Dynamicland and the Whimsical Digital Object: Analoguing the Digital | Hacker News Reader

Dynamicland and the Whimsical Digital Object: Analoguing the Digital

cabinetmagazine.org

43 pointsapollinaire10 comments

Good / bad / ugly ...

The good: unlike 20 love letters to Bret's work previously upvoted on HN, this response pays it the respect of considering whether it has inherent shortcomings.

The bad: this essay cites Donna Haraway approvingly, which places it firmly in "critical theory" territory, which for some of us flips the bozo bit. Way too much grad student virtue signaling, and while I genuinely admire her decision to think through Dynamicland's PARC heritage, the OOP bits don't work.

The ugly: the author notes Dynamicland's commitment to "working closely with the Oakland community", then blows it off. Having visited Dynamicland, I can confirm that they are serious about being a particular human space. If you're going to challenge their vision (and we should IMHO), you need to grapple with their actual vision, which is of people gathered in a room working together. No "creepy animism" there.

What do you mean "creepy animism"? You mean like SJW-type stuff? I don't understand.

The phrase comes from the article, and I think it is silly.

"creepy animism" in the sense of "Don't Hug Me I'm Scared"? https://vimeo.com/83687791

Can you explain more about how the OOP bits don't work? Also, how would you challenge their vision?

The article falls in a bit of an uncanny valley, I think. It almost makes sense, and could be more irritating than it is.

To me, the strange thing about this concept is the emphasis on textual code. There is still a screen and a keyboard, they are just projected or on paper. I think I would enjoy more something like a cable-patch interface. I would draw an integrator on a piece of paper, draw a line to a plot window, also just a rectangle on that paper, draw and an incoming signal from a generator to the integrator, then play with that. Something more like Bret Victors previous demos on dynamic exploration and simulation.

I think I would enjoy more something like a cable-patch interface. I would draw an integrator on a piece of paper, draw a line to a plot window, also just a rectangle on that paper, draw and an incoming signal from a generator to the integrator, then play with that.

That's called LabView. Widely used with instrumentation, or for simulating instrumentation.

The trouble with programming by wiring stuff together is that it doesn't scale. It's been tried many times. The Blender game engine worked that way, and you'd end up with a maze the size of a tabletop for a modest game. At some point you need abstractions, and that model doesn't lead to them.

I was going trough wikipedia after your comment, browsing different visual programming languages and paradigms. I suppose you're right that they don't scale in the same way as textual languages. Maybe someone finds a good set of constraints to make them scaleable and the elements abstractable.

Anyway, they do have niche success (if not in general programming); and some of Bret's explorable explanations ideas [1,2,3] seem to point away from textual programming.

[1] http://worrydream.com/#!/ExplorableExplanations [2] http://worrydream.com/#!/InteractiveExplorationOfADynamicalS... [3] http://worrydream.com/#!/LadderOfAbstraction

"Analoguing the Digital" title reminds me of Goodiepal's Radical Computer Music: https://vimeo.com/11152592

Does anyone know when the next open-house for dynamicland? I stopped by when I was in San Francisco, but they seemed to be closed to the public. Writing the website got me no reply.

Reply on news.ycombinator.com