This part fits the problem, this part fits the person.

September 5, 2024 10:42 AM

After a more than a decade of work (previously and previously on the blue) Bret Victor has unveiled his vision of a Communal Science Lab and a tactile, collaborative computing-without-computers environment called Dynamicland.
posted by mhoye (12 comments total)

Yay! Shawn Douglas has been working on this for a long time, too. Glad to see it come alive.
posted by lalochezia at 10:47 AM on September 5

The "unveiled" link does an astonishingly bad job of explaining what Dynamicland is.

Like, I get the gist – it had something to do with human-computer interfaces which bridge the gap between the physical/tactile and the virtual.

But exactly what has this dude created in service of that vision? Software of some kind? Hardware of some kind? Standards or protocols for integrating same? An operating system (I saw the word "RealtalkOS" in one shot)? Like, what? And can I download it or something?

I'm sure this information is buried somewhere in the links, but dude does a disservice to his project by not putting it front and center.
posted by escape from the potato planet at 2:08 PM on September 5 [2 favorites]

I am imagining the look on my facility director's face when I explain to him that we need the lab wing remodeled to provide room for giant tactile models of cell surfaces.

Not saying this is a terrible idea in theory. Saying that in practice, it's hard enough to fund science, let alone fund the infrastructure for science. Especially in government work. Would I like a "dynamic floor" so we could look at simulations of nanoscale molecules? Sure, but if I had the money for that, I'd use it to replace the HVAC system that was installed in 1964 and should have been replaced 40 years ago

(In summary, please lobby Congress for more money towards federal research infrastructure, k thx)
posted by caution live frogs at 2:20 PM on September 5 [2 favorites]

Dynamicland is very much a research project, I think of it as a modern Bell Labs doing thinking on how people could use computers in the future. This PDF describes how it might apply to a research lab. I can't find the video right now but they are actively working with scientists on using this in real labs, they used it to help with pipetting showing what went where. They used it with cardboard "molecules" having the computer show them what different combinations would look like.
posted by macrael at 2:46 PM on September 5

Their FAQ entry "is RealTalk open source?" does not inspire a lot of confidence in the actual realness of the system.
posted by Pyry at 2:47 PM on September 5 [3 favorites]

This guy is completely fascinating. I remember the earlier posts, and I still yearn to have a setup like this in my house. I get why it's compelling, and the thing is it seems possible. Indeed if you click around the site you can see some details on how they do the actual projecting (standard lampshade on a mic stand with a projector and camera) and what it looks like to use as a computational system (it's frequently a complicated multi-monitor setup where one of the monitors is a piece of paper or a table, but there are moments where he's pointing a laser pointer at a physical album of images to copy-paste the ones he wants and I get it-- imagine knowing exactly what you were looking at...) It seems feasible! You'd need a hell of a printer and some very nice laser pointers, and a willingness to code. But for something with such an anarchic, communal-ownership-of-knowledge ethos, why is the only implementation of this as a partnership with one particular lab? If he wants this to be the future of technology, why can't I try this?

He mentions on the site that people have tried to fork or imitate Realtalk in the past-- does anyone know where any of those projects have gone?
posted by peppercorn at 2:48 PM on September 5

Their FAQ entry "is RealTalk open source?" does not inspire a lot of confidence in the actual realness of the system.

Realtalk on GitHub, if such a thing were even possible, would open it up, both technically and socially, only to the computing culture that we’re trying to create an alternative to. There would be no opportunity for hands-on coaching or immersion in a community, and most people would use Realtalk in physical isolation (like a personal computer) instead of in a communal setting. Their resulting work would redefine “what Realtalk is”, and would eclipse the delicate and difficult progress that we’re attempting.

puts head in hands. so in order to create the ideal environment for the slow, human inculcation of a culture of norms as essential to your vision as the tech itself, you [checks notes] closed to the public fully and partnered with a biotech lab
posted by peppercorn at 3:07 PM on September 5 [3 favorites]

...partnered with an academic biotech lab to show one implementation of embodied computing in science

a biotech lab whose traditional-computer stuff is all 1,2 open source, and on github, including the most widely used design program for dna nanotechnology in the world.
posted by lalochezia at 5:51 PM on September 5 [2 favorites]

I stumbled on this yesterday while looking for 100 Rabbits. I love how every book on the bottom shelf goes to a page about that book - when I first heard of the internet I hoped it would be like that.

But yes this guy sure hides his light under a bushel.
posted by unearthed at 7:10 PM on September 5

As someone with esoteric interests, I recognise when communication is carefully structured to dissuade the wrong kind of person. Unfortunately, trying to make sense of these materials, I feel like I am the wrong sort of person. I think I would like what Victor is trying to do, but I am only getting a hazy idea of it and I find the site design off-putting and difficult.
posted by i_am_joe's_spleen at 8:31 PM on September 5 [1 favorite]

I'm trying to avoid the computerist fallacy of expecting this to solve All Problems to be valid as a computing paradigm, and I appreciate that they're acknowledging the hard social problems of computing that most ostensibly liberatory computing technologies elide. That said, I think I'd need to get my hands on this to understand what its purpose really is.
posted by jy4m at 4:11 AM on September 6

puts head in hands.

I'm deeply sympathetic to Victor's argument here, in particular as elaborated in his "future of computing" talk. Reducing whole new ideas of computing back into the shape of previously-understood paradigms is inherently diminutive, by which I mean specifically "diminishing them to equivalence of a historical norm" is an exercise in actively avoiding the most important possibilities of the new approaches.
posted by mhoye at 7:47 AM on September 6 [2 favorites]