You are viewing archived messages.
Go here to search the history.

Josh Justice 2021-09-13 11:07:19

Hi folks, I am new to this community but I wanted to share an interview that Ted Kaehler (who worked on both Smalltalk and HyperCard) was kind enough to do with me. I asked him about the two systems and how they relate to users being able to inspect and modify their own software.

Some of his insights that interested me most were about the impact of the HyperCard Help Stack on HyperCard’s design itself, the places he says that Bill Atkinson was right in his different approach from Smalltalk, and how a user builds a drag and drop UI and then says “I’m almost done.”

https://codingitwrong.github.io/empowering/ted.html

Kartik Agaram 2021-09-13 13:15:29

Alan Kay wanted to build a “personal computer for all ages.” Just as writing and music are each a kind of “literacy”, so the ability to build simulations and play with them would be a new literacy. Simulations help a student understand anything that changes over time. The spread of an epidemic or the growth of a forest fire are great examples. Every child should become empowered to do simulations.

❤ ❤ This helps with a long-standing question of mine: https://futureofcoding.slack.com/archives/CCL5VVBAN/p1613351462087400

Kartik Agaram 2021-09-13 13:26:55

Each version of Smalltalk was created as a simulation in an earlier version Smalltalk. It was particularly easy to make a new functioning class heirarchy along side the existing one. Smalltalk-76 was quite a bit different from ST-74, and it was vital to build one inside of ST-74. New data formats for objects were easy to simulate. Dan Ingalls got as far as running a simulated version of the compiler for ST-76 to verify that it would work once we “crossed the bridge” to the new system.

🤯

Ivan Reese 2021-09-13 15:04:31

Macintosh User Groups and stack exchange groups played a huge role in popularizing HyperCard.

Is that the origin of the name stackexchange.com?

Shubhadeep Roychowdhury 2021-09-13 14:34:37
Kartik Agaram 2021-09-13 14:50:33

I'm curious to hear more about what aspects of this article you find most interesting. Is there some research agenda that led you to it?

Shubhadeep Roychowdhury 2021-09-13 17:22:48

Well, the most interesting part is to match up the code with the theory in my head. But in general I found the article pretty well written.

Kartik Agaram 2021-09-13 17:46:32

Looking at it some more, I enjoyed the description of the Minsky UTM (LCCP)

Shubhadeep Roychowdhury 2021-09-14 04:20:57

I am thinking of writing a Python or Go version of it.

Alexander Chichigin 2021-09-13 17:11:33

I don't know if everybody already knows this project but I've stumbled upon it just today.

https://sdg.csail.mit.edu/projects/espalier

Now we can say it's a low-code "spreadsheets-on-steroids" platform. I see many features similar to Coda and Notion (but areguably better), and they explicitly reference Airtable in their video which I found pretty impressive and illuminating. It's a shame the project have halted.

Andreas S. 2021-09-14 20:16:50

🐦 💻🐴Ngnghm: In a proof assistant, is it the human assisting the computer, or the computer assisting the human?

Jimmy Miller 2021-09-15 03:43:50

Had to share this incredible chart from Jean E Semmat’s paper “Programming Languages: History and Future” (Not sure if breaking the rules/norms by sharing an image. But look at it!) https://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.86.6078&rep=rep1&type=pdf

Ivan Reese 2021-09-15 04:16:04

Image posts are great. We don't even have concrete definitions for "programming" (I say photoshop counts) or "computer" — so "link" is whatever you want it to be.

Kartik Agaram 2021-09-16 16:06:45

Definitely feels worth updating this format for c. 2020. I think I'd do the categories differently. For example, does Lisp really belong with the other languages in its category? Pretty much every HLL these days supports array/list/map literals. I think Lisp isn't really about "list handling". With 50 more years of context I think cons cells and s-expressions are best seen as a notation for trees rather than lists, thanks to their most common application -- for syntax trees.

Andreas S. 2021-09-17 09:32:39

I'd be very interested in such a thing updated until today. I tried to create something related using wikidata.

Andreas S. 2021-09-17 09:34:28

Here is what I created back then :

Kartik Agaram 2021-09-17 09:35:57

There's been a few like this. What sets OP apart is the concision and information density..

Andreas S. 2021-09-17 09:36:31

It needs to be a computational thing 😄 lets build it 😄

Andreas S. 2021-09-17 09:36:54

*dynamic visualization

Kartik Agaram 2021-09-17 09:37:24

A collaborative wiki that permits controlling both content and visual structure.

Andreas S. 2021-09-17 09:47:27

I would almost suggest a git repository which uses a webpage renderer like gitlab pages?

Kartik Agaram 2021-09-17 09:57:31

I think it'd be nice to have unit tests for this 😄

Andreas S. 2021-09-17 10:12:44

So should we use typescript then?

Andreas S. 2021-09-17 10:13:53

I think one major part is getting data from a source like wikidata, but then create a opportunity to annotate and change the semi(?) automatic graph layout?

Kartik Agaram 2021-09-17 10:14:04

Whoever puts up the first rev chooses! 🙂 This includes your gitlab pages option, btw.

Chris Knott 2021-09-15 17:05:39

Just saw as top on HN https://www.replay.io/

Chris Knott 2021-09-15 17:06:20

Looks like co-founder is here (@Jason Laster)

Christopher Galtenberg 2021-09-15 18:00:56

Def the headline news of the day!

Alexander Chichigin 2021-09-16 08:54:46

On 27-28 of September the Conference on Software Visualisation is coming: https://vissoft.info/2021/program.html

Kartik Agaram 2021-09-16 15:00:49

I'm 4 years late to this thoughtful article which delves deeply into the dichotomy between involving users ("nothing about us without us") and progressing along learning curves ("making things better by making better things"); and the role technology can play in social change (as part of infrastructure, alongside laws, markets and social norms)

https://ethanzuckerman.com/2016/06/22/the-worst-thing-i-read-this-year-and-what-it-taught-me-or-can-we-design-sociotechnical-systems-that-dont-suck

Kartik Agaram 2021-09-16 15:36:16

Some raw material for a patchy map or periodic table about the role of technology in society:

  • "Building apps instead of solving real problems" (emphasizes positive outcomes of tech capability)
  • Accelerationism; progressing along learning/optimization curves (again emphasizes positives)
  • Ensuring teams building tech are demographically balanced (emphasizes process)
  • Ian Malcolm: "you were so busy seeing if you could, you didn't stop to think if you should" (emphasizes negative outcomes of tech capability)
  • Facebook critiques; Tech should take responsibility for breaking things (again emphasizing negatives)
  • Representation in dev teams of end users; "nothing about us without us"

Not sure this is going somewhere, but improvisation ("yes, and ..") most welcome.

Andreas S. 2021-09-17 12:05:16

This is exactly going somewhere Kartik Agaram !

Andreas S. 2021-09-17 12:06:15

Funnily enough I already wrote quite a bit about this in the Alan Kay Interview.

Andreas S. 2021-09-17 12:07:00

Ivan Illich's Tool for Conviviality come to mind: https://en.wikipedia.org/wiki/Tools_for_Conviviality

Andreas S. 2021-09-17 12:09:21

In the Alan Kay thread I also mentioned ICARUS

or

The Future of Science

by

Bertrand Russell

1924

Andreas S. 2021-09-17 12:13:05

I Feel playful so I go with this: "Yes and...for me most of the points are leading down to two basic questions:

  • How do we relate to ourselves
  • And how do we relate to each other
Andreas S. 2021-09-17 12:13:56

What does sovereignty(user vs consumer, user vs programmer) mean?

Andreas S. 2021-09-17 12:16:40

For a individual in the modern western civilization, what does "normal" mean(see Alan Kays talk normal considered harmful)?

Andreas S. 2021-09-17 12:21:17

If we want to tie this somehow to computers the closest thing my zettelkasten hast on this would be this: http://viznut.fi/texts-en/permacomputing_update_2021.html

Andreas S. 2021-09-17 12:22:49

but of course this are complex and language is a slippery slope but still this connects your original inquiry to the broader picture:

Andreas S. 2021-09-17 12:28:19

(think about how Alan kay said that Martin Luther was one of the greatest user interface designers)

Kartik Agaram 2021-09-17 12:45:17

Yes, I've been following and engaging a bit with viznut on Mastodon, and on the Collapse OS mailing list. Though I don't quite see the fine details about how permacomputing connects up with this thread.

Andreas S. 2021-09-17 12:56:18

You see my helpless attempts at communicating something failed. I'll try again, A interesting thing to think about is sovereignty, agency of the individual in the society. the points you mentioned could be described as being part of a different kind of culture opposed to the main stream culture which promotes user-consumer <-> producer relationships.

Andreas S. 2021-09-17 12:57:23

So in an effort to create a culture which employs different values ( agency of the individual) I see perma computing as part of a regenerative culture.

Andreas S. 2021-09-17 12:57:27

Does that make sense?

Kartik Agaram 2021-09-17 13:12:32

Yes, it does. I just needed to go back and reread http://viznut.fi/texts-en/permacomputing.html

Mariano Guerra 2021-09-16 17:31:03

Alan Kay on the context and catalysts of personal computing

https://notionrealized.com/blog/alan-kay

Christopher Galtenberg 2021-09-16 17:43:15

"I think of civilization as being a set of processes trying to invent things that mediate, deflect, turn away, and modify most of the things that are wrong with our brain."

Andreas S. 2021-09-17 09:53:59

Thats quite ... optimistic, maybe even naive?

Andreas S. 2021-09-17 09:54:26

Here are some other people on Technology: http://geosci.uchicago.edu/~kite/doc/von_Neumann_1955.pdf

Andreas S. 2021-09-17 09:56:37

Bertrand Russel also has very good writing on this: http://bactra.org/Icarus.html

Andreas S. 2021-09-17 09:59:18

Alan Kay , he is one of my Heros regarding Computers, but he doesn't dare to go beyond his own culture ( how could he?). He is aas far as one could get from a technical/industrial perspective regarding progressive thinking related to technology.

Andreas S. 2021-09-17 10:00:38

Yet you won'T seem him debate: What is progress? What is technology? - outside of the western civilization narrative, like say for example from an indigenous perspective like mentioned here: https://www.goodreads.com/en/book/show/45449501-sand-talk

🔗 Sand Talk

Andreas S. 2021-09-17 10:03:50

A different perspective being it either from indigenous people or the eastern tradition, like daoism - wuwei principle seem alien to him ( as far as I can tell).

Andreas S. 2021-09-17 10:05:04

Thus he would always look for the right kind of progress, the right kind of technology. Non technology or non progress are non existent in his worldview.

Andreas S. 2021-09-17 10:08:45

"we're in a situation like that's where our progress is cancer"

Andreas S. 2021-09-17 12:30:20

Kartik Agram started a similar thread about the perspective on techology, in there I reference a few other resources that I feel are related to the subject, including 2 from alan kay here is the link to that thread: https://futureofcoding.slack.com/archives/C5U3SEW6A/p1631804449070900

[September 16th, 2021 8:00 AM] ak: I'm 4 years late to this thoughtful article which delves deeply into the dichotomy between involving users ("nothing about us without us") and progressing along learning curves ("making things better by making better things"); and the role technology can play in social change (as part of infrastructure, alongside laws, markets and social norms)

https://ethanzuckerman.com/2016/06/22/the-worst-thing-i-read-this-year-and-what-it-taught-me-or-can-we-design-sociotechnical-systems-that-dont-suck|https://ethanzuckerman.com/2016/06/22/the-worst-thing-i-read-this-year-and-what-it-taught-me-or-can-we-design-sociotechnical-systems-that-dont-suck

Andreas S. 2021-09-17 12:36:47

"I really can't converse very well at my age because I've lost my patience" 🙁

Andreas S. 2021-09-17 13:03:49

Lots of things could be said about this interview, I'll try one more thing: Alan implies many things things like the superiority of writing over oral information exchange or even dialog. There actually people, kind of reinventing dialog. They created a whole bunch of language to talk about certain aspects about the phenomenon of dialog in communication. Really interesting stuff ( cognitive science, If you are interested in that, which I highly recommend, you may search for: John verveake awakening from the meaning crisis)

Ivan Reese 2021-09-16 18:05:51

Nice quick demo of a "tape drawing" UI, where you use one hand to control the position of the pen, and the other hand to control the angle, allowing for precisely drawing both smooth curves and straight line segments.

https://www.youtube.com/watch?v=LvyzwN36PSw

Chris Knott 2021-09-16 19:09:24

I'd love to see this on the Remarkable 2

Chris Knott 2021-09-16 19:11:32

It's interesting that he says "we bring the computer to us, not us to the computer" but actually the user is still making quite a lot of allowances for the computer using floaty light pens with no tension between the hands. I wonder if it might be more bringing the computer to us to simply autofit bezier curves to a photo of the physical tape drawing.

Leonard Pauli 2021-09-16 22:59:05

The VR demo in the comments was neat :)

Nick Smith 2021-09-17 00:48:30

Wow, that's a very cool concept!

Srini Kadamati 2021-09-17 08:33:48

Chris Knott my interpretation was that ‘bringing the computer to us’ emphasizes human embodied skills by letting us adjust the curve ‘inputs’ / parameters at the speed of thought using our hands! Auto-fitting bezier curves to me feels like going back to treating the computer as a magical black box again instead of having both human and computer participate in the process 😄

Chris Knott 2021-09-17 13:19:01

What I mean was if you consider two input devices; the [light pens+screen+projector], and [physical tape+whiteboard+camera], then to me the second one makes better use of human's embodied skills because you can feel the tension between your hands, more intuitively peel off and re-stick mistakes etc. (it does have downsides such as creating waste).

The motivating problem they identified in the video was how to digitise the tape images. In 1999 this maybe required inputting the data digitally in the first place (with hand tools), but nowadays you could also have the option of inputting the data with a camera. The end result is the basically the same; users can draw on the wall with their hands, but end up with a vector image on the computer system.

Chris Knott 2021-09-17 13:20:34

I agree that black boxes are bad. You wouldn't want to be in a situation where the image on the computer is not what you drew, with no way to understand where you and the computer are disagreeing.

Srini Kadamati 2021-09-17 14:05:22

ah I see