My view of the IAOC Meeting Selection Guidelines

[Bob Hinden <[e-mail]>]

Hi,

I have been on the IAOC for about a year and wanted to explain my  
view how the IAOC decides to to have an IETF meeting in a specific  
location.  I thought this might be useful given the discussion about  
IETF72 in Dublin.  This is my personal view, not anything official  
from the IAOC.

First of all and for background, the IETF uses the revenue from the  
IETF meetings to pay for the meetings themselves and for the other  
fixed costs of running the IETF.  Meeting costs include meeting  
rooms, power, power strips, AV, cookies, drinks, continental  
breakfast, network, WLAN, IETF secretariat costs associated with the  
meeting, etc., etc.  The fixed costs of running the IETF include IETF  
secretariat, phone conferences, IAD salary, RFC Editor, IT  
infrastructure to host ietf.org, software tools development (those  
not done by our great volunteer developers!), etc., etc.  The total  
expenses are greater than the total revenue.  The ISOC pays for the  
difference (e.g., the deficit).  The budget information can be found  
at: http://iaoc.ietf.org/.  In round numbers total expenses are about  
$4M, revenue is about $2.5M, and the ISOC contributes about $1.5M.   
If it wasn't for the ISOC we would have a big problem.

I believe the IAOC is trying to keep this stable and not grow the  
deficit beyond what the ISOC is willing to subsidize.

We have two kinds of IETF meetings, hosted meetings and non-hosted  
meetings.  When we have a host (e.g., Alcatel-Lucent for IETF72) they  
pay for a lot of the meeting expenses.  This reduces our expenses and  
helps support the rest of the IETF operation.  Hosts usually cover  
the expenses for the network, circuits, NOC, social, t-shirts, etc.   
Sometimes they pay the IETF to do some of these things or pay for  
them directly.  When we have a non-hosted meeting, we pay for  
everything.  Not surprisingly the IAOC prefers to have hosted  
meetings.   In most cases the host has a preference for where they  
would like to host the meeting.  For example, the host for IETF72  
wanted the meeting to be in Dublin.  As long as the facilities in  
that location are acceptable we will do it at the host's preferred  
location.  Acceptable includes the right ratio of large and small  
meeting rooms, hotel room availability and price, ability to build a  
working WLAN network, availability of transit network connections,  
availability of other hotel rooms and restaurants, etc.  We won't go  
somewhere if we don't think we can have a successful meeting.  That  
means in a location where we think people are willing to travel to,  
get visas, hotel rooms are not $500 per night, etc.  We will be more  
flexible on some of these things to meet the hosts desire for a  
specific location, but not to the point where don't think the meeting  
would be successful.

In the case of Dublin, the IAOC did understand that the sites  
distance to Dublin wasn't ideal, but it was the only site we could  
find in the area that meet the other requirements.  In this case, we  
will try to ameliorate this issue by providing busses to downtown  
Dublin.  In my view we are having a meeting in Dublin because we have  
a host to wants to host the meeting there. I don't think we would  
have done it there if there was not a host.

I hope this helps the IETF community to understand the decision  
process the IAOC uses.  We are trying hard to book meetings as far  
out in the future as possible, maintain a good balance of North  
America, Europe, and Asia, and keep the IETF financially afloat.  If  
we had some other significant source of income besides meeting  
revenue, there would be more flexibility, but until that happens we  
will continue to have a preference for hosted meetings.

Bob


_______________________________________________
Ietf mailing list
[e-mail]
http://www.ietf.org/mailman/listinfo/ietf