SharePoint Evolution 2010 – The Conference That Almost Wasn’t

We just wrapped up the SharePoint Evolution 2010 conference in London. By all accounts, it was a great event. Attendance was excellent and there was a ton of new content on SharePoint 2010. The Combined Knowledge crew made everything appear smooth and effortless. But things are not always as they seem – this conference was perhaps one of the most difficult conferences to put on that I’ve ever been involved with. Just to give you a taste of what happens in the background when things go terribly wrong, I thought I’d give a behind-the-scenes account of the insane amount of effort that went into this event.

My personal travel journey sets the stage for one of the strangest logistical nightmares ever encountered for a conference of any type. On Wednesday, the 14^th of April, I was schedule to fly to London on the last direct flight of the day out of D/FW airport. About the time they were boarding business class, the gate agent made an announcement that a volcano in Iceland, of all bloody places, was delaying our departure. She wasn’t sure if we were going to be allowed to leave. As you can imagine, everybody just looked at each other as if this was some strange sort of joke. I mean, really, a volcano in Iceland? Who cares? What possible impact could that have on a flight to England? That’s like, all the way at the top of the world, right? Somebody must be putting us on.

Sadly, as the whole world now knows, it was no joke. Amidst all the grumbling, cursing and general carrying on, the agent came back on in a few minutes and told us we could just make it if we moved now. Fast. As in, get yer butts on the plane and don’t dilly dally around. We still had no idea of what she was going on about but if we could make the trip then fine, let’s get going. I travel nearly 80,000 miles a year in the air, almost all of it on American Airlines. I have never, ever, seen a flight crew move so fast. They had every seat filled and doors closed in less than fifteen minutes. Five minutes later we were on the tarmac. The pilot came on and gave us a run down of the situation. Heathrow airport was predicting they were going to close ALL air traffic by noon the following day. We were due to arrive at 12:25. If he put the afterburners on we just might make it. So get in, sit down, shut up and hang on (that’s not actually what he said but I’m not paraphrasing by much – he was definitely a good ol’ country boy in a city slicker hurry).

Naturally, we all had a good chuckle for the next nine hours. Close Heathrow airport – right, as if that would ever happen. It’s the busiest airport in the whole freakin’ world. Every few hours the pilot would come on and update us, each time assuring us that we were ahead of schedule and were going to make our gate, volcano be damned. "Step on it!" someone yelled at one point. I couldn’t resist – I looked at the guy next to me and said "Boogity, boogity, boogity!" but, since he was British, he didn’t get it (that’s a NASCAR reference, for all of you who don’t follow real racing). As it turns out, we were wheels down at 12:11 and at the gate by 12:20 – ten minutes before they shut the whole place down. Done. No more planes, no way Jose. I’ll spare you the details of what Heathrow was like that day – if you’ve ever been through there, imagine it on its worst day then multiply that by at least ten. It was a complete zoo. All I can say is thank goodness for IRIS or I would have been waiting in customs for days.

So I had snuck in the country just in time. Now what? Well, once we were able to catch up on the news, we all thought things would be back to normal in a day or two. Not so. Thursday became Friday, Friday turned into Saturday, still no flights. Now we had a real problem. Out of all the speakers scheduled to appear at the SP Evolution conference, nearly half of them were from out of the country, and most of those were from the US. That meant we had more than 75 sessions to deliver and were short nearly half our presenters. Uh oh. That’s what you call a serious logistical problem (or a SNAFU, if you were ever in the military). I rang up Steve Smith and offered to jump on the first train to London from our offices in the West Midlands to help him figure out how to handle the mess we found ourselves in. Once I arrived, I found Steve and Chandima, who had just made it in on the Eurostar from Paris after travelling from his home in New Zealand, ensconced in a boardroom we quickly dubbed the "battle room". Over the next five or six hours, we played musical sessions, with Steve scribbling a session title and presenter on the whiteboard, then erasing and moving it around each time an update came in announcing that yet another speaker was stranded. With Outlook up on one computer, Twitter up on the projector, and Steve’s iPhone going off every ten seconds, we were in a barely controlled state of chaos. We would look at the schedule, check our list of cancellations, make frantic calls to see if someone who was scheduled for Tuesday or Wednesday could get there on Monday, write on the whiteboard, erase it, move things around, and repeat the cycle. By 9PM we still didn’t have a full schedule for Monday and it was starting to look grim.

But then a funny thing happened. Calls started to come in from local UK people who only had one session scheduled saying they would be happy to do another. Microsoft folks who didn’t originally plan to participate chimed in to volunteer, some of whom even had presentations already prepared. Community participants who hadn’t done much public speaking offered to take a session, any session, if that would help keep things on track. Stranded speakers started emailing their decks and demos in case somebody else could deliver their session. Tobias Zimmergren, who lives in Malmo, Sweden, emailed to say that he was getting in a car and driving to London, offering to pick up anyone who might be on his route (that’s a 1700km journey and nearly 22 hours of windshield time). Daniel Wessels hopped on a train in München, Germany, facing lord only knows how many connections and sardine-like conditions. Other speakers were reporting similar travel insanity but were bound and determined to get to London come hell or high water. Within an hour, we had the schedule for Monday sorted. And it even looked like Tuesday might come together if everyone who said they could be there actually made it.

On Sunday, speakers started to arrive in force, and the battle room swelled to a dozen people. Steve and several others attacked the schedule for Tuesday, and everyone began working on sessions for the following day. In case you’ve never done one before, it’s safe to say that an hour’s worth of content, including slides and demos, usually takes more than ten hours to prepare, and sometimes considerably longer. Several speakers who were originally scheduled for Tuesday or Wednesday had planned to prep their content during the week but had now been moved to first thing Monday morning. Some picked up sessions with content they were unfamiliar with or had never worked with before (bear in mind this is a new product version with tons of new features – none of us has had a chance to look at everything, much less work with it enough to talk intelligently in more than a couple of areas). I was dumb enough to volunteer for the first developer session on Monday. How do I get talked into these things?

The original plan for the Keynote presentation was to have myself, Steve, Spencer Harbar, Andrew Connell, Bill English, Penny Coventry and Todd Bleeker do a "how we got here" roadmap, starting with SharePoint 2001 and moving through CMS, SharePoint 2003 and SharePoint 2007. With Andrew stuck in Oslo, Norway and Bill and Todd never able to make it out of the US, that presented a bit of a problem, especially as everyone had their own version of old images to work from (how anyone got 2001 up and running at this stage is beyond me). Thankfully, Brett Lonsdale stepped in to take over the 2003 developer piece, Spence covered AC’s bit and we pulled it together. With not a single minute of rehearsal, we managed to get all the machines running on stage at once and the keynote came off without a hitch. That could easily have gone very wrong in front of the whole crowd.

By Tuesday, things were humming along. After way too much fun the night before trying to let off some tension, Andrew Woodward and I managed to get through a ton of content in our unit testing session, working with different versions of R&D code that only a few people on the planet have ever even used. Andrew got a last-minute lifeline from the Microsoftie who actually wrote the core PEX/Moles/Behaviors code in order to get his demo working and, miraculously, I was able to live-code working unit tests with Moles. This sort of thing is kind of like being a test pilot for fighter aircraft; there’s a ton of danger and never more than a 50% chance it’s going to work – you could easily go down in flames (to be honest, I don’t think either one of us had as much as 50% confidence in our demos). Dual presentations are hard to get right without much practice time and, naturally, we didn’t have any, but lucky for us Woody and I have bantered about this stuff for at least a year and were able to get through it without looking like complete fools (or so I believe). I got in a couple of digs about TDD and TypeMock, Woody got me on a few points regarding proper isolation and the whole "fail first" thing, and we generally had a good time. My only regret is that we didn’t play the video of Woody snoring at last year’s MVP Summit (sorry mate, had to get that in J).

Apparently, at some point Monday night, Steve came to the conclusion that it would be a great idea for Ben Robb and myself to do a session on the Client Object Model Tuesday afternoon. I thought he was just poking fun at us and forgot it ever happened. Unfortunately, he was serious and posted it to the schedule without either of us knowing about it. Until Tuesday morning, that is, right before we both went on stage at 10AM. Back in the speaker room after our sessions, we had a genuine "Holy Crap!" moment. We had no slides, no demos, and four hours to put a session together. And we were both dead tired and mind numb. With a lifeline from Spence for slides and the good fortune of having a bit of client OM code hanging around, not to mention a clutch save by Woody on the javascript piece of the stack, we managed to cobble a full session and demos together in less than two hours. It wasn’t pretty but it worked. The funny bit was that neither Ben nor myself had actually written live Client OM code before the conference. All I knew about it was what I’d seen in demos and discussed with other MVP’s and the product team. The next two hours I spent researching everything I could so I didn’t say anything deliberately stupid on stage (randomly stupid I’m OK with – that’s part of my normal session content). Again, with no practice time, we hit the stage, wrote our code live on the spot, and it all compiled and worked. That was about as close to a minor miracle as you can get.

By Wednesday, we could all see the finish line. I don’t know how many sessions Spence ended up doing (I think six was the final count) but every time I turned around he was on stage somewhere. Steve managed to do an hour of live install and configuration without so much as a script to work from. Several people presented someone else’s deck without having any clue what was in it. I don’t think anybody did less than twice the number of sessions they originally had planned and several who weren’t even on the original schedule delivered more than one. By the time we got to the Ask The Experts panels at the end, we were all completely knackered. One last meal and a few drinks with the folks who stayed in town and SP Evolution 2010 was done and dusted.

It’s often said that when times are tough you find out who your friends are. It’s a credit to Steve and his crew at Combined Knowledge that they have a lot of really good friends. When the SharePoint community wants to make something happen, we make it happen, period. There was just no way we were going to let this thing go off the rails. We managed to have content in every single slot of five tracks for three full days without missing a single one. I can’t say enough about the dedication and perseverance of everyone who stepped up to the plate. Zoe Watson, rock star that she is and the real reason we ever had any chance of getting through it, deserves so many accolades I don’t even know where to begin. And the attendees, many of whom had their own challenges, deserve a lot of credit of making it to the show and putting up with a constantly changing schedule. I’m sure some were disappointed that they didn’t get to see a particular speaker or session that they had been hoping for but everyone I talked to was positive and grateful for all the effort. It was one wild ride that turned out better at the end than we thought it would at the beginning. As the saying goes, all’s well that ends well.

So there’s a bit of background on how we managed to make a conference work in spite of a silly thing like a volcano going off at the top of the world. Now, assuming I can actually get back to the States on schedule, it would really be nice to find some deserted beach somewhere and unplug from the world. What are the odds, d’ya think???