18 Mar
2013
18 Mar
'13
8:41 p.m.
I've run a more detailed cost comparison for the Portland bid,
projecting for 200 attendees and 350 attendees separately, so it's
easier to see the base costs, and also incorporating some changes from
the discussion feedback. See the attached ODS file. (The cost per
attendee seems radically lower than previous years, but these are base
costs, and don't include incidentals like printing signs/badges and
t-shirts, video equipment shipping, or optional costs like travel
sponsorships.)
The "Comparison" tab is using the 4-day figures from the "Rooms" and
"Food" tabs. This is based on a projection of sessions Mon-Tue and
Thu-Fri, with the Day Trip on Wednesday. But, in previous years DebConf
has run sessions Mon-Tue and Thu-Sat, that is, 5 total days of sessions.
So, the 5-day projections are included on the "Rooms" and "Food"
tabs.
The actual food reservations will be made much closer to the event,
based on counts of registered attendees, so these are just broad estimates.
In general, if confidence is high that DebConf14 will have attendance of
at least 250-300, then we recommend including the larger Ballroom in the
initial venue reservation. These projections are using a full-day price
for the large Ballroom, rather than specific hours for opening/closing
plenaries, so the actual venue rental total is likely to be a bit
cheaper. (The Ballroom is preferable to Hoffman Hall, because it's in
the same building as the smaller session rooms and hacklabs.)
We're figuring the decision about whether to run DebCamp will be made
later, after sponsorships start to come in. We've talked a bit about
alternatives that could provide hacking time at a lower cost than the
traditional two-week schedule (all with DayTrip on Wed):
* Two full days of DebCamp on Sunday before and Saturday after DebConf
* Combine DebCamp/DebConf to run simultaneously Saturday-Saturday
* DebCamp on the two Saturdays, and DebCamp/DebConf combo on Sun-Fri
At the moment, we're guessing that DebConf14 won't have a Debian Day,
but again, figure the wider DebConf team will make that decision
partially based on sponsorships.
HTH,
Allison
18 Mar
18 Mar
8:54 p.m.
On Mon, Mar 18, 2013 at 1:41 PM, Allison Randal <allison(a)lohutok.net> wrote:
I've run a more detailed cost comparison for the
Portland bid,
projecting for 200 attendees and 350 attendees separately, so it's
easier to see the base costs, and also incorporating some changes from
the discussion feedback. See the attached ODS file. (The cost per
attendee seems radically lower than previous years, but these are base
costs, and don't include incidentals like printing signs/badges and
t-shirts, video equipment shipping, or optional costs like travel
sponsorships.)
The "Comparison" tab is using the 4-day figures from the "Rooms" and
"Food" tabs. This is based on a projection of sessions Mon-Tue and
Thu-Fri, with the Day Trip on Wednesday. But, in previous years DebConf
has run sessions Mon-Tue and Thu-Sat, that is, 5 total days of sessions.
So, the 5-day projections are included on the "Rooms" and "Food"
tabs.
The actual food reservations will be made much closer to the event,
based on counts of registered attendees, so these are just broad estimates.
In general, if confidence is high that DebConf14 will have attendance of
at least 250-300, then we recommend including the larger Ballroom in the
initial venue reservation. These projections are using a full-day price
for the large Ballroom, rather than specific hours for opening/closing
plenaries, so the actual venue rental total is likely to be a bit
cheaper. (The Ballroom is preferable to Hoffman Hall, because it's in
the same building as the smaller session rooms and hacklabs.)
We're figuring the decision about whether to run DebCamp will be made
later, after sponsorships start to come in.
I think the opportunity to get great sponsorships here in Portland is
going to be great.
I have already had some initial discussions with at least on local
company and hope to
put my local contacts to good use ;)
We've talked a bit about
alternatives that could provide hacking time at a lower cost than the
traditional two-week schedule (all with DayTrip on Wed):
* Two full days of DebCamp on Sunday before and Saturday after DebConf
* Combine DebCamp/DebConf to run simultaneously Saturday-Saturday
* DebCamp on the two Saturdays, and DebCamp/DebConf combo on Sun-Fri
At the moment, we're guessing that DebConf14 won't have a Debian Day,
but again, figure the wider DebConf team will make that decision
partially based on sponsorships.
HTH,
Allison
_______________________________________________
Debian-portland-soc mailing list
Debian-portland-soc(a)pdx.debian.net
http://pdx.debian.net/mailman/listinfo/debian-portland-soc
--
Benjamin Kerensa
http://benjaminkerensa.com
"I am what I am because of who we all are" - Ubuntu
10:12 p.m.
New subject: [Debconf-team] [DC14] Portland team: historical cost comparison
On Mon, Mar 18, 2013 at 01:41:50PM -0700, Allison Randal wrote:
I've run a more detailed cost comparison for the
Portland bid,
projecting for 200 attendees and 350 attendees separately, so it's
easier to see the base costs, and also incorporating some changes from
the discussion feedback. See the attached ODS file. (The cost per
attendee seems radically lower than previous years, but these are base
costs, and don't include incidentals like printing signs/badges and
t-shirts, video equipment shipping, or optional costs like travel
sponsorships.)
Thanks, Allison!
FWIW, looking this over I realize there was a mistake in the bid page: the
$66/night listed for The Broadway was per-room, not per-person. I've fixed
this now in the wiki page.
I'm not sure how the totals are arrived on the Comparison page; the numbers
look low to me even without including the venue and other non-per-person
costs. Can you elaborate on the math here?
For 200 people @ $20/day for food, $16,000 comes out to 4 days. But that
covers only the days talks are running, and we probably need to figure one
additional day of food to cover the arrival/departure days, as well as
breakfast on the day trip day and some costs (TBD) for food on the day trip.
The room total I can't reverse-engineer at all. $9840/$28 -> 351.43
person-nights. While not everyone who attends DebConf has sponsored
accomodation costs, 351.43 person-nights @ 6 nights -> 58.57 people. So I'm
not sure what's going on there. :-)
And for consistency with https://wiki.debconf.org/wiki/PriceComparison, the
"total spent" should at least include the known venue costs.
--
Steve Langasek Give me a lever long enough and a Free OS
Debian Developer to set it on, and I can move the world.
Ubuntu Developer http://www.debian.org/
slangasek(a)ubuntu.com vorlon(a)debian.org
10:45 p.m.
New subject: [Debconf-team] [DC14] Portland team: historical cost comparison
On 03/18/2013 03:12 PM, Steve Langasek wrote:
FWIW, looking this over I realize there was a mistake in the bid page: the
$66/night listed for The Broadway was per-room, not per-person. I've fixed
this now in the wiki page.
Ah, okay, updated in the spreadsheet too.
I'm not sure how the totals are arrived on the
Comparison page; the numbers
look low to me even without including the venue and other non-per-person
costs. Can you elaborate on the math here?
Venue rental for 4 days plus food for 4 days, divided by total attendees.
For 200 people @ $20/day for food, $16,000 comes out
to 4 days. But that
covers only the days talks are running, and we probably need to figure one
additional day of food to cover the arrival/departure days, as well as
breakfast on the day trip day and some costs (TBD) for food on the day trip.
Yes, that's the 4-day estimate from the "Food" tab. I'll change to a
7-day estimate instead, just to be on the safe side. Though, with so
many good cheap food options nearby, will people want to be eating at
the cafeteria for breakfast, lunch, and dinner every day? I guess they
might if it's free.
I'll add a separate tab for the Day Trip when we have estimates for that.
The room total I can't reverse-engineer at all.
$9840/$28 -> 351.43
person-nights. While not everyone who attends DebConf has sponsored
accomodation costs, 351.43 person-nights @ 6 nights -> 58.57 people. So I'm
not sure what's going on there. :-)
Those are venue room totals from the "Rooms" tab, not lodging room
totals. I've changed the spreadsheet to consistently use "venue" for
session/plenary rooms to be clearer. :)
And for consistency with
https://wiki.debconf.org/wiki/PriceComparison, the
"total spent" should at least include the known venue costs.
Yes. Do we need to include lodging rooms in the total base price? That
is, are we paying any fixed up-front amount for lodging, or is lodging
paid by attendees as they register (or once we decide if it's possible
to sponsor travel/lodging)?
I had already tweaked the projections to reflect venue reservation of
the large Ballroom for opening/closing plenaries only, so that change is
also in the attached updated spreadsheet. (The full-day price for the
Ballroom is still in the "Venue" tab for comparison.)
Allison
19 Mar
19 Mar
12:49 a.m.
New subject: [Debconf-team] [DC14] Portland team: historical cost comparison
On Mon, Mar 18, 2013 at 03:45:07PM -0700, Allison Randal wrote:
> I'm not sure how the totals are arrived on
the Comparison page; the numbers
> look low to me even without including the venue and other non-per-person
> costs. Can you elaborate on the math here?
Venue rental for 4 days plus food for 4 days, divided
by total attendees.
Ah!
> For 200 people @ $20/day for food, $16,000 comes
out to 4 days. But that
> covers only the days talks are running, and we probably need to figure one
> additional day of food to cover the arrival/departure days, as well as
> breakfast on the day trip day and some costs (TBD) for food on the day trip.
Yes, that's the 4-day estimate from the
"Food" tab. I'll change to a
7-day estimate instead, just to be on the safe side. Though, with so
many good cheap food options nearby, will people want to be eating at
the cafeteria for breakfast, lunch, and dinner every day? I guess they
might if it's free.
Probably not - we can mine the data from past conferences to get a better
model for what percentage of attendees will be eating the cafeteria food at
any given time. (We'll want this anyway for purposes of giving estimated
headcounts to the cafeteria staff.)
I'll add a separate tab for the Day Trip when we
have estimates for that.
Sounds good.
> The room total I can't reverse-engineer at
all. $9840/$28 -> 351.43
> person-nights. While not everyone who attends DebConf has sponsored
> accomodation costs, 351.43 person-nights @ 6 nights -> 58.57 people. So I'm
> not sure what's going on there. :-)
Those are venue room totals from the "Rooms"
tab, not lodging room
totals. I've changed the spreadsheet to consistently use "venue" for
session/plenary rooms to be clearer. :)
> And for consistency with
https://wiki.debconf.org/wiki/PriceComparison, the
> "total spent" should at least include the known venue costs.
Yes. Do we need to include lodging rooms in the total
base price? That
is, are we paying any fixed up-front amount for lodging, or is lodging
paid by attendees as they register (or once we decide if it's possible
to sponsor travel/lodging)?
I don't know that there's any fixed up-front price, but sponsored attendance
at DebConf includes both food and accomodation. So for the big picture of
"how much money does DebConf spend", this is very relevant.
--
Steve Langasek Give me a lever long enough and a Free OS
Debian Developer to set it on, and I can move the world.
Ubuntu Developer http://www.debian.org/
slangasek(a)ubuntu.com vorlon(a)debian.org
1:24 a.m.
New subject: [Debconf-team] [DC14] Portland team: historical cost comparison
On 03/18/2013 05:49 PM, Steve Langasek wrote:
I'll start a tab for projecting sponsored attendance. At the moment, I'm
assuming we'd start at 0 sponsored attendees, and expand as incoming
sponsorship increased beyond costs. Do people have a sense of an "ideal"
number of sponsored attendees? As in "If funding is available, ideally
we'd like to sponsor X number or Y percentage of attendees."
And, I'll start with:
- Accommodation: $28/person/day
- Food: $20/person/day
- Flight: $1000/person
Aiming for an approximate average on flight price, figuring US/Canada
flights could be as low as $300, while international flights could be
$2000 or higher. (Numbers from past years will be helpful in these
projections too.)
Allison
On Mon, Mar 18, 2013 at 03:45:07PM -0700, Allison
Randal wrote:
Probably not - we can mine the data from past conferences to get a better
model for what percentage of attendees will be eating the cafeteria food at
any given time. (We'll want this anyway for purposes of giving estimated
headcounts to the cafeteria staff.)
Good idea.
Do we need to
include lodging rooms in the total base price? That
is, are we paying any fixed up-front amount for lodging, or is lodging
paid by attendees as they register (or once we decide if it's possible
to sponsor travel/lodging)?
I don't know that there's any fixed up-front price, but sponsored attendance
at DebConf includes both food and accomodation. So for the big picture of
"how much money does DebConf spend", this is very relevant. 
4:29 p.m.
New subject: [Debconf-team] [DC14] Portland team: historical cost comparison
Allison Randal dijo [Mon, Mar 18, 2013 at 06:24:11PM -0700]:
I'll start a tab for projecting sponsored attendance. At the moment, I'm
assuming we'd start at 0 sponsored attendees, and expand as incoming
sponsorship increased beyond costs. Do people have a sense of an "ideal"
number of sponsored attendees? As in "If funding is available, ideally
we'd like to sponsor X number or Y percentage of attendees."
And, I'll start with:
- Accommodation: $28/person/day
- Food: $20/person/day
- Flight: $1000/person
Aiming for an approximate average on flight price, figuring US/Canada
flights could be as low as $300, while international flights could be
$2000 or higher. (Numbers from past years will be helpful in these
projections too.)
I expected flight prices to be quite lower. Still, for DebConf10 (I
expect numbers to be closest to what we had at New York), 88 out of
306 attendees (marked as reconfirmed and arrived) were from within the
USA (and the second most popular country was "not selected" ;-) ). I
was surprised to see that, for July, ticket prices from Paris and
Frankfurt are at ~US$1500, quite higher than what I expected.
This following query does *not* mean the requested travel sponsorship
was fulfilled, but requested, and that the person attended, and was
labeled as "the amount needed for you to travel" (need_travel_cost):
acronym | avg
---------+-----------------------
dc7 | 574.9236000000000000
dc8 | 1070.1224489795918367
dc9 | 435.2346428571428571
dc10 | 819.6127500000000000
dc11 | 612.6257142857142857
dc12 | 981.0483870967741935
Do we need to include lodging rooms in the total base
price? That
is, are we paying any fixed up-front amount for lodging, or is lodging
paid by attendees as they register (or once we decide if it's possible
to sponsor travel/lodging)?
I don't know that there's any fixed up-front price, but sponsored attendance
at DebConf includes both food and accomodation. So for the big picture of
"how much money does DebConf spend", this is very relevant.
5:18 p.m.
New subject: [Debconf-team] [DC14] Portland team: historical cost comparison
On 03/19/2013 09:29 AM, Gunnar Wolf wrote:
Allison Randal dijo [Mon, Mar 18, 2013 at 06:24:11PM
-0700]:
Ah, excellent! I'll go with $800 for approximate average flight
sponsorship cost, since New York is geographically closest. It's
possible that we'd get more local applications that lower the average,
or more international applications that raise the average, but it's a
good first approximation.
So, that comes to a sponsorship total of $1,136 per person for
accommodation, food, and travel. Or, approximately every $1,000 of
incoming sponsorship above costs would mean we could sponsor one
additional attendee.
Allison
Aiming for an approximate average on flight price, figuring US/Canada
flights could be as low as $300, while international flights could be
$2000 or higher. (Numbers from past years will be helpful in these
projections too.)
I expected flight prices to be quite lower. Still, for DebConf10 (I
expect numbers to be closest to what we had at New York), 88 out of
306 attendees (marked as reconfirmed and arrived) were from within the
USA (and the second most popular country was "not selected" ;-) ). I
was surprised to see that, for July, ticket prices from Paris and
Frankfurt are at ~US$1500, quite higher than what I expected.
This following query does *not* mean the requested travel sponsorship
was fulfilled, but requested, and that the person attended, and was
labeled as "the amount needed for you to travel" (need_travel_cost):
acronym | avg
---------+-----------------------
dc7 | 574.9236000000000000
dc8 | 1070.1224489795918367
dc9 | 435.2346428571428571
dc10 | 819.6127500000000000
dc11 | 612.6257142857142857
dc12 | 981.0483870967741935
4:13 p.m.
New subject: [Debconf-team] [DC14] Portland team: historical cost comparison
Steve Langasek dijo [Mon, Mar 18, 2013 at 05:49:26PM -0700]:
Right. and for this, I would look at similar cities - e.g. I expect
the Portland food offer (and university cafeteria quality) to be much
closer to New York or Edinburgh more than to most of our other past
venues.
> For 200
people @ $20/day for food, $16,000 comes out to 4 days. But that
> covers only the days talks are running, and we probably need to figure one
> additional day of food to cover the arrival/departure days, as well as
> breakfast on the day trip day and some costs (TBD) for food on the day trip.
Yes, that's the 4-day estimate from the
"Food" tab. I'll change to a
7-day estimate instead, just to be on the safe side. Though, with so
many good cheap food options nearby, will people want to be eating at
the cafeteria for breakfast, lunch, and dinner every day? I guess they
might if it's free.
Probably not - we can mine the data from past conferences to get a better
model for what percentage of attendees will be eating the cafeteria food at
any given time. (We'll want this anyway for purposes of giving estimated
headcounts to the cafeteria staff.)
4:47 p.m.
New subject: [Debconf-team] [DC14] Portland team: historical cost comparison
On 03/19/2013 09:13 AM, Gunnar Wolf wrote:
Right. and for this, I would look at similar cities - e.g. I expect
the Portland food offer (and university cafeteria quality) to be much
closer to New York or Edinburgh more than to most of our other past
venues.
Portland is generally cheaper for the same quality food than New York:
http://www.numbeo.com/cost-of-living/compare_cities.jsp?country1=United+Sta…
So, overall, we can anticipate the food bid for Portland will be about
1/3 lower than New York.
Allison
5:19 p.m.
New subject: [Debconf-team] [DC14] Portland team: historical cost comparison
On Tue, Mar 19, 2013 at 09:47:59AM -0700, Allison Randal wrote:
On 03/19/2013 09:13 AM, Gunnar Wolf wrote:
> Right. and for this, I would look at similar
cities - e.g. I expect
> the Portland food offer (and university cafeteria quality) to be much
> closer to New York or Edinburgh more than to most of our other past
> venues.
Portland is generally cheaper for the same quality
food than New York:
So, overall, we can anticipate the food bid for
Portland will be about
1/3 lower than New York.
Right - the price should be lower, but in terms of availability of outside
food (and therefore the frequency with which people choose to eat outside
the conference), New York and Edinburgh are probably good models for what
we'll find. Banja Luka may as well, since the conference was fairly central
in the city with a number of restaurants around.
--
Steve Langasek Give me a lever long enough and a Free OS
Debian Developer to set it on, and I can move the world.
Ubuntu Developer http://www.debian.org/
slangasek(a)ubuntu.com vorlon(a)debian.org
5:47 p.m.
New subject: [Debconf-team] [DC14] Portland team: historical cost comparison
On 03/19/2013 10:19 AM, Steve Langasek wrote:
Right - the price should be lower, but in terms of availability of outside
food (and therefore the frequency with which people choose to eat outside
the conference), New York and Edinburgh are probably good models for what
we'll find. Banja Luka may as well, since the conference was fairly central
in the city with a number of restaurants around.
Ah, got it. So, in terms of projecting head-counts for the cafeteria,
numbers from those past DebConfs will provide the most accurate
approximation. If, say, 40% of attendees skipped the conference-provided
dinner in those cities, then we can build that same percentage into our
projections with a fair degree of confidence. Sounds good.
Allison
4235
days inactive
4236
days old
debian-portland-soc@pdx.debian.net
11 comments
4 participants
participants (4)
-
Allison Randal -
Benjamin Kerensa -
Gunnar Wolf -
Steve Langasek