On 20 September 2011 15:22, Nick Burch <[email protected]> wrote: > First up, when would we want to hold the event? Largely influencing this: > What's the soonest we could be ready? (We will be constrained by when the > SAP conference venue we prefer is available)
That depends on who is making it happen and what kind of event we are talking about. If this is a traditional ApacheCon style event then I would say a minimum of 9 months. If it is a more community focussed event (as in targetting people already interested in Apache projects rather than those who might become contributors) then it can be done in less time. > Moving onto the bigger part, what sort of conference do we want to put on in > the space? Monday-Friday? Wednesday-Sunday? Monady-Sunday/Saturday-Friday? Speaking entirely personally I do not like being away over weekends. Many of us have family now and weekends are precious to us, especially those of us who travel as part of the standard working week. RANDOM THOUGHT: we could put effort into providing facilities for partners and families (I'm not talking about anything significant things like connect people together, point them at suitable activities, and make socials open to partners). I'm not sure how many people have partners/families that would come along though, so not sure if this is worth the effort. > How much should be conference sessions, how much hackathons, how much > meetups, how much barcamp / ad-hoc sessions? It depends on what we want to achieve. Personally I think we should be identifying some common themes that span multiple projects. The goal should be to bring significant benefit to the employers of people who attend. Note this is different from what I think a producer led ApacheCon should look like (see my reply to your other mail). For me a community event needs to be low cost and highly educational for project committers. A user focussed ApacheCon should be for profit and highly educational for potential users (who may eventually be committers, but that is irrelevant in this context). A community focussed event should bring in key *viosionary* speakers, not necessarily ASF people, on the chosen themes but the main activities should be BarCamp/MeetUp/Hackathon sessions, some pre-arranged, some designed at the event. This would be something like: * 3-5 "scene setting" sessions per day (e.g. presentations such as "the next five years of cloud computing") * 4-8 "innovation sessions" per day (e.g. interactive sessions exploring possible collaborations between projects to address some of the issues raised by the scene setting sessions) * 4-8 "barcamp sessions" per day (e.g. impromptu presentations in response to one of the above sessions) * plenty of hacking space to follow up on the innovation ideas Speakers should not, in the main, be from within the ASF. They should be CIO/CTO level individuals, startup founders, product managers etc. They should be chosen because a) they can help point our committers at possible future trends and b) because they are our power users. The "innovation sessions" should be facilitated by people who know how to get people to find potential collaborations that might have otherwise been unrecognised. > How many people will we expect > to come for all, and how many for just one day? Impossible to say without knowing what we are trying to achieve. If we looked at the above type of event I'd say 30-40 per theme. Of those maybe 50% would come for more than one day. > Ratio of talks aimed at > users, developers, and those interested in open source > business/community/the apache way? See above - for me community events should be about adding value to our existing community. Other events should be seeking to bring in new users (see my other mail). That is not to say we couldn't have a user ApacheCon at the SAP offices, it's just that I am trying to make a distinction between user events and community events (I define user as someone who is not yet contributing to the community). > Feeding into this, I know quite a few people go to other community open > source events, and a smaller number help organise them. What works well? > What doesn't work? Who should we be learning from and emulating? Personally I'm really dissatisfied with all of the ones I have been to. The majority of events focus on dissemination. The emphasis on presentation after presentation after presentation is, for me, a throw back to a time when information was hard to come by. Today I can find out anything I want to know about any ASF project by sitting at a keyboard. I go to events because I want to meet people, I want to explore ideas with people and I want to discover *unexpected* things. There are a few that seek to break the mould, BarCamps, for example. However, even at BarCamps people tend to just do presentations. It's hard breaking old habits. However, there are people out there who specialise in running sessions that surface new ideas. They get people to interact and talk. A few months ago I was in just such a session. Myself and a complete stranger found that if we combined our skills we could come up with a really innovative new product. Yesterday I heard that the "stranger" now has seed money to build a prototype. That's what I want to come out of conferences, sitting there in sessions doesn't achieve that. These things don't happen by accident, they are managed. The room in which I met this individual was a self-selecting group who had complimentary skills and interests. As a result there were a great many possible partnerships forged at that event. Selling semi-structured events to bosses can be very hard though. That't why I want to separate the community events from the user events. Users need to learn how to use the projects, they want to be trained and they want measurable benefits. That allows high ticket prices. However, I would argue that committers want to learn what else can they do within their companies. How can they bring extra value to their employer? The risks are higher and it is difficult to measure benefit here, so costs need to be lower. > How much help do we need to offer > for speakers? And can that work with a lower registration fee, and hence > lower total budget? Speakers and facilitators should, if possible, be fully covered. Everyone else should be as low cost as possible. > And anything else I've missed that we'd need to consider? :) Are my rantings about separate committer and user events worth discussing more generally? Ross
