111. Why Software Hasn't Eaten Museums (Yet)

Welcome to Museum Archipelago. I'm Ian Elsner. Museum Archipelago guides you through the rocky landscape of museums. Each episode is rarely longer than 15 minutes. So let's get started.

If you've been to a museum lately, any museum big or small, you've probably noticed ever more software-driven experiences. Interactive touchscreens, projections, buttons, videos are all controlled by software.

Lucas Junqueira: I've seen mostly all exhibitions have at least some kind of interaction or some pieces of the exhibition that require some kind of software to enable it.

This is Lucas Junqueira, a Brazilian museum professional and software developer.

Lucas Junqueira:  Okay. My name is Lucas Junqueira. I've been working on this exhibition museum scene for quite some time right now.

It's tempting to see the increase of software in museums as another example of software eating the world. This phrase, "software eats the world", was coined by investor Marc Andreessen in a 2011 Wall Street Journal opinion piece.

The idea isn’t that software replaces everything. It’s that software absorbs the value layer of company after company, industry after industry.

In the "software eats the world" thesis, traditional music labels only exist to provide software companies (like Apple Music and Spotify) with content. The software, everything from how the app looks on your phone to what song is recommended next, sits higher on the value chain than the music itself. Even the network effects of which app your friends use, matter more than what you listen to.

Netflix, Salesforce, and the game Angry Birds are some examples Andreessen mentioned back in that 2011 essay, plus plenty of other companies I barely remember. But the core of the thesis, even if it's not explicitly mentioned, is the zero-marginal cost nature of distributing software. Whatever the up-front cost of developing Angry Birds is, it doesn't actually matter since the company can distribute it to every phone on the planet for zero dollars. Once it exists, it can be copied endlessly, instantly, and globally. Which is exactly what museums can’t do.

And this is where museums are different.

Museums aren’t infinitely scalable, and they aren’t frictionless. You can't really download an exhibition in the way that you can download a song. You have to show up. There's only one Mona Lisa.

And that's why software in museums is (so far at least) immune from software eating the rest of the world.

Yes, there's more software in museums than ever, but that software rarely becomes the value layer.

Museum software doesn't own the audience relationship, and it doesn't become the product like Spotify was able to do. Instead, it supports the product. As long as museums continue to control distribution, software remains infrastructure: a tool to help museums tell their stories.

Lucas and I are both software developers, specializing in building exactly that infrastructure.

One of Lucas's early projects was creating software to project images onto the facade of the Space of Knowledge museum, which faces Liberty Square plaza in the Brazilian town of Belo Horizonte.

Lucas Junqueira:  We provided them a tool so they can make, a projection. They do have a projection on the front wall of the museum. It's a very big projection, and it is right in front of a square here in my city that is very crowded all the time.

So they wanted to use this to show something they are exhibiting right now the museum is funded by a university, so they wanted to show some scientific evolutions. They want to use it like an information tool. They wanted to do that and provided them with the software used to produce the content to show on this big projection.

Almost always, the cost of these projects is front-loaded: museums pay Lucas, or me, to create the software that projects the images and everyone is focused on that opening day. Lucas explains that this focus on new exhibitions is related to sponsorship funding.

Lucas Junqueira:  We see that,, most of the museums had some requirements when they start a new exhibition, they usually have some kind of budget and some kind of funding to that. And that doesn't really happen when these institutions have to maintain their spaces. They can get some sponsors to new exhibitions, but they don't get so many sponsors when they are just keeping the museum open, right?

And this is what happened to the projection on the facade of the Space of Knowledge museum. Long after Lucas finished building the software, and long after the project had successfully finished, the museum wanted to update that projection to show new content.

Lucas Junqueira:  And it was these, that, brought me the idea of creating a tool so the people on museum can use it without technical knowledge, because I felt they were needing to create some interactive content, but they needed to do it by themselves.

Lucas wanted to make software that acted like infrastructure.

Lucas Junqueira:   And I felt this need to create something that the museums and the, institutions can use to both create the big exhibitions, the big interactions for the new exhibitions, but also something that they can use, on by day, day basis.

Lucas Junqueira: Because they don't have that sort of technical people or budget to fund these on a day by day basis. Right? And so when I started thinking about creating something that they can used for both things. They can use it when something new is coming and they can hire people to do it. They can hire technical people, they can hire developers to do it, but something that they can install on their computers and with minimal knowledge they can use to solve day by day problems.

The result is TilBuci, a piece of software that Lucas developed on his own. It's a tool for helping build the software infrastructure a museum might need. The name TilBuci comes from Lucas's dog.

Lucas Junqueira: His name is Busi He, he keeps company for me. He is always where I am. When I'm working, he's close to me all the time. So when I had to think about that name first I thought, that would be perfect, let's name after him!

TilBuci, which you can download for free on any OS platform you want, is an impressive suite: capable of doing anything from slideshows, interactive menus with animation, dialogue systems, survey kiosks, and quiz games.

Lucas Junqueira: It started growing slowly with the functions that started growing slowly because of course it's a free software and it's not funded by anyone. Well, it's funded by my own work. Right? So, I started creating like this. So when someone hired me to develop, for example, a kiosk and I had to build a, a in specific function instead of creating it somewhere else, I use it on that case, I create, I use it to create the code and incorporate it on the base software.And so the, the, the basis of, of TilBuci started that. It started growing from the day by day work, my of myself.

But it wasn’t without setbacks. Lucas’s early code was written to be distributed on Flash. Flash Player was, for a long time, an easy way to make interactive media work on the web and in museums. It was a browser plugin developed and controlled by Adobe, and it allowed designers to build animations, videos, games, and interactive experiences that ran the same way almost everywhere. If you wanted something dynamic, visual, and relatively easy to deploy, Flash was often the answer.

Museums adopted Flash for the same reason everyone else did: it lowered the barrier to making interactive content.

But Flash was also a closed system. Museums didn't control it. When Adobe decided to end support in 2020, thousands of interactives simply stopped working -- not because the exhibitions had changed, but because the supporting software wasn't allowed to run.

Lucas Junqueira: At  that time when we called the apocalypses, our digital apocalypse when we when the found problem of the Flash player. And what in about 2020 was kind of bad because it heavily depended upon the flesh player.

I remember the Flash Apocalypse too. That moment exposed something important. Even when software is "just a tool," the kind of tool matters. Especially when museums don’t control it.

And that's the problem Lucas was trying to solve in the first place.

So Lucas switched course and made sure that his software was made with open source technologies.

Lucas Junqueira:  I started thinking on ways to bring it back to life with something more open, some more open technologies. And that's when I started studying better the JavaScript and something like that. And then I could convert it, convert what we have done to what we have now.

While museums have so far avoided software “eating” them, that doesn’t mean they’re immune forever. It's not hard to imagine a company offering museums free interactive kiosks in exchange for visitor data, or a platform that hosts digital exhibitions and starts competing with the physical visit for attention. The moment a third party controls the interface between a museum and its audience, the museum becomes the content supplier—just like the labels became for Spotify. Tools like TilBuci show a different path forward. Instead of extracting value, they extend the life of museum systems by keeping them editable, local, and owned by the institution, long after us programmers leave. They allow institutions to adapt, update, and respond over time — even when budgets shrink and staff change. And this brings us back to the projection on the facade of the Space of Knowledge museum. That projection opened in 2012. And because of TilBuci, it’s still working today.

Lucas Junqueira: In this example,this projection without the tool for them to create themselves, it would be closed some time after we left there or there, or it would just be used to show some static videos and so on. But with the train them to create interactive content gave some life to this projection for a bigger time, bigger time spent than it was, than it could have been without it.

This has been Museum Archipelago.

 They never call it a museum. Jurassic Park has a visitor center, a tour, a gift shop, fossils on display.

It's curated, it's interpretive, it's full of didactic signage and high budget audio, visual material. It has a pulse visitor experience that begins with an introduction film, but nobody ever says museum. That's because museums are slow, they're dusty, they're institutions. Jurassic Park wants to be something faster, cleaner, cooler, a theme park, a show a miracle.

But Jurassic Park is a museum movie. The dinosaurs are exhibits, living dioramas, engineered to validate a particular vision of science and nature and human knowledge. The tour is scripted. The narrative is highly controlled. That is until it isn't.

The latest episode of our Bonus podcast, Club Archipelago is an hour and a half on 1994's Jurassic Park from a museum professional's perspective. It’s a good one. Listen at jointhe museum.club to join the club to support the show.