1298592 - [decision] What is the future of l10n hackathons?

Status: NEW

(Localization Infrastructure and Tools :: General, defect)

Description

[Jeff Beatty [:gueroJeff]]

One thought that keeps coming back to my mind about hackathons is whether we're taking the right approach for the size of team we have. I know that we've had some separate conversations going about this and I wanted to bring it up among the group.

These are my concerns:
1) The overhead of planning each hackathon is quite high. At a time when we need to be more focused on our team's alignment within the org, this seems to take away time we need to dedicate to that.

2) Do we have enough people to travel to these hackathons?  

3) Burnout. The amount of travel required for some of us to attend these hackathons can be exhausting.

4) Team off-site this summer. We mentioned in Stockholm that we wanted to have more off-sites/work weeks among our team. If a lot of us are traveling for hackathons, it's less likely that we can add a team work week between London and Hawaii All Hands.

There are a lot of pros to doing hackathons like this that I'd like to preserve. I also want to be realistic about whether we can make these scale and remain efficient.

Comment 1

[Jeff Beatty [:gueroJeff]]

From flod: 

1) I can't really talk about the organizational overhead, since I've been lucky enough of being in two events so far where the organization was basically taken care by Jeff. My feeling from the outside, looking at other people currently doing the work, is that it's big and stressful, and it doesn't really scale.

2) and 3) No, I don't think we have enough people for this much travel (both physically traveling to events, and planning for them). In the past 4 weeks we've been all over the world, and I lost the count of times I've heard the sentence "I won't be available for X because I'll be traveling to Y".  I'm personally looking forward to a few weeks of calm at home after the last crazy month.

4) Yes, please, work-weeks. Chatting on Vidyo and IRC works, but being in the same room helps a lot, especially for bears working from home :-)

Having said that, I think we need to deliver as much as possible of what we promised for 2016, otherwise we won't be treating our communities equally. And start planning for 2017. The feeling I've got from the Nordic event is that nobody will be able to organize the event next year without a serious push (and work) from our side, on the other hand Celtic folks would probably manage.

As for 2016, I can probably help for August, but I'd need to figure out the VISA situation. We also need to assess November and Paris, it might just be me and Delphine at this point (that should be relatively easy, being in Europe).

If there's one thing that works in these meetings is the size of them: an event of 10-15 persons lets you know and talk with all of them, it's hard but not impossible to find places to eat at, you can move them around the city easily. For example, the Balkans event last year felt already too big in that sense.

And we need events, that's the main complaint I've got from my own community: you're doing a lot of work for free in your spare time, in the past traveling to MozCamps was a huge and tangible form of recognition. But, again, I don't think it scales from an organizational point of view.

The first idea would be to involve the Participation team to organize 3 events (South America, Europe, Asia) and letting them take care of all the organizational aspect. But then I'm quite scared of what they would require from the attendants in terms of "show me the work you've done and what you plan to do". Also, I don't see "let's split up and work on localization" as part of an event organized by the Participation team.

Francesco

Comment 2

[Jeff Beatty [:gueroJeff]]

From Matjaz: I share all these concerns. Especially after Prague where only 3.5
localizers were talking. Seems like consolidation is the way to go. ;)

I wonder to what extreme shall we take it. 1 event to rule them all? 2
after each of our off-site? 3 "per continent" events? 1 seems like a
big event in terms of effectiveness. And 3 seems like equal if not
more hackathons than some of us have been doing per year if we
organise them as separate events. ;)

A couple of hackathons following our team off-sites (or even all
hands) could be an interesting idea, but that's just my gut feeling
speaking. We should look at the numbers and see:
- how many people attend hackathons per year
- how many l10n drivers per capita is the minimum we think we need
- what different event sizes mean in terms of cost, especially travel

-Matjaž

Comment 3

[Jeff Beatty [:gueroJeff]]

From Axel:
I agree that the organization-piece is heavy, and I'm saying that as someone that slacks through it twice.

One thing that drags on for example is the goal setting for each team.
Also, visa handling is an interesting set of intermingled dependencies, which may inverse the planning you thought you'd do. Like in Berlin, you have to find lodging first, and then you can send out visa applications.
pro-travel is OK-ish, I guess. Also visa requirements for us to go to places are "ugh".
Budgeting is black magic to me, on both creating one, and having spent money on another.

To add a pro, I actually get energy from the hackathons. I'm really exhausted quickly after, but I also find the interactions energizing with two days of distance.

Comment 4

[Jeff Beatty [:gueroJeff]]

From Stas:

Chiming in late here;  hopefully this is still a helpful thing to do.

I relate to all things said here before me.  I can't speak much about
the organization part (thanks to Jeff for taking care of that), but I
can only imagine that it's a lot of work between encouraging the
localizers to do their part, booking airbnbs etc.  I guess there's a
threshold at which it pays off to outsource that to an event company.
Do we know from previous experience and from other teams what that
threshold is?  Is it 25 people?  50?

Smaller groups are easier to handle and manage: in Riga it was easy to
find a place to eat and also easy to get focused on a particular
topic. OTOH, they mean more meetings in general for our team and also
less energy and synergy.  It's definitely worthwhile to consider going
back to the idea of bigger events.  Tying them to our team work weeks
sounds great too.  For instance, we could start a workweek on a
Wednesday (Tuesday being the travel day), work for three days, meet
with the localizers over the weekend, and then work for another two
days and debrief on Monday and Tuesday.  This way all the happenings
from the hackathon are still fresh in our minds when we talk about
them as a team.  Back home on Wednesday; the total works out to be 8
days including travel.  Not bad for 5 days of work and 2 days of
meeting with localizers. I would also suggest that Thu-Fri after such
a week would be off to make sure everyone has time to rest and
recover.

What I liked the most about the Riga hackathon was the "hack-" part.
We did some talking in the morning and over food but we also had a lot
of time to just work. I'm not sure how well it scales to bigger groups
but I'd definitely like to try doing more of that.  We should also be
careful to not send a signal that it's okay to lag behind because
eventually there's going to be a hackathon to catch up.

Something that I really liked during Summits in the past were Science
Fairs and World Fairs. We had cocktail tables for everyone who wanted
to talk about their projects, culture and community.  it's a great way
to hand off some of the talking to the participants, give space to
talk to and with a lot people over a short amount of time and to get
to know them better.  I wonder if this is something we could try doing
if we organized a larger event.  I'm imagining a table for each
community where they can demo their tools and their process, talk
about roadblocks, talk about what's hard in their language and banter
with other teams.  I can also see myself and Zibi by a table with a
demo of L20n and Matjaz with a demo of Pontoon; it works both ways.