Episode 36: How to give (bad) talks

In this episode we talk cities and urbanization. Specifically how do we measure what is urban (and rural!!)? It may be a topic you've never given much thought to but—guess what?!—there are lots of folks out there who spend most or all of their time considering how to identify what is urban and what makes a city. For this episode we are joined by one of those people: Lewis Dijkstra, from the Joint Research Centre of the EU Commission, known for his work on the degree of urbanization (DEGURBA) classification.

Today, we’re talking about talks. Specifically, about *bad* talks. Speaking in public is such a core part of the academic life. We get jobs through it, we share ideas with our community, and we sometimes even try to persuade for money with speeches. And yet, it’s so hard to come by discussion, reflection and sharing about the act of talking. Worry not, GLaD to the rescue!

On this episode, Dani, Rachel, and Levi join Elizabeth Delmelle, Associate Professor in the Weitzman School of Design at the University of Pennsylvania and Director of the Masters in Urban Spatial Analytics (MUSA) program, to talk a bit about how AI continues to change our academic practice. About a year ago, GLaD went on the record about how AI was impacting how we work, live, teach, and research. And wow! How things have changed! Or have they? The tools have definitely gotten better in some ways, the capital-”D” Discourse has certainly moved on a bit, and at least one of us has apparently made friends with a chatbot named Claude. So today is our one-year check-up. Join us to reflect on what stuck, what didn’t, and what we’re thinking about today that we weren't a year ago.  Links - UPenn MUSA program - AI & GLaD in Practice: a field report - Simon Willison's writing on AI - Matt Rocklin: "The surprising value of AI today isn't super-human depth. It's super-human breadth."

This episode is part of our ongoing and occasional series on, “A Day in the Life of…” where we talk to folks about what exactly it is that they do (sadly, the evidence across episodes is starting to indicate that much of the answer is meetings and email). Like, how did they decide what they wanted to be when they grew up? And—just as interesting (we think)—how do they organize that crazy basket of activities that makes up “work?” What does a typical workday, week, or year look like? We're pretty excited to have Sarah Williams in our virtual studio with us for this episode. Sarah is a Professor in Urban Planning at MIT and she’s also Director of the Civic Data Design Lab and the Leventhal Center for Advanced Urbanism. Join us to hear what she has to say!

Today, we’re talking about embeddings. Unless you’ve been living… well, I don’t know where, you might have heard of the term generative AI. At the core of this technology is an obscure idea, word embeddings, that powers much of its magic. Embeddings work for text, but not only so. They also work for imagery and other types of geospatial data. This idea is taking force in certain circles, and we thought it’d be worth spending some chat time with this! Join us for a (hopefully) "lay enough" conversation about one of the most interesting ideas in geo-data that the last few years have produced.

You think balancing babies and a career is tough; try recording a podcast episode about the topic! Kidding. Sort of. In this episode, Dani, Levi, and Rachel are joined by special guest, Elisabetta Pietrostefani, from the University of Liverpool, for a wide-ranging conversation about babies and (academic) work. The complicated decisions around whether and when to have kids. Things we wish we'd thought of. Tips for navigating different stages of kid and career. One of the most interesting things about this episode, we think, is just how tricky it was to navigate the conversation—an important takeaway lesson!

For this episode, join Dani, Rachel, and Levi for a discussion about Academic Love Languages. Does getting public recognition make you feel warm and fuzzy? Or would you rather the freedom to curl up in a quiet office, reading and writing? We're all different, and it helps to be aware of what makes us (and others) feel valued, cared for, or supported. So, how do you show your colleagues that you value them? And, what makes you feel valued as a colleague? Join us for a spirited discussion about whether such a thing exists and let us know your academic love language by emailing us at [email protected]  

Happy new year!!! We know, we know, we’ve been in 2026 for a while already (already too long?). But this is the first episode we recorded this year, the first one with Levi as a dad (yay!), and one where the “year” itself will somewhat feature. So we find it apt to start with a cheerful approach.   The reason behind this episode is that I’m such a sucker for “best of…” and “what’ll happen in…” compilations. Like the best sci-fi, they tell you a lot more about the present, its hopes and anxieties, than about the year they either review or speculate on. And that’s always interesting to me. So Dani spent most of December listening to “hot takes” from all sorts of podcasts… And here at GLaD HQ, we’re all made of the “hot take” stuff. So, possibly several weeks “late” (in time is overrated!), here it is: our hot takes for 2026.

Come for the catchy episode title, stay for the wide-reaching discussion about spatial data, visualization, and cartography. We're joined this time by Anthony Robinson, from Pennsylvania State University, for a conversation about what cartographers do, what cartography looks like in the 21st century, what makes a map, and the challenges of "personal cartography." We've even got some requisite hot takes on spatial analysis and data (...and cartography!).

Today, we’re joined by a close friend of the podcast (someone some of us personally credit with changing our lives forever!), Serge Rey, to talk all things open, open source, and academia. Serge is Professor for Geography at San Diego State University and BDFL for the PySAL project. He’s been one of the most outspoken voices advocating for closer adoption of the open source ethos in academia, so we thought who better to unpack the open conundrum.