I love PyCon. And until 2016, I can’t go, because the organizers failed to open a calendar before scheduling the conference in conflict with Passover—for two years running.
This is the story of how a large population was casually excluded from an event that assigns so much value to being inclusive. I say “casually” because it has been almost an entire year since the organizers were made aware of their failure, and they haven’t communicated publicly about it once—not to own up to the problem, and not to take steps to prevent it from happening again. Given the swift actions following incidents (1, 2, donglegate) at PyCon, it’s hard to feel that this failure is being treated with any kind of gravity when it escapes mention for so long, so it’s time to speak up.
Before I dive into the details, I’d like to stress that I have no personal animus with the organizers, and any suggestions of malice or (sigh) “antisemitism” on their part are simply unfounded; they’re well-intentioned people. This is a case study of how a series of technical and communication failures result in exclusion, to the detriment of both the community and the excluded. I’m hoping that it spurs some action—maybe 2014 isn’t salvagable, but there’s time before 2015. At the very least, some public communication acknowledging the problem is indicated.
For those of you who aren’t aware, the Jewish Calendar is a lunar calendar. Because the solar year and 12 lunar cycles aren’t the same length, a somewhat complicated system of leap months was introduced, with the entire system operating on a 19-year cycle. The unfortunate effect of this timekeeping hackery is that specific dates on the Jewish calendar move around on the Gregorian calendar (aka “the calendar”, for most of us). The fifteenth of Nisan, or Passover, usually falls sometime in April, but can even occur in late March, as it did in 2013. The actual calculations are also influenced by ecclesiastic judgment on whether the proper season has arrived yet—no point celebrating a harvest festival too early. If you’re used to holidays occurring on consistent dates like “December 25th”, then welcome to the mess of Judaism.
What this means for pretty much every Jew I know is the following:
- Open your calendar app
- Subscribe to an ical feed of Jewish Holidays
- Know when the holidays are taking place
Boom. It’s like that Oprah bees GIF, but for Jewish holidays. It’s stupid simple, takes one minute, and should be viewed as an obvious requirement for anybody scheduling events.
During PyCon 2013, I was hovering near the doorway to the overflowing Guido keynote with Diana Clarke, the newly-instaled PyCon chair for 2014–15. Capitalizing on the opportunity, I asked about the dates for the next two PyCons and opened up my calendar.
With growing horror, I saw that both years were either in direct conflict with Passover or scheduled immediately before—e.g. that Passover is the day following PyCon. On a practical level, this is an issue for anyone living east of the Atlantic; over 24 hours can elapse on the trip home, so you’d have to leave Pycon on the first day to be back in time for the Seder.
Even if we ignore the needs of attendees living outside the United States, no conference organizer in their right minds would even think about scheduling a large conference on December 22-24, but that’s exactly what happened here for 2014. And in 2015 it’d be even worse, like scheduling a conference on Christmas day. The conference should never have been scheduled on these dates, and it should seem reasonable to absolutely nobody that they were.
I remember the blood draining from my face when I told Diana about the conflict, expecting her to share in the mortified horror of such an epic fuckup. What I got was something along the lines of:
- “Oops.”
- “We already signed contracts for both years. I don’t think we can change the dates, but I’ll look into it.”
Fair enough—there’s no batphone to pick up and resolve things of this magnitude on the spot, and different people handle surprise in different ways. But over the next day or two, everyone I know in the PyCon organizing team took me aside and apologetically explained how the ink is dry on the contracts and there’s nothing to be done. With a growing sense of frustration, I began to feel that this was somehow being treated as Idan’s Problem, and not something that applies to a lot of people. The (very PC/American) impulse to find the nearest token member of A_GROUP and apologize for INJUSTICE made me the address for a lot of well-meaning but unwanted sympathy. I wanted action—either to fix the conference dates, or failing that, fix the process in a public and accountable fashion.
The worst revelation of all came a day or two later when Diana followed up with me about her efforts to move the conference dates. In that conversation, she explained why the current dates had such gravity to them: “They were the only ones available in April!”
The venue knew it was Passover. The hotels knew it was Passover. The caterers knew it was Passover. The groups who booked other events in April knew it was Passover, and didn’t book for those dates. None of these had any duty to inform the PyCon organizers of their imminent mistake. Getting a good rate on the “only available” dates in April was no fluke. They were damaged goods and we bought them, for two years, without doing our due diligence with a calendar. And then we compounded the failure by failing to communicate it to the community, failing to publicize a plan of action to rectify it, and failing to address how we can prevent this from happening again.
I’m murkily aware of “some efforts” underway to try and talk with Tourism Montreal, whom I suppose are key to moving the conference dates. I haven’t heard anything public on the matter, nor have I gotten any updates I didn’t solicit. It may be that there is real effort going on behind the scenes, but it might as well not exist for all of the communication surrounding it. Ten months is more than sufficient opportunity for the organizers to go public about the bug and the efforts to fix it—but they haven’t. At this point, I’d be surprised to hear that there was more than a desultory attempt to unbreak the conference dates. I stopped pestering them for updates, and the problem seems to have gracefully fallen off the radar.
PyCon is a popular event. There’s no way to count the number of people who won’t come because of the scheduling. The event will sell out anyways, because there are more people who want to attend PyCon than tickets. From the organizers’ perspective, this whole saga is an invisible problem—and that’s exactly how it’s been treated thus far. Some people can’t attend for two years, and that’s just part of doing business.
This would be a fine attitude to have at a commercial conference, but it has no place in an event with such community significance and values—an event that knits our diaspora of a community together. It has no place in an event attended by people who contribute their nights and weekends for love of their open-source work. It’s exactly the kind of dismissive attitude that we’ve worked so hard to change when it comes to gender imbalance. We talk a big game about our devotion to the cause of inclusiveness, and this incident should be setting off alarm bells in people’s heads, because we’re not delivering on our goals.
To wit, it’s hard to overstate the corrosive effect that this saga has had on my relationship with PyCon, and in a larger sense, with the Python community. I’ve been involved in conference organizing for a number of years now, helping out with six DjangoCons and two PyCons (full disclosure: often in exchange for material rewards like travel assistance and conference tickets, though the labor involved far outstrips the monetary compensation, and I would have been applying for financial assistance regardless). I don’t know if I’ll be involved in PyCon again, and my energy for contributing to the community over the last few months has been at an all-time low. I haven’t submitted a talk this year, because my chances for attending are practically nil, and I’d like to think of lost speakers as a concrete measure of community loss.
I’m more aware than most people of the hard, mostly thankless work that the conference organizers—and especially the conference chair—do every year. I’ve seen firsthand the effect on Jesse Noller and his family. They juggle an unreasonably long list of requirements in the service of us all, and I can see how they might see this as just another requirement they would have liked to juggle, but failed to.
It is hard to admit failure, but keeping schtum about it is wrong. I have sympathy for the people wearing the big shoes, but this was a sizable oversight that happened under their auspices, and carries costs for our community. It shouldn't get swept under the rug.
Mostly, I miss PyCon, and everybody in it, especially the friends I only see at these gatherings, and the new ones I am exposed to. All of those projects outside of Django, and all the people excited about using them to make awesome shit. It is what makes PyCon such a unique event, an opportunity to form bonds with people in the flesh. It’s a miracle that we manage to simulate the thing over the wire for the remainder of the year.
So, until 2016. I hope the simulacrum holds up, or I find other conferences that aren’t scheduled on major Jewish holidays. But it won’t be PyCon, and you probably won’t all be there, and I’m left holding this bag of sadness about the whole mess.
Idan is a creative geek, designer/developer hybrid, and a Django core team member. He’s idan on GitHub and @idangazit on Twitter. His badly-in-need-of-some-love website is over at gazit.me.
Dude, it sounds like you were passed over. ;)
Thank you, thank you. Please take care of your wait staff. I will be here all week.