Week 3: New office, lo-fi, papers, and a general meeting

First week in the new office

It seems like it’s been ages since we last worked on the fifth floor of the Engineering Building. I worked there for less than a month, but I was already getting used to the fake wood, the orange atmosphere, the constant strolling about for coffee (not me, as I don’t drink coffee). Not to mention the occasional weird conversations on the other side of the court.

As it is, I’m quite content with the change. Focusing is easier, the room is cozy (if a bit too hot at times), we have a (shared) fridge, among other amenities. What I like the most, however, is getting to share the space with the people actually working on the project at hand. Don’t get me wrong: I’m not saying I don’t like the people working on the fifth floor. On the contrary, they’re great. I’m just saying, we finally get to have a room for us and us only. That’s neat.

lowfi – A simple, offline lo-fi player

I downloaded lowfi, a lo-fi player with the lo-fi already embedded. It’s quite simple and runs on all major desktop platforms. Having to pull up a terminal on Windows looks odd, but it’s worth it. If you use web apps to meet your background lo-fi needs, give it a shot.

Reading a systematic review

I started reading a systematic review that examined the use of grammars to recognize patterns in images. They focused on several applications such as object recognition, layout recognition, and object construction (which piqued my interest).

I took some notes using the Keshav (2016) three-pass method. I can say it helped me focus, which is something I’ve been struggling with.

EMB, GPB and I did shared pomodoros to help us stay focused. We’re thinking of getting a physical pomodoro clock that we can all hear. For now, we just rely on EMB to tell us when it’s due.

D’s visit

EMB had a friend come over to visit. Lady D (pun intended) will teach Construction in high-school. She brought some wine from Putumayo / “Fucking May”, a Colombian department bordering with Ecuador.

English Wednesdays

Last week we settled on English Wednesdays. We, and by “we” I mean “the people from the AMSI group who agreed to it, plus others in the office”, were to speak and write in English for the entire day. It was less cringe than I expected. I quite enjoyed it. It becomes more natural as time goes by.

Right after lunch, the fifth floor peeps visited the new office and we took a group picture emulating how packed it’ll look once the six of us are settled (hopefully).

First general meeting

On Thursday, the 5th of February, we held our first general meeting. All team members, professors and students, attended the event. However, only us students had to present our ideas and work so far.

Did I do a good job? I don’t know. I wasn’t too nervous during my presentation, which was good and unusual. Professor Guarín’s only feedback was to use a different color palette for the mindmaps I made (about the mindmaps: each of us had 10 minutes and 2 slides: how to make the most of it? A mindmap per slide was my solution.)

Voronoi diagrams come into frame

Amidst the discussion, professor Guarín brought up an idea he had for a data visualization of the nearest public library to each location in Medellín. It would resemble a Voronoi diagram since it’d use the Euclidean distance instead of the travel distance. I wondered, why don’t I try and do that myself? So I’m looking into that.

OSMnx for visualizations of street networks

I started reading through the OSMnx documentation.

OSMNX is a Python package that interacts with OpenStreetMap data to create visualizations of street networks. It’s not limited to street networks, however. OSMnx also works with landmarks, elevation, orientation, trip statistics, etc.

The main author published an article about it in 2025 (last year):

Boeing, G. (2025). Modeling and Analyzing Urban Networks and Amenities with OSMnx. Geographical Analysis 57 (4), 567-577. doi:10.1111/gean.70009

Thanks to the docs, I learned about the expression “as the crow flies (or alternatively as the bird flies)”, which is “an idiom for the most direct path between two points” (Wikipedia or Wiktionary).

I know it’s not much, but I found it interesting that one can filter by tags. Like so:

1import osmnx as ox
2
3# get all building footprints in some neighborhood
4place = "El Poblado, Medellín, Antioquia, Colombia"
5tags = {"building": True}
6gdf = ox.features.features_from_place(place, tags)
7
8gdf.shape
9fig, ax = ox.plot.plot_footprints(gdf, figsize=(10, 10))

Movie Night

EMB came up with the idea to do Friday Movie Nights. The movie/film being the reward for the Viernes de Seriedad, in which we work hard to make the film worth it.

We decided on the 1998 German film cult classic Run, Lola, Run (Lola rennt is the original title in German).

We were joined by AYMV, OARC, MABB, JM, and of course EMB and I. Overall, we had a great time watching the movie as a group.