Boston Python Meetup

I've had a chance to visit Boston Python Meetup in November. Several famous schools (like MIT or Harvard) and a lot of techonological companies/startups are located in Boston Area. So should I say I was very excited to attend this event? And as long as I am involed in Minsk Python community, I'll try to compare how meetups are organized in Minsk and Boston.

The paradox of this blog post is that its target audience is members of Minsk Python community. So English is probably not the shortest path to the audience. However I'm also going to share my thoughts with Bostonians, and I have no other choice but English here.

That was the first public meetup I visited in the US, so it's hard to tell what is common for all the events organized there and what's special about Python meetups in Boston.

Logistics

Meetup.com platform is used for communication. I was warned by my friends from Boston that I need to grab an invitation in advance - sometimes they run out quickly.

Before the meetup I received a couple of emails - all signed by Ned Batchelder (the author of coverage.py). Ned also presented the speakers at the event. Seeing a real person in the emails gave some confidence (especially taking into account I've been reading Ned's blog for a long time).

Talks

1. How to scrape together a dataset using things you found on the internet by Deborah Hanus (there is a video recording of this talk from SciPy conference on youtube)
2. Lightning talk about recurrent neural network model for classifying speech pathology (I didn't write down neither the speaker name nor the title :( )
3. Lightning talk Linear Programming in Python by Thomas Keefe (that was a very good introduction to simplex method)

The first thing that surprised me was that all the talks were dedicated to data analysis. I think that the main reason why there are so many data analysis talks is actually current state of Python tools - they are becoming better and better every day, so it's becoming easier to start with them. However it's still an exciting area and some astonishing results seem to be around the corner. That's why it's interesting both to give talks and to listen to them. Well... data is the new gold, and data scientists are the new gold miners.

However we usually have more engineering talks at Minsk Python meetups. I can see two reasons for this: we have a lot of outsource development shops in Minsk, so projects' scope and key decisions are usually well defined. And because these shops usually work for European or American companies, they usually can offer very competitive salaries.

Data analysis can be useful when your company actually has data and needs to optimize its business. But in Belarus companies that have data are usually working for local markets, so it's hard for them to have as competitive salaries as development shops do. And choosing between career of a data scientist and a software engineer many prefer engineering (even if they can be very good at data science).

We have data science and business analysis communities in Minsk. But they also seem to be more focused on outsorcing needs (big data stack, requirements gathering, etc.).

But well... maybe I'm too pessimistic and Minsk is just a couple of years behind the trend. We'll see.

I can see that speakers take their talks more seriously in Minsk. They are often afraid to give talks until they become really good in the domain. This makes every meetup very interesting, but this doesn't help much to develop the community. I can see cultural background here: professionals in ex-Soviet countries want to be damn good at what they do, but they don't pay enough attention to soft skills. And we also have great demand for engineers, so carrer doesn't depend much on effective communication.

People in the US generally have better communication and presentation skills (probably from school). And they are also aware that soft skills are really important for their careers. Giving talks at local meetups is a good chance to develop such skills (even if you are not a domain expert). That's also a chance to find a job, an internship, etc. And this also makes meetups more inclusive: attenders can see that they don't need to be experts to give interesting talks.

Sponsors

In Minsk we don't have many absolutely independent communities. Many meetup events are organized by companies or event spaces directly.

For example, Python meetups have always been organized by Wargaming and annual Python conference is organized by Event Space.

Wargaming does very much to make Minsk Python community sustainable:

• they rent venues
• they buy food
• they make photo and video recordings
• sometimes they even bring speakers from other cities

But people attending the meetups have feeling that meetups are literally owned by Wargaming, so the community members are less initiative than they could have been. And this also means other companies don't want to contibute a lot into Python community development.

So what I really liked about Boston Python meetup is crystal relationships with sponsors. It looks like the community is completely independent, but events have three kinds of sponsors:

• Venue Sponsor - this is usually a company that hosts an event
• Pizza Sponsor - they just buy some pizzas and drinks so people coming to Meetup after work don't need to think about their dinner
• Beer Sponsor - they usually organize some kind of afterparty in a nearby pub

I've looked through the list of recent events and found that venue is usually provided by Microsoft, but pizza and beer sponsors usually differ. Pizza and beer sponsors give short talks about their companies (and both companies were hiring).

These relationships are clear for both community members and companies. And all the parties have some benefits.

Project Nights / Presentation Nights

Boston Python community has two event kinds: project and presentation nights. Each one takes place once a month. That means they have community events every two weeks (which is really cool).

We never had anything similar to project nights in Minsk Python communities, but I really love the idea. Here comes its pitch:

Bring a laptop. Bring ideas. Bring questions. Bring enthusiasm. The Boston Python Project Night is a time to find others with compatible interests, and work away. Code, hack, teach, learn, meet, laugh, talk: it all happens at Project Night.

I haven't visited a project night yet, but I like the idea a lot. We'll see if we can organize project nights in Minsk too.

Lessons Learned

Some ideas I'd like to try in Minsk:

• Project nights.
• Clean and clear communication (schedule should be available in advance, emails should contain real names of organizers).
• Try to be more loyal to beginners.
• Cadence - Bostonians have a good schedule.

What can Boston improve:

• Video is missing.
• Lightning talks lists should be published after the events.
• And links to the slides can also be published.

I'll probably go to Boston once or twice during the next year. I should really prepare at least a lightning talk by that time - that will be a good chance to meet with new people.

Update from Ned Batchelder

I'll correct one impression of yours: the talks were all about data analysis on that night, but that's because I collected a few talks on the same topic. If I can, I like to have themed evenings, in the hopes of drawing in new members. Someone who is interested in data science is more likely to attend their first event if all of the talks are about data science. It is a very hot topic these days, so it's easy to find three talks about it, but other nights will have other topics. For example, take a look at the seven lightning talks we are doing next Tuesday.