Channels
of-ai
in-sf
in-london
in-ontario
in-seattle
in-socal
in-alberta
of-functional-programming
reading-together
wormholes
of-end-user-programming
random-encounters
in-germany
in-israel
bot-dark-forest
of-music
present-company
of-logic-programming
share-your-work
devlog-together
introduce-yourself
thinking-together
in-boston
in-nyc
of-graphics
announcements
administrivia
two-minute-week
linking-together
Title
j
Jan Ruzicka
12/16/2022, 4:57 PMA new Bret Victor’s demo (Bret starts at 14mins; learned about this from @stevekrouse’s tweet):
c
Chris G
12/16/2022, 5:46 PMIn the Q&A, he didn't want to talk about the computer at all (how to program, how it's built)
Thrilled by question about a new aspect of this particular application
i
Ivan Reese
12/16/2022, 8:05 PMWelp, this was super cool and a delightful surprise.
That the video-of-a-projector-of-a-video-of-a-projector effect made it feel like I was hallucinating just added to the impact.
d
Duncan Cragg
12/16/2022, 9:50 PMI'm a bit worried - sounds like he's had a stroke - he is having trouble speaking and breathing! Hope he's OK
Anyway - on the techie stuff - what is in the physical doodads that couldn't be implemented in VR or AR?
i
Ivan Reese
12/16/2022, 11:25 PMhe is having trouble speaking and breathingI think he's always sounded like this. Just in the other talks, I don't remember him ever using a handheld mic. I think if you take someone who's not a confident public speaker and give them a pretty hot dynamic mic, they sound kinda like this. But who knows! Hope he's okay.
what is in the physical doodads that couldn't be implemented in VR or ARTactile feedback? The ability to specify constraints like "these blocks need to snap together according to a certain tiling pattern" using magnets instead of having to code up some constraints? The fact that you can make eye contact with the person across the table from you? I mean, sure, we could eventually figure that stuff out with AR / VR. There's lots of cool potential there. But this DL-esq stuff feels (A) like a much nearer-term reality (ignoring market forces, granted) and (B) like it's pursuing a fundamentally different goal from just "fill the space around you with virtual graphics"
k
Kartik Agaram
12/17/2022, 4:19 AMIf this isn't AR, I don't know what is.
w
wtaysom
12/17/2022, 6:27 AM"I want to clarify here. It's not VR. It's not AR. It's just R." http://vimeo.com/115154289
There's a question about the system crashing at the end, and I did wonder earlier how much of the program state lives in the arrangement of objects on the table? It's kind of a measure of how tangible the system is.
d
Duncan Cragg
12/17/2022, 9:33 AMThese objects will get worn, dirty, lost, broken, and the papers torn, dog-eared and lose their crispness, but humans being humans they will still be in use long after they should have been replaced - even the paper won't be reprinted because of Printer Hesitancy!! And finding the thing to print again. Imagine 3D printing the medical-looking bits when they break or get lost
j
Jack Rusher
12/17/2022, 9:34 AMIshii has done so much great work.
c
Chris G
12/17/2022, 10:03 PMTrue - this platform is so local-first you wonder where the state data actually is - in theory you'd want to be able to replace the table, projectors, all the non-solution components and still have it work - but as Duncan says, if a solution artifact goes away, then what - ideally there's a redundancy somewhere
w
wtaysom
12/19/2022, 1:55 AMLet me add that the domain here (protein imaging) is perfect since the need for a wet lab plays very well with idea of physically extended computers. As for the binders, I always think Dynamicland is actively trying to escape the gravitational pull of the way we usually do things with computers. In that case what affordances do binders have? Maybe it gives you a tangible measure of how big your program is? We could, in principle, go a lot further. For example, in text a big ball of mud and a nicely layered lasagna don't look all that different.
n
Naveen Michaud-Agrawal
12/19/2022, 11:17 PMDynamicLand seems to take an approach similar to Smalltalk, where there is a small kernel of functionality (providing input/output modalities, access to computational time and space) and the rest of the system is built on itself (from the few photos I've seen of the Real talk OS most of the core functionality like parsing, eval, and even output are generated by Page objects)
Is there a systems level description of the overall approach/architecture and the underlying virtual 'machine' that drives the system?
i
Ivan Reese
12/20/2022, 6:02 AMNone that I've seen. The most complete collection of links and project info is Christoph Labacher's translation / analysis of Götz Bachmann's ethnography of DL, so if you wanted to go poking around looking for such info this is where I'd start.
n
Naveen Michaud-Agrawal
12/20/2022, 1:38 PMThanks, I have read that a few times and followed all subsequent links. Unfortunately some of them publications are behind paywalls so I haven't been able to read them.
Also most descriptions are from greater than two years ago, I assume that Bret's 'burned the diskpacks', so to speak, and has rebuilt the system from scratch based on the learnings of the original version.