Show more

Interesting tidbit in Mozilla's speech-to-text: it is now starting to emulate breathing.

I want to use/create the term "cryptofacilitation" to refer to facilitated discussions that use vagueness and incomplete information as a technique to advance the conversation, but I know everyone will expect it's just about cryptocurrency.

I guess to work you need a second chapter. Do you embrace your evil deeds? Lash back against whoever sent you on this journey? Create an army of minions that you throw against their attackers, only speeding their demise?

Video game concept: a simple shooting grinder where you collect treasure and experience points, and as you get more experience points you develop skills, until you eventually learn to understand your enemy's voice, hearing them say "no! You've killed my entire family, how could you?" or "please, just leave us alone." Or "spare me, please"...

Do any programming languages define their own semantically aware patching tools? As I understand it version control tools don't actually encode diff, merge, or conflict resolution tools, but the defaults are the line-based tools we all know. But you could do something else...? Even just for JSON (this could avoid many unimportant package.json conflicts). Or for ipynb.

Wisdom, the 68-year-old Albatross, has hatched another chick. Presumably a widow (probably many times over), she met her current mate (Akeakamai) at the age of 56. She's known to have at least 31 children, each of which takes over a year to raise (including 7 months of incubation).

staradvertiser.com/2019/02/11/

I'm taking a class where we have several multi-hour chunks of time set aside for us to meet in small (5-person) cohorts and do our work. It's been nice, we get to discuss lots of details, talk through formats and approaches.

It struck me: I never do this. I spend eight hours at work every day, working on hard problems co-owned with other people, and we never spend hours working on solutions together. Minutes sometimes? Lots of coordination time. But working... no. And this seems normal. Why is it normal?

As I become familiar with an issue where there is a diversity of subjective feelings people experience, I can learn to frame my imaginative empathy more correctly. But there's always new frontiers, there's more for me to learn and more for others to learn, and we're all just somewhere on our journey... so I think I should think more about this empathy trap when interpreting other people's interpretations.

Obviously how I first frame the question, and the fixed-vs-variable sense of my own identity is going to radically change the outcome of my imagination. But this all happens really quickly, it's like the very first moment of empathy, I'm not thinking rationally or even consciously yet. Next up I just have to decide if I even want to invest any energy in considering my inference.

An aide in a Vi Hart video I was watching: if I see someone doing X, and I want to figure out why, my first attempt will be to imagine I am doing X and why I might do that. Her example then: if I see someone talking a lot about their gender identity, then why might I imagine myself doing that? To... get attention? Because I'm relating this to my personal lived life I'm not going to come up with much. (Con't)

Separately I was thinking about libertarians vs. anarchists – they have many purportedly shared values, yet are far apart from each other. This definite vs. indefinite optimism (or maybe definite vs. indefinite good) could be a way of distinguishing them.

Libertarians are very indefinite, they studiously avoid making any assertion about what a good life is, what anyone might want to do with their freedom. Anarchists... much more definite, with opinions about what freedom actually means.

I find Peter Thiel to be a disturbing figure, and yet with some interesting perspectives.

In this review – slatestarcodex.com/2019/01/31/ (section IV) he proposes a dichotomy:

"Definite optimism" is the belief things get better because of doing good things. Specific good things, like you figure out something that makes things better and you do it.

"Indefinite optimism" is structural: what is good? What is better? How do we make things better? Eh. We can only structure the world to hopefully ratchet up.

I was listening to something about LaTeX equations embedded in a notebook, and I thought ugh, why can't you just write your equations in Python and have those get nicely displayed. And of course that exists!

github.com/erwanp/pytexit

In fact aspects of this argument are quite important at Mozilla, but from a different perspective. Many at Mozilla are very aware that the impact of Firefox itself is a relatively small percentage of users. But Firefox is also what makes Mozilla relevant, and Mozilla can have a greater effect on Chromium acting as a peer rather than as a contributor. And if Gecko is not as widespread as Blink, we use our mission to assert ourselves as peers anyway.

twitter.com/auchenberg/status/ : "Thought: It's time for @mozilla to get down from their philosophical ivory tower. The web is dominated by Chromium, if they really *cared* about the web they would be contributing instead of building a parallel universe that's used by less than 5%?"

The actual argument behind this is an interesting one. I think he's pushing an essentially utilitarian approach: this is where you can do the most good.

print(): a temporal ad hoc dynamically scoped return value.

"Dynamically scoped return value" isn't actually a thing, but I don't think there's any other name.

The pattern is common: setup some data structure, call a function, anything in the system can modify the data structure, when the function returns check the data structure. Now the function can "launch" activity without having to understand the entirety of the activity. stdout is just one such data structure, but the pattern is bigger than print.

Per a comment on HN, the terminal Pike used/created was called Blit: en.wikipedia.org/wiki/Blit_(co – it was basically a smart dumb terminal, which had extended escape sequences that allowed graphics and mouse interaction. But those were extensions of the terminal and login, so it's like opening a terminal window and then programs you launch run inside that window.

It does feel a bit closer to a browser than modern windowing.

"[...] This is hard to explain (it is, as I said, philosophical), but a trite explanation might be that you look outwards through our windows but inwards through theirs."

I'm actually not sure what he's comparing to Smalltalk here, i.e., what is Bell Labs interface he's used to? Terminal windows?

This does make me think of browser tabs, which are content-neutral, at times for better and at times for worse.

Show more
Mastodon for Tech Folks

This Mastodon instance is for people interested in technology. Discussions aren't limited to technology, because tech folks shouldn't be limited to technology either! We adhere to an adapted version of the TootCat Code of Conduct and follow the Toot Café list of blocked instances. Ash is the admin and is supported by Fuzzface, Brian!, and Daniel Glus as moderators. Hosting costs are largely covered by our generous supporters on Patreon – thanks for all the help!