Updated with complete Sydney figures. One of the toughest aspects of encouraging and enhancing public participation is figuring out who exactly participates with ICANN and who wants to - or would want to.

Because the community is by its nature so diverse, it is difficult to get a handle on who is involved. There is of course plenty of anecdotal evidence but that comes with obvious flaws and is not sufficient to build an organizational plan around.

And so for Mexico City and for Sydney we added some questions to the registration form in order to gather some information on who is attending and what their level and degree of participation is. The stats are below in graphic form, along with an explanation and quick analysis.

So, we asked (among other questions):

* How many ICANN meetings have you attended? (0, 1-3, 4-6, 7-10, or 10+), and

* How many ICANN meetings do you go to each year? (0, 1, 2, 3)

The questions were only added to the Mexico City registration form a few weeks out from the meeting, so the data covers just over 500 attendees - roughly half those that attended in person. And the Sydney data is presented below in both a preliminary and final form, with the analysis created after just 200 responses and then updated after the meeting.

Below are four graphs - one shows the actual figures for each question, and the second the percentage of the total.

Meetings attended in total

So this shows something interesting: the biggest percentage of attendees have been to very few meetings.

The Sydney meeting shows far fewer people that have never been to an ICANN meeting - but that's to be expected as we are still two months out from a meeting and so the first-timers will probably not even heard about the meeting yet.

The updated graph with full Sydney stats below continues to demonstrate the same thing: that a very large percentage - over a third - have never been to a meeting before.

What does this mean? Are ICANN's efforts over the past two years at increasing awareness of these meetings paying off? We'd like to think so. But at the same time, it is possible that alot of the people that do attend end up coming to one or two more meetings and then stop coming. It is very difficult to know without stats running over several years.

What is clear is that we need to put in more effort in catering for those that have been to very few or no ICANN meetings previously.

Of course, just as significant is that a large proportion of those coming to meetings are clearly very regular attendees. Considering that ICANN has had 34 meetings in total over 10 years and right across the world, for people to have attended 10 or more meetings demonstrates that there is a hard core of attendees.

We already knew this - and these are the people that form the community aspect of ICANN - the same people having dedicated long periods of their time to helping ICANN work as a decision-making body.

Let's look at the stats as percentages of the total:

This demonstrates that basically we are looking at the same profile for both Mexico City and Sydney (although this will of course change over time).

The very regular attendees have registered early for Sydney - giving a slight increase - and the first-timers haven't registered yet - because they won't have heard of the meeting yet.

But our expectation would be that once Sydney registration closes, the two meetings' data are effectively the same.

And that is exactly what happened when we ran the full data - an almost exact fit for both the Mexico City and Sydney meeting. Clearly this is significant because it shows that the same percentage of people that had never attended a meeting in Mexico had also never attended a meeting in Sydney i.e. a whole new set of people.

Perhaps worryingly, if we were being more effective at involving people, we would see a shift in the graph with a larger percentage of people having attended 1 to 3 meeting i.e. we were bringing people along to the next meeting.

This might be a good set of data to measure our effectiveness at increasing and encouraging participation.

Again, this is but a snapshot in time but it clearly demonstrates the same points as mentioned above.

* There is a very significant number of new attendees at meetings

* There is a hard core of ICANN attendees

Meetings attended per year

We also asked how many of the three annual public meetings people attended. Above is the graph of actual figures. Just below is the graph from full Sydney attendee data.

And below is a graph taking the same figures represented as percentages of the total.

What do these graphs show?

Well they would appear to back up the idea that the past two years or so have seen a lot of new people coming to ICANN - and that they are coming to meetings fairly regularly. The reasoning behind that assumption: the higher percentage of people coming to three out of three meetings a year than those that have been to 10+ meetings.

It is of course possible that a significant number of those that have been to few meetings think have the intention to attend three meetings a year in future and so they choose that option when registering (we can dig into the data to test that theory out).

But what it does show is that, again, there is a determined core of people that come to every ICANN meeting - which is quite surprising when you consider there are three of them a year, when each one takes a week, and when they are, by design, dotted all over the world.

If you look at the percentage diagram below:

It is clear that again we are likely to end up with the exact same pattern for both Mexico City and Sydney once the first-timers have registered. Again, since these are back-to-back meetings, this is hardly surprising. In fact, it would be more surprising if there was much of a variation.

And with the full data, that is exactly what we found - an almost exact fit. This is less surprisingly than the fit of meetings actually attended, but even so it is significant. It means we have a clean snapshot of what we are faced with at this moment in time with respect to participation in ICANN meetings.

What we should be seeing though, if we are doing our jobs effectively, is these graphs changing over time - more people coming to meetings more regularly. We will keep running these figures over time and use them to determine efforts we put in to encourage continued participation.

But overall we see the same overall pattern in attendees and participation:

* A large percentage of new attendees

* A core community group that follow ICANN closely

It is encouraging that such a larger percentage of attendees put down three meetings a year - it implies a certain degree of loyalty and an intention to keep attending ICANN meetings.

It does of course also suggest there is a general sense that you need to attend all ICANN meetings - we may need to look at the possibilities and impact of effective remote participation.

And we may need to ask more questions about whether attending three meetings a year is viewed as a positive thing or a drain on time and resources.

Of course it is also important to recognise the other answers given. While it is remarkable that such a high percentage of people say they attend three meetings year, at the same time there is still a very significant number that attend one or two a year - over a third.

In fact if you want to do a very broad breakdown of the level of participation at ICANN meetings, the graphs above clearly demonstrate a pretty clear and equal breakdown of groups:

* One third of attendees are relatively new

* One third of attendees are fairly regular

* One third of attendees are very regular

It is impossible to view trends with such a small sample but we will see if we can load up the registration data we have from other meetings and try to pull out comparative participation information to see if more can be learnt.

What we know is what we already knew - that the ICANN community is broad and it is diverse.