Recording of today's This one featured a demo of LabView + AI from @Jim Kring, a smalltalk-esq browser for Lisp from @Konrad Hinsen, plus a nice discussion about collaboration within the community led by @Kartik Agaram.

Here's my chance to comment my own presentation, with something I forgot to say: I'd really love to have something like my Common Lisp inspector for JavaScript, in the browser. Something going beyond the typical element inspector in letting me define custom GUI representations and interaction elements for my code and data. Maybe that something already exists, e.g. as a browser extension. Any ideas?

@Kartik Agaram - FWIW: first impressions, while I can still remember them, completely orthogonal to your talk: a) When I saw your diagram, I immediately thought “this is a Kinopio diagram”. I.E. like brainstorming, flying off in multiple directions. One thing that attracts me to Kinopio is that it lets you elide busy-ness by linking to other diagrams (the “/“ operator) and replacing whole sub-diagrams by a single rectangle (albeit, manually). (http://kinopio.club) b) More agency over computers. My first impression is that you were talking about /all/ people. I don’t believe that /all/ people want more agency over their computers - if anything they want more agency over their own problem domains. Programmers want more agency over computers, dentists’ receptionists want more agency over scheduling and patient records.

As for agency, I see three categories of people: 1. Those who some agency over computers and want more of it (roughly: us here) 2. Those for whom computers are tools like many others and who just want more agency in their domain. They often see computers/software as increasingly damaging to their agency. 3. Those who have no idea that one can actually have agency over computers and who would profit from it if they knew.

The definition of agency is "being able to do what you want to do." It's about optionality. So "people don't want agency over their computers" seems like an oxymoron. You can want agency over computers and also your scheduling and patient records. It's not either/or. Or in Konrad's framing I believe in categories 1 and 3 but not 2. Imagine you're a peasant in medieval Europe or Japan. Would you like to have agency over armies or consider armies damaging to your agency? I have to think the answer is both. You can think realistically that you have no agency over Facebook and still fantasize about being king of Facebook.

Agency is not static, we can increase our agency in various ways: learning, acquiring material resources, acquiring power over other people (directly via hierarchies, or indirectly though money). My three categories express different levels of knowledge about computers and how they can increase someone's agency. Category 2 is people whose agency wouldn't increase much if they had more control over their computers.

Hmm, interesting. I don't quite follow this but this is good. Can both of you state the problems you care about? That might clarify things more. @guitarvydas feel free to use Kinopio to do it 😄 (ability to inline diagrams does seem very useful)

My main issue is transferring a big dose of agency from infrastructure providers (meaning developers of operating systems, programming languages, Web platforms etc.) to users, in particular power users (my category 1). I'd claim (though it might be out of self-interest) is that this would give all users more agency, because even someone who doesn't want to write any code would find it easier to pay a local expert to do some coding for them than to get Google or Microsoft to do anything specifically for them. Note that this is just the IT subset of my generally somewhat anarchist attitude to technology, which I think is much too centralized and un-convivial nowadays.

Ok! I think we're entirely agreed on the latter half, and I now appreciate refining the problem to a specific category. More focus is always better.

Agency: I contend that most programmers cannot understand what “normal people” want to have agency over. Most people don’t want more options, most people don’t really want to control/improve their own processes. They want to control you, instead of their process (“bureaucracy”). For example “My computer screen does not allow me to enter two middle names, even though you say that you have two middle names”. The phone clerk has 2 options: (a) modify and improve their computer screens, or, (b) get you to buckle under and just pick one middle name, so that they can move to the next screen. Availability of “agency” seems to be more of a marketing / re-education issue, rather than a technical issue, the phone clerk does not even consider the possibility of modifying their input screens, they see only one option: get the caller to buckle under and to pick only one middle name.

@guitarvydas Good example for my third category. Technical issues are indeed only a small part of the problem in this case. The main issue is social structures, more specifically hierarchies, that actually require the phone clerk to be powerless about this question to let the upper layers of the hierarchy go on with their business. They don't want to be bothered by people having more than one middle name.

I get the sense there are philosophical depths here I haven't plumbed when it comes to the word "agency". I was trying in the talk to cover a space of problems I see, but unavoidably my picture is still informed by the problems I care about: • Individual people • interacting with individual devices • that don't need to communicate with other devices or people to do their work. The support network I care about is rich relationships between people. So the clerk intermediating for a faceless bureaucracy is a harder problem on 3 levels. My goal is more limited, and I think should be relatable to anyone who has asked or fielded questions like, "how can I change the color of text in Notepad?" Certainly if you care about these harder problems and try to work on them, you should make sure your problem statements include them. So far I'm not seeing the connections in what y'all have said in that thread.

I can't say I care that much about the phone clerk, but I think it's important to understand the role of the phone clerk because it's roles in hierarchies that business software and their support infrastructure is created for. That same support infrastructure powers every mass-market device, so we can't just ignore it.

Conversely, why not ignore it when there's nothing I can do about it? It's a situation rather than a problem, I think.

If you can just ignore mainstream, fine. I am constantly talking to people for whom mainstream is the obvious way to do things, so I have to explain to them why I disagree, and that means I need to understand the roots of mainstream.

Conversely, why not ignore it when there's nothing I can do about it?

Is this not an example of lack of agency?

> Is this not an example of lack of agency? Of course! Agency is never infinite. We don't sit around complaining we can't move the Sun. "Let me warn you, `important problem' must be phrased carefully. The three outstanding problems in physics, in a certain sense, were never worked on while I was at Bell Labs. We didn't work on (1) time travel, (2) teleportation, and (3) antigravity. They are not important problems because we do not have an attack." -- Richard Hamming, https://www.cs.virginia.edu/~robins/YouAndYourResearch.pdf

A point I was trying to make, but didn’t manage to, is that the bar for agency vs. lack-of-agency is based on /belief/. Providing more technical ways to improve agency has no effect unless the subject’s beliefs are changed, also. The given example shows that “anti-gravity” was considered (believed) to be unimportant at Bell Labs, but,,, “anti-gravity” was researched elsewhere (see “Thomas Townsend Brown”) where the belief structure was different. Likewise, it was believed that nothing better than the scissor-kick or roll, etc., could be used in olympic high-jumping until Fosbury came along. Likewise, it was believed that humans could not run a mile in less than 4 minutes until Bannister proved that that belief was wrong. Bureaucrats boil my blood when digging in their heels because I know, in certain cases, that their beliefs about their computerized forms are untrue. Henry Ford is said to have commented that people didn’t know that they wanted automobiles, they believed that they only wanted faster horses. People believed for 1,000s of years, that the Sun orbited the Earth. It took 100’s of years to alter that belief and to get smart people “to look through the telescope”. I was taught that you can only add and multiply matrices, but, this belief has been shown to be incorrect by a branch of calculation that shows how to perform matrix division (“Vortrix Algebra”, yet to leak into the mainstream). Many of the preceding examples show that hard proof works to rapidly change beliefs only some of the time. I don’t know why, but the answer (if any) might be interesting. Maybe we also need to study https://en.wikipedia.org/wiki/Edward_Bernays.

Ok, I see where you've been going ♥️