Götz Bachmann’s Ethnographic Research on Dynamicland

10. January 2022

The “Dynamic Medium Group” led by Bret Victor which is developing “Dynamicland” is much admired and closely watched in the interaction design community. Yet the information about their work is closely-guarded and usually only available in a highly moderated form. The work of ethnographer Götz Bachmann, who spent much time with the group to observe them, offers a new perspective and deep insight into the group’s ideas and practices.

While it would be inaccurate to describe the Dynamic Medium Group as secretive — they do share videos of their prototype and there are many detailed reports online from people who visited their space See below for an overview. with whom knowledge is shared freely — it is obvious that the publicly available information is highly moderated. Little is available on the history, inner workings, and processes of the group. A notable exception is Carl Tashian’s excellent blog post on Dynamicland.

The Dynamic Medium Group’s vision is rooted in the idea that the computer revolution of the ’70s and early ’80s was cut short, primarily by premature com­mer­cia­li­za­tion. While the computer as a medium was still unfolding its potential, and way before it could do so entirely, it was solidified into commercial products, thereby stifling its free growth. Once corporations had built their businesses on the ideas developed so far, they were only interested in incremental change that could easily be integrated into the products, rather than revolutionary new ideas.

Because of this cautionary tale and to avoid repeating history, the group is wary when it comes to public attention and deliberately reserved in what it shares. It is therefore equally surprising and delightful that they gave German anthropologist and ethnographer Götz Bachmann full access to visit them, watch and interview them, even read their internal communication and write about them. Between 2015 and 2017 Bachmann spent eight months in total with the group during four field visits. In the meantime, he has published several reports and articles See below for a full list. Several are available in English, some only in German..

Bret Victor and many of the engineers of the Dynamic Medium Group are highly renowned in the interaction design community and the default mode to look at their work is admiration. Bachmann’s reports are particularly interesting because of his outsider’s perspective — not only from the group but more importantly from the wider “tools for thought” community in general.

In this note, I want to summarize findings from the various publications by Bachmann and put them in context. The focus here is on the history of the Dynamic Medium Group and what can be learned about the way they work — not the thing they are building. This note draws mainly on Bachmann’s writing and a radio interview with him but includes information from other sources. I added citations when quoting from Bachmann directly, but am paraphrasing his writing liberally throughout the note.

Table of Contents

  1. Founding & Funding
  2. Research Agenda
  3. Team Members
  4. Scaffolding, Staircases, Stacks, and Bootstraps
    1. The First Two Years — Scaffolding
    2. Rising Tensions — Towards Bootstrapping
    3. A Staircase Towards a Dynamic Spatial Medium
    4. Bret Victor’s “Yak Stack”
    5. Megalomaniacal Techno-Solutionism?
    6. The First Generation Leaves
    7. Building Dynamicland
  5. Links & Literature
    1. Publications by Götz Bachmann
    2. Publications by Members of the Dynamic Medium Group
    3. Notes from Visits to the Lab
    4. Interviews with Members of the Group
    5. Other

Founding & Funding

The research group was initiated by Alan Kay, who proposed it to Bret Victor in March 2013. It was one of the multiple research groups initiated by Kay at the time, as part of his newly founded “Communication Design Group” (CDG). The goal of the CDG was to create a research lab in the spirit of Doug Engelbart’s “Augmentation Research Center” from the ’60s or the Xerox PARC of the ’70s and ’80s, where Kay did some of his most impactful work. Some articles published around the start of the CDG were “5 Steps To Re-create Xerox PARC’s Design Magic” and “SAP Looks to Xerox for R&D Inspiration”.

It can also be seen in a line with Kay’s efforts with “STEPS Toward the Reinvention of Programming”, a project of Kay’s “Viewpoints Research Institute” (VPRI) which ran from 2006 to 2012. Victor is listed as an author in the project’s final report from 2012, though he is not mentioned in any of the project’s previous yearly reports. In STEPS’ first annual report from 2007, Vishal Sikka, CTO at SAP is listed as an advisor. It was Sikka who told Kay in 2013 that SAP would fund the CDG and that he could hire several principal investigators to start new projects, one of which was Bret Victor Another one was Vi Hart, whose blog post on the history of her CDG project offers another perspective on the story..

After Sikka surprisingly left SAP in 2014, the relationship with the company continued until early 2016, when the CDG projects became a part of YC Research’s “HARC” (Human Advancement Research Community). The relationship between Kay’s VPRI and CDG is not quite clear. The VPRI also became a part of HARC in 2016. This change in funding coincided with an internal crisis within the project, which will be discussed below.

The relationship with YC Research would not last very long: barely a year later, in July 2017, HARC was defunded rather unexpectedly The last reports on HARC’s website are from June 2017 — Victor is the only Principal Investigator whose report is missing. In 2020 YC Research became OpenResearch, which lists Dynamicland under “Past Groups Supported”.. I wasn’t able to find any details about the funding of the group since then. The concept of fundraising seems to be difficult given the nature and timespan of their work, particularly because the goal is to resist the commercialization of their ideas. Nonetheless, it is necessary to keep the project going: Bachmann mentions that the group published their website in early 2018 “as part of their search for funding” after they previously “avoided exposing Dynamicland to the online public as much as possible” Bachmann, 2019b, p. 38.

Research Agenda

Bachmann tracks the development of Bret Victor’s ideas and goals through his influential talks and points out how a sort of “radicalization” is happening.

The goal of the project was now clear: To help this new “dynamic spatial medium” come into being.

Team Members

According to Bachmann’s writing, the members of the group in the beginning were Bachmann, 2019a, p. 279:

The second generation of members, joining after most of the first generation left in 2016, was, in addition to Bret Victor and Toby Schachman Bachmann, 2019a, p. 292:

Bachmann last visited the group in 2017, so some more recent members are missing from this list:

Further contributors were:

As of early 2022, only Bret Victor and Luke Iannini are mentioned as staff on the about page of the Dynamicland website, all other members of the second generation are listed under “former staff”. May-Li Khoe and Dave Cerf were advisors to the group throughout. [Bachmann, 2019a, p. 292]

Scaffolding, Staircases, Stacks, and Bootstraps

As a social scientist, Bachmann pays particular attention to the metaphors engineers and designers like to throw around rather liberally.

The First Two Years — Scaffolding

During the first two years of the Dynamic Medium Group’s work, the strategy to approach the new medium was to have individual members of the team work on their own projects without much coordination and create many different prototypes, hoping to get a shared sense of the possibility space.

Bachmann describes the lab at this time as feeling like a “giant brain”: “The abundance of prototypes, books and manifests spread around the room made me feel as if I was sitting within a ‘giant brain’ […]: in a shared, associatively interwoven, growing room of ideas.” Bachmann, 2019a, p. 281 [translation mine] Seemingly contrary to this feeling, Bachmann writes that there appeared to be little direct collaboration, or even communication, among the members of the group: “The members of the group often didn’t seem to notice on another. They did not greet each other. They went to lunch together, but this was often a silent affair. When they finished their work in the evening they left wordlessly.” ibid. [translation mine] In Brown University, 2016, Bachmann points out that the lack of talking is out of respect for each other’s “deep work”. There were, however, some forms of collaboration, centered around the prototypes: “riffing”, “dropping” and “jams”. Bachmann describes them like this:

It was the prototypes created during jams, in particular, that seemed to have a specific quality: In contrast to the prototypes for the member’s own projects, the goal of these prototypes was not to eventually get “ready to ship”. They also weren’t “proofs of concept” meant to substantiate anything or “demos” meant to impress people. These prototypes were “working artefacts” and “traps for potentialities”, a form of poking in a “space of possibilities” Bachmann, 2019a, p. 284; citing Lucy Suchman and Alberto Corsin Jiménez: “It was about speculating together and in the form of artifacts, in order to gain some intuitions for properties of the new medium.” ibid. [translation mine]

An important aspect of the prototyping culture was that all of the prototypes were documented through their email announcements and the subsequent chain of reactions.

Bachmann describes that during this time there was a sense that their primary work at the moment was “scaffolding”. He points out that there were three distinct (and somewhat conflicting) notions of what that meant: On the one hand, their prototypes were the scaffolding, and they were hoping to develop a sense of the medium as it exists in-between that, much like scaffolding is erected around a house to be built. On the other hand, scaffolding could mean that they used technologies that were opposing to the principles of what they were aiming to do because it was “only” scaffolding. Lastly, scaffolding could also mean that they were developing tools, with which further, more advanced tools could be developed, in the same way, scaffolding is stacked on top of itself.

Rising Tensions — Towards Bootstrapping

Be that as it may, most of the members’ time and energy went into their individual projects. After two seemingly productive years, tensions started to arise.

One of the things that became an issue was a trend one could observe in the general “tools for thought” community during those years too: Bret Victor’s early talks had worked too well. His impressive demos and strong case for his personal principle had indeed inspired people — to try and build similar things, or variations of his demos. These prototypes often were focused on improving the way software is created, or on (the creation of) explorable explanations — both of which are very interesting and important topics, and many of the prototypes built by members of the group during that time have since become inspiring software projects themselves For example “Apparatus” or “Flowsheets”. Some more can be found on the CDG’s GitHub.. But they were not necessarily steps towards a dynamic spatial medium.

There were some early prototypes that made some progress in that direction. In particular the “big board” — which contained the idea of code printed out on sheets and information projected around it — and the “room system” — which connected several prototypes throughout the room —, later further developed into “Hypercard-in-the-World” — which combined these ideas into a prototype that is already clearly a precursor to the Dynamicland known today Glen Chiacchieri’s writup on Laser Socks, a game for the system developed in 2015, contains a description and videos. — seemed promising.

However, these prototypes failed to gain natural traction with the other engineers in the group, who kept working on their projects (using their screens, rather than the spatial medium). This led to a crisis in the group — in early 2016, Bret Victor had grown frustrated: After two years there were prototypes for lots of different things, but the few that moved towards a dynamic spatial medium weren’t at a point where they could be shared with others, and it seemed like the other engineers weren’t intent on getting them there. Writes Bachmann: “Victor grew even more desperate: He was unable to assert himself within the group, and he believed that his external prestige was suffering because he still had nothing he felt he could show for his work. All the prototypes were either for something other than a spatial medium or were so sketchy that he did not think they had the necessary features to substantiate his ideas and authority.” Bachmann, 2019b, p. 44 Victor felt that a change was necessary. In order to make progress towards a dynamic spatial medium, they had to start taking “bootstrapping” seriously.

A Staircase Towards a Dynamic Spatial Medium

In April 2016, Victor sent an email announcing that there would be some far-reaching changes to the way they worked. He introduced two new metaphors: that of the staircase that they were trying to move up on, and that of the stack that they were trying to drill down into.

Victor declared that they’d have to come together to develop a “platform”, which would be the first of a “staircase” of such platforms, which would eventually lead them “up” towards the dynamic spatial medium. Once they had built this first platform, they could return to a phase of experimentation in order to figure out what the next platform, the second step of the staircase, might look like. By building platform on platform on platform and so on, they’d finally live up to the ideal of bootstrapping they’d aimed at from the beginning and have a chance at eventually reaching the dynamic spatial medium.

The platform we’re building this year is not the ‘dynamic medium’. The platform will be ‘technology’, not timeless or transcendent. But it should make possible the exploration and perhaps invention of new kinds of rep­resen­ta­tion-for-understanding, which will then make a bit clearer what the medium for these rep­resen­ta­tions should be, which will then inform the design of the next platform ad transcendum. Bret Victor describing the platform principle in personal communication from April 2016; cited from Bachmann: “Strapping und Stacking” (2019a, pp. 287 – 288)

But this would require all of them to spend one year solely focusing on building the platform, rather than working on their individual projects. Victor believed it was the only way to go and reminded them of the Dynamic Medium Group’s proclaimed goal: “Our goal is to invent an in-the-world dynamic medium, capable of hosting in-the-world dynamic representations of systems, which enable the people, who need to understand these systems to mutually understand them together.” Bret Victor in personal communication from April 2016; cited from Bachmann, 2019a, p. 288 Putting their individual projects on hold to invest their energy into this shared goal was a big ask, so Victor framed it by formulating a new, higher aspiration for the work of the group: Their motivation wasn’t the invention of the dynamic spatial medium for its own sake, but to save the world.

Bret Victor’s “Yak Stack”

In his email Victor explained his reasoning by describing the metaphorical “Yak Stack”:

On a given day, you might be designing a messaging protocol to implement a pub/sub system to make a network of IMPs
to make sensor data easily accessible throughout the room
to implement an object model, based on observation
to enable authoring of dynamic media in-the-world
to represent complex systems that used to be unseeable
to understand complex systems that used to be ungraspable
to introduce new forms of human thought and communication
to expand humanity’s thinkable territory
to escape tribalism and lift humanity to a sufficient level of enlightenment that technological power can be used responsibly, not destructively
to prevent the world from tearing itself apart Bret Victor’s “Yak Stack”, described in personal communication from April 2016; cited from Bachmann: “Strapping und Stacking” (2019a, p. 289) [highlight mine]

The goal is not to get lost in the day-to-day business on the top of the stack, but to keep the eyes on the motivations in the lower part of the stack.

Victor himself points out how the final part of the list is prone to be seen as being ridiculous: “The items at the top of the yak stack are easy to talk about, they lend themselves to lunch chat and status reports.”, he writes, “The items at the bottom are abstract, hard to articulate, emotionally charged, unfashionable, easy to mock, even embarrassing. At least, I often feel like I can’t discuss these things with anyone.” Bret Victor in personal communication from April 2016; cited from Bachmann, 2019a, p. 289

Megalomaniacal Techno-Solutionism?

He is right with this diagnosis: In a radio interview with Bachmann, the journalist somewhat smugly references this passage as being equally cocky and naive, not that different from Silicon Valley’s mantra of “make the world a better place”, which is usually shallow and capitalistic. But Bachmann responds by offering a more nuanced perspective: Doesn’t almost everybody who got to choose their profession freely claim to aspire to a somewhat naive higher goal? As a professor at a university, he has to believe in the values of the enlightenment and believe that his students will become better people through the education he gives them. Of course, he knows full well that this is a romanticized and idealized perspective that does not reflect what is really happening. But even while being aware of this, he still has to somehow believe in it one way or the other. Doesn’t the journalist believe, I might add, that his work is in service of truth versus structures of power and is integral to a functioning democracy, even though what he is actually doing is interviewing an anthropologist about his work on a group of engineers for a general interest story?

I’d be lying if I said that the logic reflected in the stack doesn’t appeal to me personally in a way.

The essence of the stack is often understood as saying: “I want to save the world and I figured that becoming a technologist is the best way to do that.” — and thereby on the one hand looking down on the work of other people, who might have a much more direct impact on their community, on the other hand saying: “My technology will save the world”, or even worse “I’ll be the one that saves the world”. But that would be reading the stack from the bottom up in a form of reverse reasoning. A more accurate way of phrasing it is this: “I became a technologist because it is something that I am good at, that I enjoy, that pays the bills — in this capacity, what can I do to leave the world a better place than I found it?” A question one absolutely ought to ask oneself, I’d say. It will not be answered by the stack, but the stack is a call to remember to ask this question more often in our daily work.

I am also wary of the superficial critique of this being techno-solutionist thinking: Of course, thinking that (a specific) technology will be or directly lead to the solution to all our problems is naive — but thinking that technology won’t be any part of the solution is equally naive. In particular if you understand things like language as “cognitive technology”.

Bachmann describes the group’s worldview as being radically pessimistic — the world will tear itself apart when we don’t do something — while being desperately optimistic — we might be able to actually prevent that if our ideas work.

The First Generation Leaves

But of course, Victor’s introduction of his “Yak Stack” was also a power move, meant to substantiate his decision to focus on building the “staircase” at the expense of the other member’s personal projects. Who could say no to saving the world?

Well, apparently most of the members of the group could — except for Toby Schachman, all of the members of the first generation left the project during this time. The reasons varied: some felt that they were asked to be mere assistants working on Victor’s ideas, rather than being able to contribute their own. Others started to doubt that the dynamic spatial medium was the thing that would enable them to reach their lofty goal. Bachmann mentions that one engineer decided to stop being an engineer at all, thinking that this was not the best way that he could contribute to “human happiness”. Here he is most likely referring to Glen Chiacchieri, who wrote an insightful, emotional blog post about leaving tech­nology to become a therapist.

Building Dynamicland

Over time, the second generation of members joined the group and they began working on the first “platform”, which turned into Dynamicland.

Once again there were conflicts within the group regarding the exact nature of their undertaking: “Some engineers emphasized, like Schachman, the social aspects of the emergent medium (often described with the adjective ‘together’), others, like Victor, the ability of understanding complex systems, others, like Paula Te or Luke Iannini, were particularly interested in the medium’s political potentials.” Bachmann, 2019a, p. 297 [translation mine] But “[…] in time, a fragile compromise was found. An important factor was that Victor now permitted — initially to a large extent against his beliefs — the new spatial dynamic medium to have more objectives than just establishing new ways of thinking. A new objective was to promote new types of ‘togetherness’, not only as a means for common understanding, but as a goal on its own. The research group also prevailed with their wish of involving other users outside of themselves, such as kids. The group increasingly started to speak of themselves as a ‘community’. These compromises remained fragile, but they were sufficient to enable the group to work together on an especially demanding part of their endeavor: the construction of a technical system that would finally be a prototype for exploring a dynamic spatial medium, a bootstrapping system that would shape their work, and a demo for the principle-driven stance of the group and the newly achieved compromise regarding ideas and sense-making processes.Bachmann, 2019b, p. 44 [highlight mine]

And so they started building their operating system, called “Realtalk”, based on the ideas of the “Hypercard-in-the-World” prototype. The core idea was to take object-oriented programming literally, by turning the objects in the system into actual physical objects (or the other way around). After a while, Realtalk was built in itself recursively, allowing for constant open manipulation and adding another aspect to the idea of “bootstrapping”.

Realtalk became operational in the summer of 2017 but had to be rewritten because the first version was too slow. The second version, which was more minimalistic in some aspects, was able to deliver on its promise — Dynamicland was born.

Since then, a continuous flow of visitors was invited into the lab to play with the prototype and build their own ideas using the system, or often “riffing” on other people’s prototypes. Pages of code are kept in binders that are available as libraries to build more things on. The new forms of social interactions, togetherness, and spatial collaboration seem to actually be emerging.

It was around this time that Bachmann last visited the group and apart from enthusiastic reports on Dynamicland from visitors, little information about the group’s internal goings-on has emerged since. Bachmann reported that a process of an alternating cycle of one year of playful exploration, followed by one year of working on stabilizing the system has been established. Carl Tashian writes that in July 2019 work on Realtalk-2020, a new iteration of the platform started. The last tweet from the Twitter account of the Dynamic Medium Group is from October 2020 and states that they continue working on the project — quietly. @Dynamicland1 on 21. October 2020 responding to someone asking if they have become inactive: “No, just quiet. In December 2021 a private letter was sent to friends of the group, announcing the next phase of the project.

Publications by Götz Bachmann

Publications by Members of the Dynamic Medium Group

Notes from Visits to the Lab

Most of the visits shared here seem to have taken place in 2018.

Interviews with Members of the Group

Other

Imprint