Backstage note: I'm writing the RSS by hand, as a learning experience. If you encounter any weirdness with this show in your podcast player, I'd love it if you DM'd me and let me know so I can try to fix it.

Beyond that, I welcome any and all feedback about the show, Hest itself, or anything else.

Searching for Ivan Reese works best on Overcast.

There will not be show notes, transcripts (sorry, Jack),😹

I enjoyed the first episode 🙂

Very inspiring and good ideas! I love the length of the episodes!

Just finished the backlog. Really enjoying this and definitely hope you continue to make episodes. I have gotten rather tired of interview style podcasts and happy to see someone doing something else in the programming space.

I particularly liked your contrast with visual and graphical programming and your explicit notions of time. I find the latter really interesting. This is something that things like factorio have that I do think helps people understand their systems much better.

I also, like this idea of participatory execution, it puts words to an idea I have had for a while as well. The opaque nature of execution is most often at the root of why I have a hard time debugging software. Being able to participate, definitely seems to go a long way towards tackling that.

Finally, my suggestion might be a little bit weird, but I couldn’t help but think about the philosophical concept of hyper-time as you were talking about the various levels of time you have to think about. The typical metaphor for hyper time is actually, exactly the laying down rail road tracks you gave in the podcast. Imagine you are laying down tracks, tracks already laid are the past, the leading edge of the track is the present. The future doesn’t exist yet but are the tracks you will lay down.

Now imagine that all the tracks are red except there is 1 blue track 3 segments away from the leading edge. Suppose we decide to remove the latest three tracks (rewinding time) and replace the blue track with a red track. Now how do we talk about that? If we say something like “in the past there was a blue track”, we are speaking incorrectly in some sense. The “past” refers to the tracks that are not at the leading edge, and if we search those tracks there are no blue ones.

Instead we can say that in the “hyper-past” there was a blue track. Basically, hyper time is the time of the observer of the system.

Not sure if that was a good explanation, but I’ve found this paper on time travel without paradox to be a pretty good explanation. https://andrewmbailey.com/pvi/Changing_the_Past.

Not suggesting you need to use this in your programming system directly. But figured it might be a fun new angle to consider things from :)

Anyways, keep up the great work, really excited about this podcast.

That's wonderful! Thank you for the feedback. This idea of hyper-time sounds super relevant. As you noted, I do like having names for concepts, and I don't have good naming or framing for the different levels of time. That link seems to be a 404 — is this the link you meant? https://andrewmbailey.com/pvi/Changing_the_Past.pdf

Yep, don’t know why that .pdf got cut off.

This breckyunits scroll is only one column 😡

Does it look terrible? Improving the typography and layout is a priority. Screenshots welcome.

I think true logeracy is even rarer than you think. I worked at Google, where all engineers were tested for their ability to analyze the computational complexity of algorithms in order to get the job, but if you asked them to use the same math in any other context, they couldn't. In particular, any attempts to get anyone to apply the same mathematics to software architecture, was futile: https://www.youtube.com/watch?v=3Ea3pkTCYx4 (I demonstrate my own illogeracy in the video when I say something is exponentially better because it has a better exponent, but the correct term would have been asymptotically better.)

Hash and mean spirited? Not sure this is the right forum, but we can try. 😉

Does numeracy compete with logeracy? Becoming numerate, are we inclined to treat all numbers as similar because operations work uniformly for all numbers independent of their interpretation? (Thinking of Richard Feynman on the textbook selection process.)

Also this line is too good: "Someone literate can read and write an address on the front of the envelope. Someone logerate can use the back of the envelope."

Meh, it's hard to know if something is linear, quadratic, exponential, or even a derivable PDE without a lot of data points.

Also, you talked about Taleb so you know that exponentials have a way of leveling out.

So thinking in logs can be a useful skill, but IMO usually up/down is good enough.

Alternatively, if I was going to add one thing to math curriculum, I believe getting comfortable with probability / Bayesian would be gold.

@Don Abrams There is hope for getting people comfortable with Bayesian update. The trick is to use odds (3:2) rather than probabilities (60%). Then Bayes Rule becomes a lot easier to apply. See Grant Sanderson's recent video on this, which has quick become one of my favorites https://www.youtube.com/watch?v=lG4VkPoG3ko.

Agreed, though Grant makes everything look easy :)

Sanderson does have a singular skill, so compare his previous Bayes theorem video https://www.youtube.com/watch?v=HZGCoVF3YvM. Also ties back to the original topic, "to be clear, anyone who is asked this question is not expected to have perfect information about the actual statistics of farmers and librarians, ... but the question is whether people even think to consider that ratio enough to at least make a rough estimate. Rationality is not about knowing facts, it's about recognizing which facts are relevant."