you
>have to decide what core theme the technical program presents. For
>example, is the USENIX program to be "top papers on the UNIX operating
>system, plus key utilities [shells, window packages, networking,
>compilers]" or "top papers on anything related to UNIX, however
>vaguely?"
I don't need to wait for a "Retrospective" theme before I go hear a
retrospective talk. I've never understood this desire to have a
theme. I suppose it gives the person making the brochure something
to tout, but it sure doesn't change the way I submit a paper, nor does
it change whether I (or my management) decide whether we can attend.
("Challenge of Innovation"? - well that seems sufficiently vague to
cover just about everything.) Do we really get enough technically
good papers that the program committee even thinks about the theme
when choosing papers?
It would be helpful to get some feedback from the past couple program
committees on this.
6) The armadillo was cute, but no thanks. For a permanent USENIX
mascot, I'll stick with the BSD daemon.
7) Putting email addresses on badges is ok, I guess, but is no
substitute for the facesaver. I was sorry to lose the faces from the
list of attendees, and now the proceedings. If you calculate the
statistics, you'll find the highest attendence correlates with the
best face-service.
Don Libes li...@cme.nist.gov ...!uunet!cme-durer!libes
Fuat C. Baran
>7) Putting email addresses on badges is ok, I guess, but is no
>substitute for the facesaver.
When I'm looking at a badge I can usually glance up and see the face.
Seeing an email address on the badge would be very useful for those of
us that recognize net acquintances by address.
Putting the email address (which we already supply on the registration
form) should not really cost a whole lot of money, whereas the face
saver project costs real money.
> I was sorry to lose the faces from the
>list of attendees, and now the proceedings.
Me too. I liked it, but I can understand why it was eliminated.
Perhaps it will come back sometime in the future.
> If you calculate the
>statistics, you'll find the highest attendence correlates with the
>best face-service.
Huh? Are you saying that the the quality of the face-service is a
result of the attendance figures? (I suppose if you equate quality
with quantity...) Or were you suggesting that the high attendance is
a result of the face service quality? I seriously doubt that too.
The current economic environment, which it can be argued is causing
the lower attendance, may also be a factor in why we used to have
facesaver in the past and no longer do (i.e. Usenix can't afford to
run it).
--Fuat
--
Advanced Network & Services, Inc. fu...@ans.net
100 Clearbrook Road 914-789-5328
Elmsford, NY 10523 914-789-5310 (Fax)
Brian Kantor
>hotels. I don't understand how you stay comfortable in shorts for an
>hour and a half (listening to a technical session) without any
>physical exertion.
Simple, dude. You *CAN* bundle up. I can't take any more clothes off
than I already have without running into legal problems. This is,
regrettably, USENIX, not NUDENIX. At least, not in the presentations.
- Brian
Berry Kercheval
Rich> That would be a good thing; the BSDI talk was the most blatantly commercial
Rich> BOF presentation I attended.
I've seen some that were even worse. Apple once had one that was
really awful. Even the Apple people I was with walked out...
--berry
--
Berry Kercheval :: be...@pei.com :: Protocol Engines, Inc. Mt. View, CA
"And there's more where that came from!"
Win Bent
>One the subject of ribbons: One person's (large) ribbon collection is
>another person's clique.
Gee, that should be "One (large) person's ribbon collection..."
Were Ed Gould a few pounds lighter, he'd have blown away in the
pre-reception storm, carried off by the 20 yards of material
hanging from his badge.
:-)
--
Wilson H. Bent, Jr. w...@hoh-2.ATT.COM
AT&T - Bell Laboratories (908) 888-7129
Disclaimer: My company has not authorized me to issue a disclaimer.
Steven Bellovin
> Putting the email address (which we already supply on the registration
> form) should not really cost a whole lot of money, whereas the face
> saver project costs real money.
Let strongly second the request for email addresses on the badges.
It's hardly a new idea -- some of us were writing them in more than 10 years
ago. Many of us know each other mostly by the net, and logins are of
more use there than names or faces...
dom...@ox.ac.uk
> Maybe instead of (or in addition to) selling conference t-shirts,
> USENIX should set up a booth to sell customized ribbons? They should
> be a different color then official ribbons, but the same size and cut.
> Or many different colors (including a selection of colors and patterns
> selected by Larry Wall).
Well, you can get a neat-o numerically-controlled sewing machine which
embroiders Macintosh color quickdraw images... Cook up your own ribbon
on the screen, select print, and stand back! Dunno whether it does
PostScript or not, though...
--
Dominic Dunlop
Henry Mensch
->> Really though, after attending the conference from
->> 9 to 5, who truly wants to go back at 7, 9 or 11?
->I do. It's also a good way to seque to the bar or other socializing.
what he said. usenix has enough going on that i make choices ... i
don't sit through the 9-2-5 parts if some of them aren't useful to me,
and i do choose to attend BOFs in the early evening if they are. it's
possible that i might miss an entire morning or afternoon of talks if
those topics aren't interesting to for me. the result is that i'm not
stuck sitting through stuff i'm not interested in, and i have more time
to schmooze (and let's face it, folks: usenix is a great place to
schmooze).
--
# henry mensch / booz, allen & hamilton, inc. / <he...@ads.com>
Kent Landfield
>In article <1992Jun24.2...@ans.net>, fu...@news.ans.net (Fuat C. Baran) writes:
>> Putting the email address (which we already supply on the registration
>> form) should not really cost a whole lot of money, whereas the face
>> saver project costs real money.
So why do we expect USENIX to pick up all the cost for Face Saver ? I
never have understood why USENIX didn't offer a discount coupon in the
registration packet and the attendees who wanted to could pay a few dollars
more to get their face "saved". I have used past face savers for book
labels. They are great. There has been times when I would have liked to
get more than one sheet and would have been willing to pay for that as well.
Face Saver is (was?) a great service and alot of fun. Lets bring it back
real soon.
>Let strongly second the request for email addresses on the badges.
>It's hardly a new idea -- some of us were writing them in more than 10 years
>ago. Many of us know each other mostly by the net, and logins are of
>more use there than names or faces...
There you go. We can revert to doing it the original way next time. (Since
my handwriting is sooooo bad, I'll have to ask someone else to write mine
in...) ;-)
-Kent+
---
Kent Landfield INTERNET: ke...@IMD.Sterling.COM
Sterling Software, IMD UUCP: uunet!sparky!kent
Phone: (402) 291-8300 FAX: (402) 291-4362
Please send comp.sources.misc-related mail to ke...@uunet.uu.net.
Ellie Young
> > Putting the email address (which we already supply on the registration
> > form) should not really cost a whole lot of money, whereas the face
> > saver project costs real money.
>
> It's hardly a new idea -- some of us were writing them in more than 10 years
> ago. Many of us know each other mostly by the net, and logins are of
> more use there than names or faces...
>
The Association will most definitely look into having e-mail
addresses printed on the attendee badges. Thanks to all for
this suggestion.
Ellie Young
Executive Director
USENIX