JupyterCon 2023

This is the first time I participated at a conference in Paris, also the first time participated at JupyterCon. It’s exciting to attend a conference without any expectations. I have made great connections, learn something new and had nice conversations with different people. The most fun part? The venue is a science museum and it matched the theme of the conference perfectly.

Keynotes, Talks and Tutorials

There are 3 keynotes in total, every day the conference kicks off with a keynote. The first day, it was a keynote by the team from GitHub, I was not too interested so I gave it a pass.

On the second day, we have Professor Alyssa Goodman giving a talk about the projects related to Glue - a 3 dimensional VR tool that is very useful in astronomy and beyond. She reminds us to not use our computer during her talk and show us why it is important to see things in 3D.

On the third day, we have Nobel prize winner Paul Romer giving his “lecture” style keynote - there were equations, quite a lot of them (considered you have been warned). It was very tough for me to attend a lecture at 9 in the morning during my university years and listening to his keynote at 9 am reminded me of this. Nether the less, he has delivered a very good message and idea - make Jupyter Notebook an interactive pdf by adding the SHA sign feature.

Other than the keynotes, I didn’t go to many talks, there are all recorded and I was carried away with chatting with people a lot of times. There are a few that I enjoyed.

One of the talks was by Jeremy, the maintainer of JupyterLite. It inspires me to try using it in workshops but this brings me to finding a bug during the sprint.

Another one is by Pavithra, who is the developer advocate of QuantSight, she is so sweet and it’s nice to see her in person. And she delivers her talk very well.

I also gave a talk about memory profiling on Jupyter, I was given the room that is used for the keynote so it is only 20% full - I think there may be around 80 people there. I also gave too many lighting talks at the conference… I hope the competition will be keener next time.

I also went for one tutorial (actually half) about security in Jupyter - I think it is interesting but when it get to the hands-on exercise part, I ran for some fresh air in the hallway.

Hall Track and Social event

There are quite a lot of sponsors at the venue, including Anaconda - our booth is a good place to meet colleagues and chat with people. It is a small booth and is at the back next to the NumFocus booth so it is not busy a lot of times.

I also talked to other sponsors and see if they are interested to sponsor other conferences that I organised. There’s one interesting booth with a Robot that can talk to you - it is connected to ChatGPT so it is fun to interact with it. After interacting with it for a while I found it quite adorable and now I want a robot too.

There’s also a PyLadies booth on the top floor, I often go there to chat with my friends who organise PyLadies Paris and Berlin. It is also a gathering place for PyLadies Lunch.

I also sometimes take calls at the side ally of the hallway, it is mostly quiet - and even quieter than the quiet room sometimes (I am not supposed to take calls in the quiet room anyway) cause people were chatting in the quiet room occasionally - I think the organisers need to remind people not to do that.

On the second day, we have a poster session before the reception (there are food and drinks there already) and I like how it is organised, it encourages people to go see the poster presentations.

After the poster session, it is the drink reception and the food is absolutely amazing. It is in the museum space and there were some creative ways of presenting the food - one canopy dish looks like a fruit that got picked from a plant; there’s a chocolate egg that was cracked with a hammer before serving… I think it’s one of my favourite social events.

People I met

Usually what makes me think the trip is worthwhile is because of all the people that I met and connected with. It’s great to meet up with old friends and make some new ones.

I am so glad to be able to have a chat with Carol again! She is so so so kind and inspiring and sometimes I feel like there’s a queue of people wanting to talk to her. I am glad to be able to “get a seat” and she is very warm and kind and has given me good advice.

Besides, I have a friend that I have not met in a while - Sandrine who founded Humble Data with me is also organising the conference, she is super busy and we didn’t get much time to hang out but we had a good chat during the conference.

Other conference friends that I have not met for a while are Vishal and Tereza, together with Noa we went for dinner one evening and it was a very enjoyable experience.

It was also great to chat with Pravitha from QuantSight and Sarah who used to work in the Alan Turing Institute. She is also leading the project Jupyter.

I also had dinner with the PyLadies Organisers, folks in the Scikit Learn team, Matt - code maintainer of Dask and Paul - the developer advocate of Capital One.

One more, I met Stephannie, a speaker who’s from Colombia, I got a handmade Jupyter ball from her.

Sprints

Since I was trying to use JupyterLite, I found out that some magics do not work as it does in JupyterLab. It brings me down a rabbit hole of looking at the iPython kernel code and JupyterLite code. Jeremy showed me how to see the message getting sent from the Jupyter frontend to the kernel at JupyterLab and how it is replicated in JupyterLite. It is fascinating. I am not sure I will be able to find the problem and fix it but it is a study opportunity for me.

I also realised that Jeremy is very kind and helpful. A few people are using JupyterLite and needing his help, he just keeps helping people and was so busy at the sprint. I didn’t work on any Scikit Learn issues but the person next to me did, he is a new contributor and Guillaume was so patient and give him a one-to-one mentored sprints - walking through all the basics of open-source contribution principles. I wish I can invite them both to the mentored sprint in the future.

Overall, I had a good time (I think I said that a lot going to Python community conferences) and I think the conference is a success. I made some good connections in the community and learnt some new things.

A small episode when I try to get my train home - it was cancelled and I had to sneak on the previous one, I was so scared that I will not be able to get home and got sucked into Paris.



About Cheuk Ting Ho

After having a career as a Data Scientist and Developer Advocate, Cheuk dedicated her work to the open-source community. She has co-founded Humble Data, a beginner Python workshop that has been happening around the world. She has served the EuroPython Society board for two years and is now a fellow and director of the Python Software Foundation.

Related

Cheuk Ting Ho -  - building the open-source community