jcurbo

Hi, I'm the poster. Sorry for the unexpected attention! I saw this on Bluesky and thought it was interesting enough to share here. If anything the style of your writing makes it stand out in a good way, we shouldn't always have polished/corporate-speak posts here.

I've had similar thoughts lately while using agents for coding. It seems like if I give the agents external tools and sources of truth they perform really well. In my case it was a lot of standards, extensive use of Rust language features (e.g. traits), formal methods tools, large-scale architecture and specification documents (accessed via a special tool I also wrote), and using the beads_rust tool for tracking session state and giving the agent a list of issues to work. I hadn't thought of this as 'backpressure' but I really like that framing. It gives the agents guardrails and context and external validity that it can't make up.

This is pretty much my approach. I started with some spec files for a project I'm working on right now, based on some academic papers I've written. I ended up going back and forth with Claude, building plans, pushing info back into the specs, expanding that out and I ended up with multiple spec/architecture/module documents. I got to the point where I ended up building my own system (using claude) to capture and generate artifacts, in more of a systems engineering style (e.g. following IEEE standards for conops, requirement documents, software definitions, test plans...). I don't use that for session-level planning; Claude's tools work fine for that. (I like superpowers, so far. It hasn't seemed too much)

I have found it to work very well with Claude by giving it context and guardrails. Basically I just tell it "follow the guidance docs" and it does. Couple that with intense testing and self-feedback mechanisms and you can easily keep Claude on track.

I have had the same experience with Codex and Claude as you in terms of token usage. But I haven't been happy with my Codex usage; Claude just feels like it's doing more of what I want in the way I want.

This is great stuff! Civ3 is still by far my favorite Civ. And a nice use of Godot.

People interested in similar experiences should check out this podcast/series of videos where two Windows/Mac users try desktop Linux and report out on their experiences.

https://www.pcworld.com/article/2888221/check-out-pcworlds-n...

Docker on Linux. Gonic, navidrome and airsonic (and airsonic-advanced, the fork I used) all have docker images available.

For me folder view is partially just what I’m used to, but even with good tags some kinds of music is easier to look up by folder (clsssical).

I also use gonic over navidrome (and formerly airsonic) because Navidrome doesn't support folder view (and apparently never will). As nice as Navidrome is, that's a dealbreaker for me. Gonic works great though.

That's wild, I was in LA recently for work and drove by that area and wondered what was up with the street grid. I figured it must be something like this given the airport.

Battlefield's anti-cheat isn't supported in Proton/Wine, so you're stuck with Windows there.

This is a talk from DistrictCon. There is video available here: https://www.youtube.com/watch?v=wdfe2BnXQjU

Likewise, I used it for a while as well. I also used a console client named micq, good times.

https://github.com/icloud-photos-downloader/icloud_photos_do... works great for me to do this.

I just want to pour one out for CalSky, that was a great page for setting up custom trackers for lots of astronomical things. It was very old school web too (for better or worse).

High five. 6 yrs in the USAF here.

Yes; APL is a University Affiliated Research Center (UARC) and while we sometimes do work with industry, our whole purpose is to support the US Government. That doesn't necessarily mean you need a clearance, but much of our work is national security related.

https://en.wikipedia.org/wiki/University_Affiliated_Research...

Nice to see fellow APLers here! I have been at APL 5 years and like it a lot.

Echoing others, thank you for writing this (as someone doing an applied math masters and digging into ML - I have used ESL for a class but not the others you mention)

"But can it run Linux?"

Of course it can, 486s were common in Linux's early heyday. During my senior year of high school and freshman year of college ('97-'99) I ran Debian Slink on an IBM 486SX. X ran a little slow on it so I used it in console only mode. I mostly used it to do compsci homework. Before I settled on Debian full-time (still using it to this day) I used Red Hat, I think v5 (the original numbering, before RHEL). And before that probably Slackware on floppies. I eventually got an AMD K6-2 which ran a lot faster...

Of course, this article is about running modern Linux, but Debian Slink is still there to download and install and I'm sure it works just peachy.

The section on configuring the kernel really gives me nostalgia as I used to build my own kernels back then, something I haven't done in years.

Related: https://hub.jhu.edu/magazine/2021/spring/apl-interstellar-pr...

(Disclaimer, I work at APL, but not on this)

Are there really PDP-11s out there still doing stuff? That is wild

Hit "Next" at the top and it will advance a day.

https://www.gi.alaska.edu/modules/custom/aurora-forecast/ima...

That's a direct link to the map as of this post. Unfortunately we have rain in the forecast here (I also live in MD).

Not too far off from circuit design languages (VHDL, Verilog) to me.

There's also this book by Emily Lakdawalla, "The Design and Engineering of Curiosity"

https://www.amazon.com/Design-Engineering-Curiosity-Performs...

This is a solid read, with good book recommendations. After several years of tinkering with self-learning I bit the bullet and applied to a MSc in Applied Math program. (via ep.jhu.edu) I've had to take some pre-reqs to get started since it's been almost 20 years since I have been in a college math class, but it's been an enlightening journey re-learning calculus and now dipping my toes into differential equations. I don't think I could have gotten this far with self-learning, but I realize YMMV.

I will say I don't feel like single-variable real number calculus tells the whole story. I had taken that and linear algebra in undergrad but never any further, and now that I've taken single and multiple variable calculus, with real and complex numbers, plus integration of linear algebra ideas, the mathematical model feels a lot more like a cohesive whole to me, highlighting fundamental ideas that only barely peek through in a typical Calculus I class. I would encourage anyone talking to calculus to at least do the typical Calc II class, if not Calc III/multivariate. There is a beauty and structure to building up from calc I through III that I was missing before.

Your site is exceptional, thanks for creating it. I used it a few weeks ago to see a Starlink pass over my area. I like how you added the simulated view in Streetview, I think that will help a lot of people understand what to look for that aren't necessarily astronomy types.

byuu, thanks for doing more writing. I have enjoyed following you on Twitter and look forward to reading more in depth work from you. Will definitely sign up for the RSS feed.

The Air Force has a similar guide that has been around for decades, the "Tongue and Quill." We were introduced to it starting in officer training.

https://static.e-publishing.af.mil/production/1/saf_cio_a6/p...

I can't speak for Android but on iOS there are several apps that can do this. Pocket Universe and SkySafari are two I use frequently.

Is it this one? https://github.com/b3ll/Ignition

Same here, except with astronomy/astrophotography and outdoors stuff. I like Instagram in that way and don't follow any big celebrities or big influencers, so never see any #ad sort of stuff.