Browse files

## Rushed BV deep dive

### Quick progress

Quick & dirty progress with little to no note taking was made today on a number of BV things:

* Media for Thinking the Unthinkable
* The Future of Programming
* Seeing Spaces
* The Humane Representation of Thought

As well as a few selected references that he linked to from those (or the notes for those), but you'll have to see my notes in this commit for all that jazz. 

### Remaining things

Thought I didn't get to them today, I think it would be useful to me to read both Bret's _2013 links_ and _What a technologist can do about climate change_ before I show up at Dynamicland. I have plenty of time for both of these things before Monday at 1pm PST, but it's also not a tragedy if I don't get to them.

As far as my BV deep dive goes, I'd say that it's as done, if not more done, than the deep dive I did into Alan Kay. There's more to be looked at for both men, for Alan there are many videos, and for Bret, there are past websites, projects, and journal entries. But I suspect that after next week, I will feel content in my knowledge of Bret Victor for now, and would be more happy spending time with his sources, than with him directly. And of course, re-reading and re-watching is always useful so I can always comes back to this if I need to.

### Self-directed masters program

On a trip to the bathroom, I picked up Richard Rorty's _Philosophy and Social Hope_ and was blown away by the first few pages. He alluded to being able to dissolve [my current paradoxical thoughts on the distinction between inventtion vs discovery](/essays/invented-or-discovered). It feels like Rorty might do for Dewey what Deutch does for Popper. 

Anyways, I have a stack of books in front of me ranging from Piaget to McLuhan to Raskin to Vygotsky to Brooks to Rheingold to Minsky to Hofstader. And I have at least another dozen such books elsewhere that feel like essential reading for me to live a fufilled life. Thus had the thought to structure all of this reading and reflecting into an organized structure of some sort, which I am currently referring to as a "self-directed masters program." Here are some of the tasks (unordered) that represent my current thinking of the next steps here if I choose to continue planning in this direction...

* look into how other masters' programs, PhD programs, or other programs of study are organized, such as [Open Masters]( and [Steph Jang's post](

* think about the topics / questions / goals (in other words, why do this over getting a job or starting a company, or any other possible thing?)

* will there be a thesis or end product, one or many?

* mentorship?

* physical location?

* time

* budgeting
  • Loading branch information...
stevekrouse committed Jan 12, 2018
1 parent 4c09562 commit 74851490e6a04231cccdd4c96c28bfb884c92c52
Showing with 133 additions and 5 deletions.
  1. +1 −1 _data/git-log.json
  2. +130 −4 notes/bret-victor/
  3. +2 −0 notes/bret-victor/

Large diffs are not rendered by default.

Oops, something went wrong.
@@ -86,34 +86,160 @@ Vision without implementation is hard to imagine. I reacted negatively to this p
## _April 4, 2013_ - [Media for Thinking the Unthinkable](!/MediaForThinkingTheUnthinkable)
I've seen and thought about these ideas and examples sufficiently to be not terribly impressed with my nth viewing of them in this video. Probably the most exciting thing I found was Bret's work visualizing Nile:
I of course came across this system in my deep dive of Alan Kay, but as with most mathmatical inventions, I found it intimidating and unapproachable. Yet, potentially these sites will make it easier for me to gain an ituition for it. In a more general sense, this makes me wonder about all sorts of mathmatical notations that intimidate me, but probably hold all sorts of beauty, such as maxwell's equations (which Bret often references) or the abstract mathmatics that inspires functional programming, such as category theory. I guess this sort of thinking is what inspired Chris Olah to explain Machine Learning.
### [An Ill-Advised Personal Note about "Media for Thinking the Unthinkable"](
> Any concept, technique, or tool that is specific to software engineering is guaranteed to have a short shelf life, at least on any time scale that I personally care about. (Which is totally fine if you’re into that, but this is my ill-advised personal note, not yours, and I personally care about mattering 100 years from now.)
Wow, that’s fascinating. Makes me wonder about where this desire to attain a lasting legacy comes from.
## _April 26, 2013_ - [Stop Drawing Dead Fish]( -
Beautiful. Yet unclear what problem it solves for whom.
## _May 2013_ - [Drawing Dynamic Visualizations](
Beautiful. Reminds me both of JoyJS and and [Programming with Comics]( Crazy how many tools this man inspires!
### [Additional Notes on "Drawing Dynamic Visualizations"](
Haven't read.
## _July 2013_ - [The Future of Programming](
This video has one theme and 4 points. The theme is that people are often slow to move from one technology/representation to a better one, but that we should keep striving for better abstractions.
The four points are that:
1/ We should move from symbolic coding to the direct manipulation of data
This seems reasonable, yet of course as Bret would agree, symbolic reasoning is powerful too.
2/ we should move from lower level to higher level instructions to the computer
Nothing that inflamatory here, yet its implecations are far-reaching. For example, variables are way too low-level. Listing instructions -- sequential programming -- also seems too low level, which he touches in his 4th point.
So much of what makes programming difficult for beginners, as team behind Eve figured out, is scope. This is not a low-level concern, but a interface design concern.
3/ Sequential text should move towards richer visual displays.
Pretty straight-forward in principle, yet we haven't figure this out in practice -- potentially because we haven't found the right "glue" to piece non-textual programs together. Copy and paste and Unix pipes are powerful.
4/ Sequential instructions should move towards a concurrent model.
Funny how so many of these goals point towards functional programming, yet how few of Bret's demos look like functional programming.
I guess [the Lambda Alligator]( do but I never got those...
### Langauge vs Tool
Just re-read Lamda Alligators - still above my head - but I found [this neat Oliver Steele article that explains that some people are more IDE-focused and others are more langauge-focused](
Reminds me of Paul Chiusano's talk on whether langauge or tools is higher leverage, i.e. would you rather have a great brain-to-binary interface or emacs-to-Haskell. However, now that I think of it, that's a misleading analogy, because just being able to "mentally type" assmebly really quickly is basically just really fast fingers in a text-editor. What Paul misses with this analogy is that what's powerful about building better tools is that they can help you see things that you can't see with just text, such as a debugger or other of Bret's demos in visualization. It's a fine point - because at what point does the langauge stop and the tool begin? I feel like I am personally a langauges guy who is trying to build tools as powerful as langauges.
One final note about this DBX talk. One of the most memorable parts of this talk is when he ridicules the idea of an API. Clearly these programs will negotiate how to talk to each other.
### [References for "The Future of Programming"](
This is a reassuring note for me at this point in my thinking:
> Lastly, here's some advice Alan Kay gave me (as I was going through a small personal crisis as a result of reading Jerome Bruner's "Toward a Theory of Instruction"):
> I think the trick with knowledge is to "acquire it, and forget all except the perfume" -- because it is noisy and sometimes drowns out one's own "brain voices". The perfume part is important because it will help find the knowledge again to help get to the destinations the inner urges pick.
As I wrote in the third paragraph of [this section of my journal yesterday](
> [Reading Tufte without] any immediate concrete, practical use for these skills [...] speaks to the question of pre-loading vs just-in-time learning. I feel like most of the value in my reading today will be in knowing where the answers to questions about data visualization will likely be found – and that there are answers here at all – rather than the actual rules or techniques that I read given that I am likely soon to forget them in the absence of applying them anytime soon.
There are way too many references in this talk! Even if I had more than just the rest of today, I'm not sure I'd want to read them all. Some of the ones that seem promising.
TODO - find a place for these links:
* [Alan Kay -- Programming and Scaling](
* [Gerry Sussman -- We Really Don't Know How To Compute!](
* [Alan Kay: Doing with Images Makes Symbols Pt 1](
* [Alan Kay: Doing with Images Makes Symbols Pt 2](
* [J.C.R. Licklider (1960) -- Man-Computer Symbiosis](
* [Tony Hoare: Communicating Sequential Processes](
* [Danny Hillis -- Richard Feynman and The Connection Machine](
* I _really_ need to at least watch Engelbart's "mother of all demos," even if I don't have time to make it further than the halfway through the paper I got. Both can be found [here](
## _Dec 2013_ [Links 2013](!/Links2013)
TODO this feels valuable
## _June 2014_ - [Seeing Spaces](
Makerspaces++. I didn't realize he explained it so explicitly in terms of Makerspaces.
Makes me wonder if there's a business to be built in consulting with schools on the design of science, math, and engineering classrooms. Maybe I wouldn't hate teaching robotics so much if it was in such an environment. In other words, it's Lego Mindstorms++.
> representations [such as Playfiar's data plot and Heaviside's equations] weren’t mere scientific “discoveries”. Each of them essentially enabled all subsequent scientific breakthroughs thereafter. A powerful new form of representation affects everything, forever.
Damn. That makes me feel a lot better than a mere "user-interface designer."
Dynamic conversations could lower the miscommunication between people, encourage empathetic connection, understanding, and more directed, and accurate discussions. Imagine if we could have meaningful conversations casually. We currently cannot just because verbal words are so powerless as compared to written mediums.
Imagine museams that are build with dynamism in mind. Or libraries. Woah. Spaces that allow you to really explore ideas with your body.
Dynamic Objects-To-Think-With. This makes me wonder why we don't have children use a clay or legos of some sort to learn about mathmatics beyond the younger years.
> Every one of these projects is about designing a thinking medium to fit the human, instead of deforming the human to fit the medium.
This quote is another angle to view the floor vs ceiling distinction. The langauges with a low floor fit the human well but not the machine, and usual the reverse is true with langauges with a high ceiling.
> The culture can start to develop a dynamic multimodal literacy.
Wow, this is a fun cause to rally behind: _deep literacy_, in many different ways of thinking, mathmatical, scientific, systematic, iterative, with your body, with symbols, with your senses, etc, etc.
## _Oct 2014_ - [Humane Representation of Thought](
## _Nov 2015_ - [What a technologist can do about climate change](!/ClimateChange)
Thinking is very phsyical. We think in our muscles.
He makes it very clear in the intro to this talk, and the following paragraph, that he is abstracting over the past few years of talks and demos to figure out what it is that he's trying to accomplish. This is profound for me because in _Inventing on Principle_, he sounds pretty sure about his mission, but it seems like he is still working on making it more and more clear:
> It's based on a piece of writing that I started over a year ago, a long-term research agenda (and floor plan) for myself and my then-nascent research group, an attempt to unearth the demons that have driven my work over the last decade, and draw a map of the destination they're trying to get to.
Science and engineering _fundamentally depends_ on the Playfair data plot. It was so powerful because it repurposed the ability of our monkey eyes to think more abstract thoughts.
> Lebniz was the UI designer of the 17th century.
The periodic table was the foundation of chemistry!
It would be a fun research project to find the representations that enable certain breakthroughs to happen. I imagine that independent simultaneous invention would be a good clue that it was casued by a new medium.
Bret's motivation is the growth of intellectual ideas, the ascent of man.
Inhumane. The analogy of putting a puppy in a cage is so memorable. I think abou this all the time, and use this analogy to communicate with everyone.
I want to promote Jerome Bruner in my list of reading, given how important Bret thinks he is. I guess I'll also want to do Kieran Egan... just bought!
Wow, the MIT Media Lab [inFORM]( project is cool!
The most exciting category of representations to design is for abstract ideas, such as for mathamtics or systems.
> Humane is never the default. The incremental thing will lead you to a tighter and tighter cage. If you believe in the humane medium, it's up to us to fight that trend.
### [A note about "The Humane Representation of Thought"](
## _Nov 2015_ - [What a technologist can do about climate change](!/ClimateChange)
TODO this feels important but don't have time for this now.
## [Quotes](!/quotes)
Haven't read.
@@ -9,6 +9,8 @@ In _Learnable Programming_, you mention "Long and careful thought was given to t
How do square your admiration for functional programming with your desire to "kill math", given that FP looks a lot like the math you want to kill? (My guess is that he likes the abstractions of FP, but not the notation, and conversely he likes dislikes the notions of mathmatics, not the abstractions.)
You often reference, directly or indirectly, how many of the principles of functional programming could improve programming generally. Yet few, if any, of your talks, papers, or demos, seem to embody those principles. Why is this?

0 comments on commit 7485149

Please sign in to comment.