Telephone +44(0)1524 64544

Announcing Hallway++

Wed May 23 21:00:00 2012

Announcing Hallway++

Sad news, good news

So, as many of you have discovered to your great dismay, YAPC::NA 2012 has, sadly, sold out. Now, on the upside, YAPC::NA will be televised so even if you can't make it, you'll be able to watch any and all of the talks as they're given (although likely with the odd exception since certain speakers/talks may have reasons to opt out).

However, a number of people have already made the decision to turn up anyway, because firstly there may be cancellations which will free up main conference tickets (I think last year had something like 40 unclaimed badges in the end), and secondly they, as with many of us, are there for the hallway track as much as for anything else.

Which got me thinking ...

Hallway, oh hallway, how we love thy walls of beige

For those of you not familiar with the concept, the "hallway track" is what people tend to use as a shorthand for, well, all of the things that happen out in the hallways of the conference - there might be five tracks of talks but there's likely to be a dozen conversations going on in the corridors and rooms (and outdoor spaces) around the presentation rooms, and a lot of those conversations will likely turn out to be at least as rewarding to the participants as attending any particular talk would have been.

On top of that, it's a great way to meet people who care about the same things you do, and hopefully also have sufficiently compatible tastes in alcohol (or lack of alcohol) that you'll find people to while away the evenings with, talking about life, the universe and everything ... and probably also at some point Perl.

I love this part of conferences. Honestly, I love it more than I do attending talks most of the time, because in the time it would take me to attend one talk I could have read the source code behind two talks and still had enough time spare to have a fascinating five minute conversation with each of the speakers.

I also, personally, really enjoy meeting people who're at their first community conference, or people who're relatively new to perl, and spending time talking to them about what excites them, what problems they've encountered, and what cool things they've built. It's a great way to not only realise where the CPAN ecosystem stuff I'm working on is leaving cracks that the unwary/inexperienced can fall through and hurt themselves, but also to find people who'd love to contribute to fixing that.

Destructuring the unstructured

Now, so far I've basically achieved my personal part of these goals by doing my best to be extremely visible (I'll be the 6 foot blond guy wearing a utilikilt and unnecessarily unkempt sideburns) so people talk to me; by walking into other people's conversations if they sound interesting; and by ensuring I grab a bunch of newbies and footsoldiers as well as people who're CPAN contributors of years' standing when I'm about to lead people forth to a better pub.

That's a great approach for me, but it totally doesn't scale, and I'm sure I'm missing people.

In recent years my good friend Yaakov has attempted to bring structure to this process by organising Very Important Person badge inserts for the people who're on their first and second YAPC, and by having a slice of time set up where everybody's invited to mingle, and where newbies in particular are encouraged to walk up to people whose names they recognise and introduce themselves.

It didn't really work. A few conversations started, but then quickly sizzled out because there was a timelimit, because people weren't sure what they wanted to talk about, and because once two people got into a conversation about CPAN code usually other people didn't join in because they didn't want to interrupt.

Propriety and pervasiveness

I have an approach that I think will scale.

It's a plan so simple even I could manage to organise it, and yet so cunning you could brush your teeth with it.

This needs to be everywhere. The whole conference needs to be a VIP event.

If it's everywhere, we can get into conversations on the first day. We can start meeting each other right from the start. We don't have to worry that we're taking too much of somebody's time, because it's easy to say "hey, let's chat about this later over a drink" if you're in the middle of something else.

If it's explicit that people want to be talked to, the propriety issue goes away as well - if we have a system wherein you can join a conversation to listen, and be incorporated fully into it at a convenient moment, then people can add their ideas to a discussion without disrupting it, or just soak it up and see if it gives them ideas.

So, here's how it works.

The Hallway++ Manifesto

People actively participating will have HW++ on their badge or something - really, just write it on with a pen, we're not trying to be clever here.

Groups actively participating will have a sign on the table that they're sat at, or somewhere next to the group - probably a piece of paper folded into a makeshift sign with Hallway++ written on it with a pen. Still not clever.

When encountering a Hallway++ group, come join us. That's the whole point.

If people seem to be mid-conversation, sit down anyway - or stand with them if they're stood up.

When there's a convenient break in the talking, introduce yourself. If the group just does not shut up (I'm likely to be guilty of this) then if you're part of that group, you should try and remember to say "hi" to the new arrival when the conversation comes to a convenient conceptual pause point.

If you're likely to be around in the evening, asking the group where they're planning to go is always acceptable. We'll likely end up descending on the same two or three bars for the whole conference, so the mingling cycle should take place there as well.

Don't be a problem lest either mst or an mst-trained bastard happen to you.

And that's it.

But I don't have anything to say!

Don't worry about people knowing more than you. Everybody knows different things, and everybody has something to contribute. From one side - some of my best code ideas have come from listening to somebody who was stuck with a problem tell me why it was driving them crazy. From the other side - I spent a while explaining the perl5 VM to a guy who'd been talking to somebody else about C optimisation; he had some interesting thoughts, and mentioned running OS/390 on the Hercules Emulator as an interesting environment for testing parallel C code. Later, I realised I'd been chatting to Tridge.

Great bloke.

Where to start.

The Hallway++ track will begin at the Arrival Dinner where those of us who've already arrived will be present and already flying the flag.

If you're going to be getting into Madison in time, please RSVP to help Uri Guttman (who is organising this as a volunteer to help people get to know each other, as in many years previously) gauge numbers.

Where to find out more

We'll be putting information on a page on the YAPC wiki.

We are going to be on IRC on #yapc - we're part of the conference, and it's relatively quiet in there since this year's organisers don't seem to be big IRC people.

We are going to try and find good beer places; we'll post 'em up on the wiki when we do (anybody arriving early and a member of is encouraged to join in with this research, in the name of science).

We are going to try and find a venue to do unconference-y stuff - i.e. people turning up and saying "hey, I could give a lightning talk" or whatever - and hopefully somewhere we can stream the main conference if nothing else is going on. If we do that, it'll be wednesday and thursday afternoons only, so it won't overlap the keynote or Larry - and because frankly most of the people currently looking at this hate mornings.

If you have any ideas, suggestions or criticisms, please do share them. This is the community's layer on top of YAPC::NA, so if you're attending, if you're part of the CPAN community, or if you just want to laugh at me for thinking this could possibly work, get in touch.

I think this is going to be awesome. I'll keep you posted.

-- mst, out