## New city, new framing

* TOC
{: toc }

### Move to London

My girlfriend and I moved to London this past Monday. I haven't done any proper work since then, I've been so busy getting our home set up. The trickiest bit was the internet, both wifi and mobile, but I solved it for now with a combination of a mobile wifi hotspot and Google Project Fi with a Google Pixel 2 XL. the jetlag is also annoying. I am hopeful that I'll get back to a real work cadance the week after next, with some work done next week.

### New framing customization conversation

I’ve been (very much in the back of my mind) developing a new framing for my work that I have (somewhat successfully for the first time last night) explained at parties to lay people. (Halfway down this partly-true account, I delve into more technical details than I would not do with a lay person.)

Me: I am working towards a world where people can modify the apps they use while they are using them. The dream is to abolish the “settings menu” of each app, because you can literally modify anything about it. For example, what’s an app you use every day - email? Could you imagine any ways you’d want to change it to suit your workflow better?

Party Person: I can’t think of anything... I’ve never thought of it before.

Me: Exactly! When you can’t change anything, it doesn’t occur to you how it could be better changed. Do you ever question changing the speed of gravity? That’s the first step: enfranchisement, empowerment. Once you realize that you can change things, you won’t be able to stop yourself for coming up with ideas for improvements. For example, I would like to be able to have my email remind me in 3 days when someone I emailed doesn’t email me back, but only when my phone's location is at the office because I don’t want to be bothered at home. That’s custom logic that’s impossible today for no good reason.

PP: That's neat... but many people can barely even use technology as complicated as it already is. Who would really want to change things? 

Me: That’s a great point. I don’t think my grandparents would ever customize anything. Neither my parents probably. However, if customizing software became as easy as I hope it can be, I can imagine a world where my father, a businessman, would hire a tech firm to customize much of his firm’s software to their needs, like they now hire a consulting firm to customize their Salesforce for them. But that’s all down the line...

Realistically my main users will be other programmers like myself. You'd think that programmers today could customize the apps they use, but not at all. Today if Gmail open sourced itself, it would be as impossible for me to change any part of it. The codebase is simply to large and the coding style too unwieldy for a single person to comprehend and change it. However, with the new programming language I envision, it should take me a reasonable amount of time (a couple hours, depdening on the task) to customize the software I use in the course of using it -- similar to using a settings menu, but on a bigger scale. 

This new language would unleash the creativity of millions of programmers to improve the apps they use all the time! In the past, open-source software only worked for developer-facing projects like operating systems and programming languages, but with this new language, I think we could build open-source versions of BETTER quality than company-created apps. We have more people and time at our disposal than any company! If this sounds crazy, people thought that a regular-person-created encyclepedia was crazy, yet all the world needed to make an encyclepedia better than any private company's was the platform of wiki software. I believe a similarly democratizing platform could exist for software itself. Do I sound crazy?

PP: Um, a bit < ... nervous laughter ... > But don’t all those apps have proprietary licenses and content deals in place?

Me: Yeah, I don't know how we'd get around liscening deals, such as the ones Netflix have. But as far as other proprietary stuff goes, we could rebuild it in this new language, such as rebuilding social networks into open, federated protocols like Mastadon. (I do worry that Facebook is so big and high-quality already that it'll be difficult to compete with it, but I hold out hope!) It was the same when open-source started. At first, all the offerings were made by large companies and that was assumed that was how it had to be. Then Linux and git, distributed version control changed everything. I think making software customization 100x easier would cause an even more disruptive shift in the way software is consumed and produced. If software is eating the world, I want individuals, not massive organizations to be doing the eating.

PP: Wow. I guess that makes sense... But it’s also nice to have things uniform, such as a back button in the same place in all my apps, and all my connected Google services. 

Me: That’s a good point. There’s a lot to be had from centralization as well (until Google shuts down or ruins your favorite service). There are always trade-offs in these things. It’s hard to imagine the solutions to those problems now -- like it would’ve been impossible to imagine Two-Factor Authentication before the web was created -- but I believe we can make a decentralized world as convenient, or even more so, than a centralized one because everyone will be empowered to improve things that they think need improving. Maybe paid services would pop up that would help transport your data between apps.

PP: When you say “decentralized”, do you mean “on the blockchain”?

Me: <... laughs and shakes my head ...> Honestly I have no idea. It's a complex question: who has control of what. Another essential problem: how to manage millions of slightly different versions of the same piece of highly customized software. That alone is an incredibly difficult research question! I’m currently just working on a piece of a piece of a piece of this puzzle.

PP: What piece are you working on now?

Me: I’m working on the visual side of things, building an interface that’s similar to a WYSIWYG like Squarespace or Weebly or Microsoft World but can create any arbitrarily complex interface that code could create.

PP: Oh, that would be useful.

Me: Yep. It’s the holy grail. People have been working on this for decades. My old boss Lloyd Tabb worked on it at Netscape in the 90s. Then he was optimistic as myself, but after beating his head against this problem for years, he believes it’s impossible. I’m still optimistic. There has been a ton of amazing research in the past decades since then, particularly in the field of functional and functional reactive programming.

PP: Do you really think someone who doesn’t know how to code would ever be able to make any significant change to the apps they use?

Me: I really do. For three years I taught coding to people of all ages and skill levels, but mostly beginner children aged 8-13 years old. I learned that the right programming environment makes all the difference. We used MIT Scratch which allowed many students who had never coded before to finish a full whack-a-mole style game in a single class without much of our help! I believe that with the right programming environment and proper motivation, anyone that can read can learn to code. I am hoping to solve both problems with a single solution: if we build a programming environment that allows people to customize the apps they use, we will have also provided them with the motivation to learn how to do it.

PP: So is this even possible?

Me: I intend to spend my career finding out.

PP: Well, good luck with all that.... I see my friend. Great meeting you.