It is my understanding that HARC has stopped funding Victor’s [1] and Hart’s [2] research.
Is that accurate? Have other groups in the list suffered from the same fate? If so, how does that fit within HARC’s long term research strategy? These sort of initiatives rarely bear fruits in a single year.
[2] I was at Dynamic Land earlier this week, and they said they were closing down for the winter, actively looking for funding, and not reopening if they cannot provide said funding.
I'm struggling to see how why Dynamic Land would be no longer be funded by HARC. Doesn't it fit the HARC goal of: "HARC’s mission is to ensure human wisdom exceeds human power, by inventing and freely sharing ideas and technology that allow all humans to see further and understand more deeply."[1] pretty well?
Maybe HARC is just a year-long thing? If so Bret/Dynamic Land team can you please put up a Patreon account and let us fund you. I will gladly give money to you as it is the most human form of computing I have ever seen and would like progress on it to continue.
Realtalk/Dynamicland is fascinating to me, though there's something about it which I haven't been able to get over, which is: how can it provide the kind of generality we're used to in computing without giving some set of re-combinable primitives? In other words, something like pixels where the primitive thing is simple and easy to configure algorithmically, and yet the range of configurations is vast enough to represent a large fraction of human imagination.
I love 'being in the world' as opposed to sitting statically at a desk—but I would need something that could give me the similar feeling of vast possibilities when I consider what I'd like to build next.
I wonder if some kind of re-combinable primitives for representation are part of the plan, or if it's an explicit aim of the project to avoid them—a kind of stance against abstraction, reduction, and uniformity. That's part of why it's so fascinating to me: I can't see a general enough system being built without that—but if it were to succeed, I'm fairly confident it would be one of the most interesting systems I'd ever encountered :)
Edit: I should add: this is from what little I actually know about it. I've just seen a few few videos, read the blurb from the article, and read another article on the laser socks project which talked some about Realtalk.
I saw them both a while back, and reviewed the poster for Seeing Spaces just now. However, I haven't been unable to find anything that addresses the issue I brought up.
I agree very much with most of the general principles Victor talks about (you can see my own extensive work on making program behavior visible here[0] for example)—but when it comes to the space concept, I have difficulty finding something concrete about how the space would be used beneficially. I get the flavor of this idea that your workspace should be 'extension of your body', but at this point, years after becoming familiar with it, I'd like to get more concrete on the subject. Dynamicland/Realtalk certainly are more concrete—but my thoughts on it are reflected in my first comment.
I'm pursuing similar lines myself, and am only able to work in spare moments due to lack of funds, so I'm in no position to donate.
> it's foolish to speculate on the possibilities of a new medium on this extraordinarily limited medium
I'll be happy to be shown otherwise, but I haven't yet found any counter-example to the idea that the 'verbal medium' is sufficient to describe any principle. So my original question still remains: is there some principle which would allow them to achieve similar generality to traditional computing without using some kind of re-combinable primitives?
Edit: to be more specific: without that, how could this ever compete with AR? People could build their own spatially interesting, room-scale workspaces without having to find physical materials to build every application out of. The fact that it can deal with actual physical things has an appeal to it, but I can't see any way of escaping the tradeoff of a massive loss in generality.
"How could the written word ever compete with pictograms?" It's a different medium with real benefits and tradeoffs. Right now for you it's all theoretical, but one day hopefully you'll be able to see for yourself if the medium has the time to grow, breathe, expand, and develop. Funding Dynamicland means that we as a species have the option of seeing how that medium might work if it was allowed to flourish.
Thanks for the link. In that paragraph (the one that was removed), there are two links to Bret’s website and both of those are now 404 as well... what’s going on?
It was the Elevr team (four people) who lost funding. http://elevr.com/elevr-leaving-ycr/ Vi said they could continue for $15,000 per month but couldn't find any backers.
Interesting. Is there anything more to the story? Why were they accepted into HARC, and then let go after such a short period? Has this happened to other people funded by HARC as well?
Just a note that there is an unclosed quotation mark in the "physics" link in Vi Hart's section, causing it to not show and it also breaks the "hyperbolic space" link.
Incredible work from an incredible team! Not sure why Bret Victor's section disappeared.
While I know so many programmers are vehemently opposed to "visual programming" I wonder if we can make programming multimodal. Such as a programming language that supports both visual and text modes, where both are first class citizens to the language. This would allow us to use visual tools to see macrostructure; a place where text falls short. But still let us have our tools for the micro-editing that is modern programming. I know some work has been done in this area: http://ravichugh.github.io/sketch-n-sketch/.
> While I know so many programmers are vehemently opposed to "visual programming" I wonder if we can make programming multimodal.
"Visual programming" so far as been about visualizing program's static structure. Scratch and Labview do this, and this has some advantages, especially for beginners. But there's never been much advantage in understanding program behavior from seeing its structure, and I think tools for understanding program behavior are probably the most effective way of making all programming better. See my project Flowsheets for some experiments in this space: https://www.youtube.com/watch?v=y1Ca5czOY7Q
i am firmly of the belief that this is possible and the right direction to go. i primarily use labview, which is arguably the best general purpose visual programming language. there is a lot of power in the visual approach, and i think it could be improved and extended.
as you say, many people are allergic to visual programming but often due to tunnel vision and short sighted reasons. but the visual paradigm is a step in the right direction in terms of seeing and interacting with your program rather than just writing it.
Is that accurate? Have other groups in the list suffered from the same fate? If so, how does that fit within HARC’s long term research strategy? These sort of initiatives rarely bear fruits in a single year.
[1] http://elevr.com/elevr-leaving-ycr/
[2] I was at Dynamic Land earlier this week, and they said they were closing down for the winter, actively looking for funding, and not reopening if they cannot provide said funding.