[Board] [Foss4g2013] [OSGeo-Conf] FOSS4G rotation

Thu Apr 18 11:52:40 PDT 2013

On 13-04-18 2:01 PM, Steven Feldman wrote:
>
> It is difficult to be precise about cannibalisation by regional events
> unless we survey the delegates at these two regional events and ask them
> whether they plan to also attend Nottingham and if not whether they
> would have considered Nottingham if there had not been a regional event.
> Is that worth doing?

Peter Batty wrote an interesting analysis based on stats and surveys 
from previous events about the potential impact of FOSS4G-NA on Beijing.

http://osgeo-org.1560.x6.nabble.com/Re-OSGeo-Conf-North-American-FOSS4G-conference-tp3714127p3714128.html

I also included a copy of the text below because it seems that Peter's 
original post did not get properly archived on OSGeo's servers (that's 
why the link above points to nabble and not osgeo.org).

Daniel

>  Oct 07, 2011; 4:53pm Peter Batty Peter Batty Re: RE: [OSGeo-Conf] North American FOSS4G conference
> Hi all,
>
> Obviously the question of a North American FOSS4G / OSGeo conference has pros and cons. There are some who are concerned that it would have a big negative impact on the global conference. Based on a pretty detailed look at data from previous conferences, I personally think these concerns are overstated. In this email I just wanted to share that data. I include some of my observations, but please draw your own conclusions (well I'm sure you all will anyway :) !!).
>
> There are already multiple strong local conferences, including FOSS4G Japan and FOSSGIS in Germany. According to the 2012 FOSS4G RFP (at http://svn.osgeo.org/osgeo/foss4g/rfp/2012/osgeo-conference-2012-request-for-proposal.pdf), attendance at FOSSGIS has been as follows:
>
> 2011: >400
> 2010: no data provided
> 2009: ~500
> 2008: >400
> 2007: ~250
> 2006: nothing - I'm guessing it started in 2007?
>
> Note that in 2009, the global FOSS4G was in Sydney, and had 436 attendees, so FOSSGIS was larger than the global event that year. I haven't heard anyone requesting that OSGeo should somehow try to prevent FOSSGIS from happening as it is damaging the global FOSS4G (and I don't think there's any way OSGeo could or should do that anyway).
>
> Let's look at how many people from Germany have attended FOSS4G, to see if having FOSSGIS there deters Germans from attending. This data comes from Cameron's spreadsheet at http://bit.ly/p5LZuJ (I will add 2011 data when I can - using my own records for this email). For comparison purposes I also include percentages from France, Italy and UK, as countries that you might expect to be somewhat comparable but that don't have a strong local conference established (so far as I know ... is there one in Italy? ... nothing mentioned in the RFP document though).
>
> Percentage of attendees from Germany, France, Italy, UK
> 2011 (Denver): 1.6, 1.8, 1.9, 1.7
> 2010 (Barcelona): 7.6, 4.9, 6.1, 3.4,
> 2009 (Sydney): 3.0, 2.1, 2.8, 1.4 (note strong attendance despite FOSSGIS being larger than Sydney event that year)
> 2008 (Cape Town): 2.0, 3.4, 3.4, 1.7
> 2007 (Victoria): 1.5, 2.5, 2.9, 2.1
> 2006 (Lausanne): 8.8, 7.6, 9.5, 3.5
>
> Looking at these numbers, German attendance at FOSS4G does not seem to have been negatively affected by FOSSGIS compared to other countries that one might expect to be broadly comparable that don't have a large local conference (to the best of my knowledge).
>
> Now let's look at US and Canadian attendance (percentage) at previous FOSS4G events:
>
> 2011 (Denver): 68.6, 6.3 (=74.9 total)
> 2010 (Barcelona): 8.1, 3.4 (11.5)
> 2009 (Sydney): 6.4, 5.5 (11.9)
> 2008 (Cape Town): 5.2, 2.8 (9.0)
> 2007 (Victoria): 36.7, 35.0 (71.7)
> 2006 (Lausanne): 9.2, 6.5 (15.7)
>
> So for non-North American FOSS4G events, the average US+Canadian attendance has been 12% (or 10.8% based on the most recent 3). Even if North American attendance took a major hit as a result of a North American conference (which the German experience doesn't support), say it drops by 50%, you would only be looking at a 6% drop in total numbers. Maybe some will argue that other people won't come to the global event if not as many North Americans are coming, but I think that's a pretty weak argument. I still think you will have a strong core of key people from the main projects who will go to the global event regardless of other regional events (both from North America and around the world).
>
> Another interesting stat from the 2011 conference is that a huge 67% were attending their first FOSS4G. About 16% had attended 3 or more FOSS4Gs (so around 150 people), and I think it's fair to regard them as a "FOSS4G hard core", most of whom who are likely to attend the global conference wherever it is. That number obviously includes a lot of the key people on projects who help give FOSS4G continuity from one year to the next, and who are also people who help draw in other attendees who would like to meet them, hear them speak, etc.
>
> Lastly, on the survey from 2011 we asked how likely people were to attend Beijing. 71.8% answered 1 or 2 (where 1 is definitely not) and 14.4% (46 people) answered 4 or 5. Interestingly that 14.4% is very close to the "hard core" number I just mentioned who have been to 3 or more FOSS4Gs. If you extrapolate that to all attendees (which may or may not be valid), that suggests that maybe around 130 people who attended in 2011 are reasonably likely to attend Beijing. The Beijing organizers are budgeting on an attendance of 500. Based on all previous events, we should expect between 60 and 75% of attendees in Beijing to be from Asia, most of whom will be attending their first FOSS4G. So looking at it from multiple angles, both the survey and historical data suggest a similar number for people from outside the host continent.
>
> I looked at which countries those 46 people were from who answered 4 or 5 (likely to attend Beijing), and 7 were from the US and 6 from Canada, so that's 28% of the people who responded that way. That suggests a likely attendance of 36 people from North America in Beijing, even if there is no North American event. My feeling is that most of those 36 would go regardless of whether there is a North American event or not. Even if half of them chose not to go because of a North American event, I don't think the loss of 18 North Americans, out of a total of 500 people, should cause anyone to panic about the nature of the global event. And I think the attrition would probably be a lot less than that.
>
> So anyway, I think I've gone on long enough. The more I look at the data from previous conferences, the more I think that having a North American event would have little to no impact on a global event. But I encourage you to look at the data and decide for yourself.
>
> I also think that having a North American event more often than every 3-4 years would be very beneficial in growing the use of open source geospatial here, and I could go on about that a lot more, but won't in this email :).
>
> Cheers,
>     Peter.
>