Personally I'm learning Elixir, and it's such a pleasant language. It feels great to write, and the packaging/build tools feel refreshing compared to the mess of Python.
Now, handing you the mic. Is there a new stack or language on your mind?
We ended up in cloud, I handed over control to my instructor and spent the next 10 minutes wondering what I would have done if he wasn't with me (hint: not good).
So I slept on it, text my instructor a couple of days later and told him I want to be able to do what he did. He told me I need an Instrument Rating (Restricted) rating, which allows me to fly in and above cloud.
So that's my learning for the next 10 weeks or so. I've got to take one written exam, one flying exam, and then I'm cleared for flying cloudy days.
Low visibility flight is frequently dangerous even when you have proper training. A lot of people that fly are surprised how easy it is to get disoriented in low visibility conditions.
Learning to use only instrumentation is a great and valuable skill, but don't forget that even if you know what you're doing the best option is to avoid bad weather if you can help it.
Thanks for your input, I'll bet your father is a cool dude!
It seems like an IFR cert would be really good to have if you were up longer than intended and had to land at night.
Edit: More info about night rating: https://www.caa.co.uk/General-aviation/Pilot-licences/Applic...
The chain of events were we were circa 2000ft VFR. Instructor took control to climb in to cloud to get on top of it. Got in, instructor realised we were not going to be able to get on top of it and so IFR'd us back to the aerodrome with an ILS landing just for show.
Then I tinkered with some basic functions: how to strip a list of URLs for a specific Product ASIN, automating some Photoshop image creation, and searching a 10k row Excel doc for specific phrases. All were bad at first, but they worked eventually.
I then did the Automate the Boring Stuff course on Udemy, super fun and deepened my knowledge. Helped me improve some of the programs and start working on others. I started to work with the Facebook Ads API and Pandas to automate reporting. So fun. That program is just getting to the Excel/Sheets automation part which will save me a ton of time every week. I spend a solid amount of time manually analyzing data each week. If I can cut that down, it'll get me quicker insights so better for my clients.
Again, nothing works incredibly well but it all works. And all of them save me time going forward. Automating image files will save a team member 3-6 hours/month and reduced errors by probably 90% (And their stress. We've already used it for 2 months so that's reduced their stress level from the errors they made manually doing it).
I'm basically just carving out an hour a week at this point after shooting for 5+/week for the first 6 months. I might increase it if I slow down on client work or hit a blocker that needs more time.
I subscribe to Always Be Learning, since that is a cornerstone of my own well being, so there is no real goal. I figure if I have a system for it then I'll make progress. And everything I learn is really a bonus for myself, my clients, or any developers I work with.
There's no shortage of interesting ideas to pursue with it so that won't be a problem anytime soon.
I know a little bit about databases so that gave me a good foundation to work with. If I was 18 and starting from scratch, some of those concepts may have taken longer to learn.
Other than that, pretty awesome. I went in thinking that I would need a few hundred hours of work before I would feel comfortable and I think that was about right. So I just enjoyed it along the way!
I've been binging books on the topic as well as exchanging dialogs with a colleague over email. Books list:
_Completed_
The Origin of Wealth: The Radical Remaking of Economics and What it Means for Business and Society - https://www.amazon.com/gp/product/1422121038/ref=ppx_yo_dt_b...
Increasing Returns and Path Dependence in the Economy (Economics, Cognition, And Society) - https://www.amazon.com/gp/product/0472064967/ref=ppx_yo_dt_b...
COMPLEXITY: THE EMERGING SCIENCE AT THE EDGE OF ORDER AND CHAOS - https://www.amazon.com/gp/product/0671872346/ref=ppx_yo_dt_b...
Complexity Economics: Proceedings of the Santa Fe Institute's 2019 Fall Symposium - https://www.amazon.com/dp/1947864351?psc=1&ref=ppx_yo2_dt_b_...
_To Be Completed_
Information Rules: A Strategic Guide to the Network Economy - https://www.amazon.com/dp/087584863X?psc=1&ref=ppx_yo2_dt_b_...
https://www.amazon.com/Misbehavior-Markets-Fractal-Financial...
I began reading a book on analytic number theory on my own sometime last year. While re-reading some chapters of the book again, I decided I might as well do the re-reading with a group of other folks who are interested in this subject. So I began hosting book club meetings for analytic number theory since March this year. I had made a Tell HN post about it back then.
Those meetings are still going on consistently. We have a tiny but regular group of participants who meet daily for 40 minutes to read the book together. In fact, we now have a small community around it. We call it the Offbeat Computation Club. See https://offbeat.cc/ for more details about it. We plan to add more topics of discussion, such as Common Lisp, SICP, etc., soon.
We will begin reading a new chapter (Quadratic Residues and the Quadratic Reciprocity Law) on Monday. This is a pretty self-contained chapter and quite accessible to someone who has not read the previous chapters but has some basic knowledge of modular arithmetic. If this sounds like fun, you are very welcome to join us. See the link in the previous paragraph to find our IRC channel location, meeting link, and other details.
There's a great youtube channel with hundreds of videos of guitarist Nelson Faria chatting and playing music at home with a different Brazilian musician each time, a lot with English captions. https://www.youtube.com/c/umcafelaemcasa/videos
And if you want to virtually travel to Portugal, check out Antonio Zambujo. Here he is singing a song [3] written by both Márcio and Pierre, as luck would have it.
https://www.youtube.com/watch?v=8QRdaL3GEiQ&list=OLAK5uy_lCU...
https://www.youtube.com/watch?v=Vr2qTlFkytk&list=OLAK5uy_n3k...
"E boa sorte nessa aventura, meu amigo!"
So I'd suggest you Egberto Gismonti, a Brazilian musician that released some very good work on the excellent ECM label (same as Jack De Johnette and Keith Jarret). My favorite of his albums are "Sol do Meio Dia" and "Carmo".
I'd also recommend the Argentinian Astor Piazzolla, I love the album "Cumbre-Reunion" that he made with Gerry Mulligan.
https://tedgioia.substack.com/p/the-most-musical-man-in-the-...
He's pretty incredible!
My repository for this (code finished, just need to add photos and add a write up about how to use):
https://github.com/whyboris/Digital-Piano-LED
I added a feature that the left and middle pedal buttons navigate through sheet music (PDF left/right button). And I'm also running Pianoteq which makes any (even dinky) digital piano sound like a $100k grand piano (or any piano you pick for that matter) https://www.modartt.com/pianoteq
Still WIP, but easy to edit: https://github.com/whyboris/Digital-Piano-Visualization
I used Processing because of how easy it is to use and how performant it is for smooth animations.
Even so, you are right, it is definitely not a growth market by any stretch of the imagination, it takes a ton of effort to learn how to play and in a world full of distractions and ready made music more than you could ever listen to in a lifetime the enthusiasm for doing this is waning. There are exceptions to this though, for instance, in Asia where playing the piano is a status thing and still in its ascendancy phase.
I have this pipeline where I take a youtube video, download it as an MP3, convert that to MIDI and then edit that midi to split left/right and to ensure it is all set to time and of the proper length so I can show it as sheet music.
I knew next to nothing about music theory when I started this beyond the very first basic stuff, but after watching many youtube videos on the subject and reading a lot of stuff it is starting to make sense. This is a subject that is overcomplicated to the point that it seems much harder than it really is, there are only very few good teachers out there to make that which looks complicated but it ultimately relatively simple simple again.
If you want I can do a write-up of all the resources that I've used over the last year and a half.
Currently practicing Ennio Morricone's "Chi Mai" arranged for piano.
Progress is still slow but if I compare my sightreading spead, accuracy and general quality of play with a year ago the difference is huge even if you can't see it day-by-day. Each new piece brings new challenges, and teaches me something that re-inforces the pieces that I already know how to play which then all get a little bit better.
Overall I'm having a ton of fun with this and the joy I get from playing a piece end-to-end without mistakes, at speed and in a way that is nice to listen is hard to describe.
It's obvious that an experienced pianist would probably laugh at the level of my accomplishments but that's fine with me, I'm enjoying this and that's what counts. It's also one of very few things in my life that I've done without any commercial goal, and which is just for myself. Overall if I could do this life over I'd tone down the business career in favor of making more music. And the programming skills came in handy while making pianojacq.com, so in a way this allowed me to combine two things I love.
1 month ago I was responsible for the entire IT function of an SMB. I was the guy who knew everything required to get stuff done, and I reported directly to the CEO. A lot of processes were not formalised because the overheads of doing so weren't worth it.
New job, now working for the government in a skyscraper full of colleagues. I know virtually nothing, there is no onboarding to speak of, and I am second from the bottom in a huge hierarchy strangled by red tape. Every day I find out that there is another whole team dedicated to doing something I thought I was going to have to work out myself - which has pros and cons. I haven't even seen an org chart yet, but I've just been told "those teams over there are getting outsourced next month". So I think I know why there is no up-to-date org chart...
It's different. But it's a pay rise, and I'm not bored anymore. The level of coordination required to get anything done here is a learning exercise in itself. After my first week I was worried I wasn't cut out for this sort of thing, but after a couple more weeks I'm finding it... fun? Still weird though.
- Soft skills matter, but performance (against your metrics) is still what matters in the end. Group dynamics only comes into it when you have a bunch of people at the same level of performance.
- You'll be ahead of a lot of your peers if you stay positive and don't gossip and don't do office "politicking" (in a bad sense).
- Plan your career around three year stints. At the end of each stint, evaluate your career progress against your long term goals. Don't be afraid to transfer jobs or take time off for learning (e.g., by getting a graduate degree).
This is important. I've worked with people who are constantly criticizing the company/other people, and while they are usually right, it undermines the morale of everyone else.
At the very least, reading this book is a way to pick the brain of one of the most brilliant managerial minds of our time (he passed away a few years ago). It’s usually hard to find good mentors in any company, so reading books is one way to get mentored by someone outside the company.
Also re-learning modern JavaScript, and TypeScript too - I've been using JavaScript professionally since 2003, but always begrudgingly. And spending some time picking up Flutter (and Dart). We're doing more and more work on the front-end and on our mobile app, and as a technical founder it's good to have some idea of what's going on even if you're not writing the code yourself.
Learning to be a better manager too, one mistake at a time, as I've been doing for the past few years.
I didn't know about Sadoway, the EdX course has a different instructor. But I checked some of his videos on OCW now, and I really like his enthusiasm!
Topological Manifolds (Lee: Topological Manifolds)
Abstract Algebra (Dummit and Foote)
Geometric Algebra (Doran: Geometric Algebra for Physicists)
Differential Geometry (Fecko: Differential Geometry and Lie Groups for Physicists)
Though I'm less invested in these last two currently, as the diff geo's a prep for the masters I'll be doing and the other is something I'd really like to pursue but can't motivate myself to start working the problems out on.
I play around with some tech stuff, but nothing major -- just doing FreeCodeCamp and trying to keep my Python skills from going too rusty.
I'm fixing to start an Applied Mathematics and Theoretical Physics masters, so might review QM and CM as well as it's been a while since I've done them in school and I need a refresher, even if I hope to lean towards the 'Applied Math' (which is mostly numerical algorithms related, etc.) side of things in the masters.
Otherwise, I've been continuing working with Irish (Gaelic), and keep dabbling in some other languages, ranging from Spanish to Latin to Japanese. I also try to read widely in various topics from religious studies, philosophy to linguistics, psychology (I know...), sociology, pop science, etc. I've also bought an ocarina I need to try to practice since I'm off for the summer and don't have to worry about disturbing my roommate. But this is all mostly dabbling at this point, as I'm fixing to move countries for (at least) a year, so...
Our hypothesis is that there must be enzymes that works similarly with plant proteins as chymosin (part of rennet used in cheese making) works with casein.
So, I need to catch up on my bio basics to be able to better understand what's going on :-)
did you find that to be a hindrance for yourself?
I wouldn't say hindrance, but it would probably be a bit easier if I had a biochemistry or food engineering/science background.
My partner has an environmental science background, so that helps a bit and she is the one doing most of the experimenting.
But in the end, it's about trial and error. That's also what an expert would need to do. I think if you're interested enough in the topic it's possible to acquire the necessary knowledge.
We plan to hire experts later on once we have the cash to do so (we're not raising money, all bootstrapped).
- French which has to be done but is quite hard.
- Some math because the more I work with computers there more I understand how it would be very useful.
- Combat sports because it keeps me fit and its good for your posture if you know how to throw a punch.
- Traveling. Never learned it, work too much and would like to see the world before being an old fuck.
- Basic EE digital circuits and microcontrollers. One should really have a grasp how computers work on a fundamental level.
- Cooking. Because it is social, full of culture and makes you independent.
I built a small site for myself to help in learning the 5000 most frequently used French words. It is, and always will be, free and with no account sign up needed.
Check it out, you might find it to be useful:
- Duolingo(don't do it for more than a few months)
For beginners
- https://www.languagetransfer.org/free-courses-1#french
For all others, some sort of immersion approach is the only way if you don't live in the country. Figure out the general approach below and adapt for your target language.
- AJATT http://www.alljapaneseallthetime.com/blog/all-japanese-all-t...
- TMW https://learnjapanese.moe/
- Refold https://refold.la/
"Learning OpenCV 4 Computer Vision with Python 3: Get to grips with tools, techniques, and algorithms for computer vision and machine learning"
by Joseph Howse, Joe Minichino - 2020, Packt Publishing
available here (library genesis): http://libgen.lc/item/index.php?md5=9208FA33E5C1F93918E128F8...
Thanks for recommending it! Just this week I started a sabbatical devoted to putting together a computer vision + ML project. So this will be useful. Adding it to my learning journal.
Shoulders are very flexible but also extremely weak and sensitive, they should rarely be used out of neutral position. I've taught martial arts for a long time and most issues I see with people's shoulders is due to (improper) overuse.
My point is that maybe the cause of the problem isn't in bed at all. My experience says that sleeping on the side makes everything easier, from breathing to offloading the spine.
Believe there's studies on amyloid plaque clearance differing b/n different positions. And I believe back sleeping has less of such than side.
Got a background in 3D and Programming, but I think the 3d industry has the same approach to problems as programming. Couple of frameworks and libraries and poof, you have an applications.
In 3d you grab a render engine, a light setup, some assets and poof, you have an image/game/animation. I miss the days that I 3d was exploring and experimentation. The alternative would be to dive into 1 subject (modeling, rendering, lighting, FX etc). But in my experience that just made me feel like a factory worker. Piece of concept art, here you go.
3D can really trap you into polishing a soulless turd, so I'm learning drawing. where you can't cut as many corners and enjoying the creative process again.
On the programming side I'm just enjoying work and learning on the job, also going back to the fundamentals like shell, sql and regex. It's amazing how much you can automate.
Especially resources for deliberate/edge practice.
* The other thing I've been investigating is meta-learning techniques for predictive modeling, like ensembling, stacking and boosting (the kinds of techniques used in ML competitions) -- basically ways to combine statistical models to extract more signal without overfitting. At some point individual statistical models are limited in their ability to extract signal from data (even though the signals are there) -- and neural networks are a step too far because of the data volume requirements -- so learning how to combine multiple models is the way to go. None of this stuff is new and I'm a few years late to the game but better late than never.
I routinely keep a web page updated with all my learning projects, both the ones I did and the ones I'd like to do [1]. I found that it keeps me accountable, plus it might be useful for some.
This thread is a gold mine of ideas to expand it!
https://pragprog.com/titles/smgaelixir/genetic-algorithms-in...
So 6 months ago, I quit my job, published a short book, then studied AI. It's funny how so many material makes the vast assumption that you already know so much. They gloss over activation functions, loss functions, vanishing gradients, or the most fundamental things like properly loading your data. No one even tells you how to load the data!
Anyway, I got my Deep Learning Specialization certificate, I watched all 3blue1brown videos, I built a smart assistant that runs on raspberry pi in my car. The goal is to have it Assist the driver, with blind spots and keeping the eyes on the road.
Can you please share how you started and the projects you worked on?
So I turned to coursera, took the deep learning specialization course by Andrew Ng. In a month, I had a working prototype. Sorry I haven't published it yet, but it will also be posted here. Hopefully it makes it to the front page.
What is this? Will it help you get a job?
Some specialization courses offer a certificate on Coursera. I don't know that it is recognized by institutions, but I believe completing an Andrew Ng class, with the addition of building a project with this new knowledge can be helpful.
By the way I'm in the process of interviewing with FAANG, so let's see how much it will matter :)
Think of Facebook. You create a post and see it on your wall right away. The rest of the world doesn't need to see it right away, but you should or else you think the post wasn't saved. Does that post get saved in some db/cache specific for you, and _then_ an event is emitted?
But yes, I'm learning things like that.
It depends on what you're trying to achieve - are you trying to increase reliability by going fully event driven? Are you trying to improve just one particular flow? Are you wanting to expose certain data for data science?
Unfortunately I don't think there is a good prescriptive answer to this - it all "depends" on your situation. If this is a publicly exposed service - you probably would want to commit/record that a user signed up or posted something and THEN emit an event which will start the chain reaction of other services reacting to the event.
If the user created a post - it probably has an ID and you would probably want to communicate that in your event - but if you did not commit anything to the DB, then what are you exactly communicating?
Your goal should be to try to be as async as possible - but in some cases, it just doesn't make sense. For example - receive an HTTP request to do something is a 100% synchronous operation. But _after_ you've handled that event - most, if not all things, can be async.
Contrived example:
1. User signed up -> UI hits public API which is backed by "Main API service" which commits to DB -> emit USER_SIGNUP event
2. Billing service consumes USER_SIGNUP event and creates a subscription in a 3rd party billing system (and maybe emits another event)
3. Metrics service consumes USER_SIGNUP event and starts recording metrics for this user
4. Audit service consumers USER_SIGNUP event and creates an audit log trace
5. --- Finally, the "Main API" service listens to all kinds of other events that causes it to update its "view" of the user.
Finally, some non-abstract advice - if you're building an event driven system from scratch, I would probably avoid doing CDC as the primary event emission mechanism and instead have your application emit events instead. You'll have a lot more granular control over what is in an event and not be tied directly to the DB schema. Your event schema will evolve and maybe it won't fit what's in the DB.
I would lean towards CDC if you are trying to retrofit an existing system.
Anyway, hopefully that's a bit helpful! I've been working on this sort of stuff for a while (and actually have a YC-backed startup that focuses on event driven: https://batch.sh).
I recently presented on event driven design at a few local meetups and made some slides, it might be useful: https://docs.google.com/presentation/d/1j6Cyid88Ca1shwEN6uyI...
Running - also relearning, which happened quickly this time. I'm already back to running 5ks after just a few weeks (though I also spent 6 months building up my cardio with a rowing machine).
How to be offline and focused. Computers (and smartphones) are fantastic tools of distraction. I'm trying to regain my focus overall. Reading physical texts is helping immensely with this. I've mostly reinstated my personal "no tech in the sitting room" rule, aside from my Kindle Fire which is sufficiently anemic to be useful as a reader and not as major distraction with other media.
I already took care of the weight I'd gained through improved diet and 6 months of rowing five times a week (got up to 30 minutes of moderate to high intensity). So my cardio condition is great, not where I want it, but great overall. It's really my running form I have to improve, and getting my feet and knees used to the impact again. I'm now back under 180 (versus 210 this time last year) which is a much healthier weight to be running at.
I'm being deliberate about adding distance slowly to minimize the risk of injury (from running). Recovering from an injury at 26 wasn't too bad, at 36 it took a lot longer, and now I'm about to hit 40 so I'm being more deliberate than I would've a decade ago. I'm doing 2x3km runs with a 5-8km run once a week. I plan to add 1-2 total extra kilometers a week and only if the runs have gone decently (don't need to be great, but need to be complete). I have no intention of pushing the short runs past 5km or the long run past 10km. Then I hope to find a soccer league and team I like to get my extra running in on weekends.
When I said I'm running 5ks, it's just the one a week. But I can actually finish it, which was a pleasant surprise (I thought it'd take longer to reach that point).
Full reference is: Astronomy Across Cultures: The History of Non-Western Astronomy ed. Helaine Selin and Sun Xiaochun (ISBN: 978-94-010-5820-9, DOI: 10.1007/978-94-011-4179-6)
• I’d quite like to learn how to do ‘low-level stuff’, particularly microcontrollers — I rediscovered an old Arduino which I’ve had some fun playing around with. I particularly want to get to the point where I can write low-level Forth code and flash it to the microcontroller. Also, I should learn assembly language at some point, but it’s proving difficult since I’m on Windows.
• Relatedly, I want to learn the basics of electrical engineering. My physics degree has given me a good overview of the underlying concepts, but I still can’t even design a basic circuit for my Arduino.
• I have very little knowledge of how the networking stack works, so I’m thinking of getting myself a Raspberry Pi and learning how to build and host my own website.
• Wrt linguistics (a favourite subject of mine), I’d like to learn more about morphosyntactic alignment, specifically split intransitivity and alignment in person marking. I have a whole stack of papers stored up which I’d like to read.
• Time management — I feel like I have so much time but am wasting most of it.
Luckily, university vacation started this week, so I may even get enough time to learn some of these!
There’s a great (very low-level introductory) IoT course from Curtin University on EdX[1] that will teach you the absolute basics of circuitry and get you started building things with arduino. MIT also has almost all of their courses available for free (search “MIT OCW” or “MIT open courseware”) if you’re looking for something more technical, given your physics background. I believe their introductory circuits course for EE/EECS students is 6.002.
The EdX course is cool though because it uses a site (I wish I could remember the name) that simulates circuits to let you build things with an arduino right in your browser.
[1]https://www.edx.org/course/introduction-to-the-internet-of-t...
(Further question: do you by any chance happen to know of a good textbook? I personally find that I learn better from textbooks than I do from online videos.)
Edit: In the past I've tried to do too many things at once and ended up making only limited progress on any of them as a result, so I'm trying to focus on just 1 or 2 things at a time now.
Clojure is simple.
Nearly everything is just (function arg1 arg2 ...), even at a fundamental language level. It's actually our background experience in complex languages with all kinds of wacky syntax that makes Clojure seem hard at first. But it's very simple. So if you can't remember the syntax or why something is the way it is, step back and you'll realize it's just (function arg1 arg2 ...).
For example:
(defn add-these [x y] (+ x y))
is still just
(function arg1 arg2 arg3)
Where:
function = defn
arg1 = add-these (a symbol)
arg2 = [x y] (a vector)
arg3 = (+ x y) (an expression, which itself is just (function arg1 arg2))
You may already know all that, but regardless I hope it's encouraging because the bottom line is Clojure is easer than you think and you can do it! :-)
Clojure is actually simpler (and eventually way, way easier) than most mainstream languages. It's just a bit different from what most people are used to and what is commonly taught as "programming".
1. There are several fantastic books on Clojure. The Joy of Clojure and Elements of Clojure are some of my favorites.
2. Watching a professional using an editor/IDE with a REPL attached to it. Look for online videos. This is the biggest reason to why people are „addicted“ to it. The language and it’s ecosystem are made in a way that enables this „living in the code“ experience.
I am interested in the actual Clojure bits. I mean, I know Lisp syntax and functional programming basics. But one also needs language-specific things: standard libraries, namespaces, etc. What is a nice resource for these?
I think planning is extremely neglected and undervalued for what it is. The field is a mess, but there's real opportunity now with all the progress in reinforcement learning.
Working in 3D has opened my eyes to new possibilities. I believe that having skills in 3D will become very important over the next five years with adoption of mixed reality hardware.
It's been fun seeing the world in a new way again. I like the nostalgic feeling of looking around and seeing the material nuance in everyday objects, how light behaves, and the uniqueness of everything around us.
1. 'Write a Interpreter in Go' by Thorsten Ball
2. 'Write a Compiler in Go' by Thorsten Ball
3. 'Crafting Interpreters' by Bob Nystrom
4. 'Ruby under a Microscope' by Pat Shaughnessy
Also, I'm trying to learn Racket in my spare time.
Kitesurf is really nice but you need to account for a few lessons before you can even stand on the board and prob a few more before you can rent on your own. In my case I did 6h base course, then another 10h with the teacher, typically 2h/day.
Windsurf you can start cruising around much faster, so I guess it's more rewarding immediately. This said many ppl told me that the progression is the opposite, meaning that with Kite you get better quickly, but with Wind it takes a lot of effort to grow from beginner.
On the language/stack side I'm deepening typescript and nestjs (nest with a S, not next) and will prob play more with rust soon.
Creative Focus: Since the beginning of the pandemic, I’ve been reducing the number of projects I take on to the ones that bring me the most joy. I manage an engineering org during the day and compose music by night, and have often felt I needed to —be- all aspects of each world. I felt I couldn’t lead my team without being fluent in every language, framework, ideology that they employ. In music, I felt like I needed to run multiple bands reflecting my range as a player, book those, publish a podcast on a regular cadence, make noise about what I’m doing every 2 weeks on the socials,etc. Now I practice, work on the projects I love, and invest time in my next learning area:
Teaching my team Conscious Self-leadership: I’m responsible for amplifying the superpowers of the humans on my team in an org that has a lot of 'old-world' patterns - expecting a parent to come in and solve your problems, expecting one person to issue an edict on technical direction without collaboration, expecting articulation of a goal to be enough, etc. I'm sure there are jargon words you can point me to for the patterns I'm seeing - I'm still on my first cup of coffee. In a remote-first world, people of all personality stripes need space to lead themselves where their most energized/joyful, and making that space can feel byzantine, but I don't think it has to. So: I'm sharpening patterns and techniques that I'm seeing work.
Villa-Lobos Etudes: I've been a guitarist for just about 40 years (yikes) and finally dug into these last year. I feel I'll be exploring them for the rest of my life, finding new things to work through every week.
Music Engraving: I'm publishing some pieces of mine (including https://www.youtube.com/watch?v=hmvI6H64SPI and https://www.youtube.com/watch?v=2M7vOIHOeeU) and am learning a ton about how to make a beautiful page of music for guitarists with my mentor John Stropes.
Ideally I'd like to press a button on the site and have it generate a manuscript - so far I've been looking at a combination of markdown, pandoc, and LaTeX for the formatted form. From what I understand that's frowned upon for Real Publishers, but works well for the DIY process (amazon, book baby, Ingram). I'm playing around with LaTeX templates now, and have a subsection of the book laid out manually now, using various guidelines I can find for font, font size, margins, etc.
I'm also writing a separate book more having to do with how people can respectfully share reasoned conclusions with each other - for that, I'm also using a LaTeX book template and am just writing it in LaTeX, since it's kind of fun to see an approximation of what the actual book would look like as I write it.
There's a long way to go on learning this though and I haven't found a lot of good resources, so I'm definitely open to suggestions for folks who have written good technical dynamic pipelines for book publishing - particularly anything that can be wired into a web backend that starts with markdown data.
I learned Japanese for several years and decided to learn something new, but I wanted the same experience I had for learning Japanese. There's just an insane dearth of GOOD learning tools for Japanese.
It turns out there's good stuff for Piano too, but none hit the right spot. So I made my own learning tool called Piano Gym.
Piano Gym is flash cards + midi piano + spaced repetition to teach you music theory and performance using actual sheet music and music theory tools.
It's SUPER cool, and I stream myself using it every day. I have a whole curriculum using a Piano method's book, so it's exciting to know that the content I'm going through is actually made by someone else.
I made a ShowHN post, but this site is weird, so if you're interested you can see that here.
https://news.ycombinator.com/item?id=27304875
Either way, it's pretty cool to learn using a tool I created. Excited to keep it going!
I'm surprised by those who commented that they're learning multiple unrelated subjects at once. Are you learning anything beyond a superficial level? Or is the goal to pique your curiosity on a new subject and then decide if you want to make it your core focus?
But, really, it's mostly because I have more time and few other hobbies currently (not starting anything new before I move countries in a few months).
I had spent a lot of time prepping before starting my training, so some stuff has been expected but other stuff has been a pleasant surprise.
One of the things that crosses over to engineering is the concept of flows. This idea that you’re constantly wanting to check on certain items throughout flight, and then at different phases of flight work through a flow.
You have checklists that you use and reference, but certain flows you commit to memory. This is very similar to incident response work as someone who has been in infrastructure for the last decade. How I triage and work through a problem has a very specific flow to it that helps me quickly sort out where the issue is (even if it doesn’t immediately tell me how to fix it). I always love seeing how two skills, like engineering and flying, can have correlated patterns.
For anyone with a love of aviation, go take a discovery flight. General aviation flying is a wonderful thing.
Then, 6 weeks ago someone stole my laptop from me while sitting in a cafe in San Francisco. Without thinking, I chased him down and was able to wrestle my laptop back from him. Without the BJJ, I probably would have just yelled at him. Anyway, I'm back at the BJJ studio!
I much prefer a C++ lite, with stl library, interfaces instead of classes, and c++ syntax sugar + procedural programming, please.
I am using C/C++ to program the library while learning essential concepts.
https://venam.nixers.net/blog/unix/2021/06/23/pipewire-under...
- PC Modding, Who would say that soldering and re arranging power supplies would be fun!
- Gardering. Gives me mental peace.
Need to build a stack consisting of Kubernetes (zero downtime, volumes), Docker for images, Django Rest Framework (backend), Nginx (reverse proxy for static files), Uwsgi (for Nginx-Django coordination), Next.js (front), Postgresql (db), Flux (for syncing with private docker registry images using maybe semver and for github repository polling), Ansible (for agentless initial configuration for test/qa/production servers), and Github Actions for CI (maybe later CD; servers allow only polling at the moment).
I hope that I can meet an artist some day and participate in a game jam.
- Databases : in the past I feel that I've learned too heavily on ORMs, so lately I've been trying to learn how databases actually work. Things like how to interpret execution plans, the data structures used for indexes, etc. Found some really cool tools like Postgrest & Postgraphile that are making me more interested in database-centric apps.
- Math : as a self taught dev that studied English in college, I was never exposed to much math. But I love thinking about the underlying patterns and structures of anything, and math seems to be an invaluable tool in doing this. Not sure how I'm going to self-teach math (or where to begin), but the little bit of discrete math I've learned has been super fascinating.
In case you haven't seen it already, http://8bitworkshop.com/ provides an online IDE.
It feels good to be a total beginner and then slowly learn different tricks/moves.
Sadly, it's gonna take me some more learning and experimenting to get there.
[1] - https://itp.uni-frankfurt.de/~gros/StudentProjects/Projects_...
[2] - All the online code examples I can find are using three.js or WebGL shaders and I have a stubborn belief that this is doable in vanilla JS using the 2D canvas API. Just because I've failed so far doesn't mean it's a Bad Idea!
For non coding related stuff: acoustic guitar and French.
(I know it is supposed to be pronounced 'geeks' but I can't help but want to say 'goo-eeks')
So far I've had more failed prints than successes over the last week or two, but I'm still excited to be doing it. I'm learning the ins and outs, which were different from the last printers I worked with. The wheels are churning as to what I can make, and I'm very excited to continue exploring and "resharpening" the skills that I once had.
Upgrade the bed to the glass bed. Get yourself some hairspray and spray the bed every few prints. You'll never get failed prints again.
On a broader level, I'm learning about creating APIs/microservices.
On an even broader level, I'm learning how to make (architectural) decisions and run projects.
So I'm learning Dart and Flutter. Flutter (which uses Dart), now has some full support for web in its latest version, so it's a real possibility I might get to propose we create some front end parts of our app in Flutter web, and even suggest we adopt it for the mobile version of our product.
We are going to do some kind of mobile version, but are currently thinking about a PWA using either React or Vue.
Apart from that, I really need to improve my knowledge of PHP design patterns.
At home I'm looking at Common Lisp and APL (which are so far out of my frame of reference I can feel my brain hurting when I sit down to them).
I used to be afraid of the water - especially natural bodies of water. This comes from learning to swim late due to tubes in my ears. So rivers carry a lot of mystique for me. I got into class 2-3 whitewater kayaking as a teenager, gained confidence in rivers, then did a bit more after college.
Recently, in my 40s, I realized “it’s ok to have interests your partner/family doesn’t”. Somehow this was a eureka for me to find more hobbies away from coding. So for me kayaking has been a part of that and my family has been supportive. Hopefully I'll find a local group that shares my interest at my beginner skill level :)
Right now I suppose I'm semi-actively learning Erlang & OTP (I vastly prefer the syntax compared to Elixir).
In the future I want to learn more about the BEAM (how it's implemented, my basic details are good but I'd like to get to internals). I want to learn about compilers too, to be able to for example contribute to BEAM. Oh, and I want to learn lisp + work through SICP.
edit: oh also just the small matter of Mathematics (from like half way through GCSE to A Level). Russian would also be interesting.
- FoundryVTT. My friends and I have tried a couple of times to run some remote DnD games over the last several years on Roll20 and Virtual Table Top, with mixed results. Discovered Foundry and am very excited to try it out, though there is a bit of a learning curve to get a game up and running.
- Procreate. I almost majored in Graphic Design as an undergrad but since switching to STEM haven't done much artwork. I've been using Procreate on my iPad to create isometric assets for the above-metioned Foundry VTT DnD session.
It's elegant but it's build-times are pretty terrible.
It has the RPC stuff built-in but I've never seen anyone use it in production and instead folks are doing redis or
something else traditional.
It has elegant process management but it's cloaked in magic (that you have to carefully learn).
Started implementing some algs in Rust, just to learn it. Liz Bradley seems super impressive as a teacher and researcher, haven't heard easier explanations for complex concepts :) The course is mind blowing to me so far.
There's more on the https://www.complexityexplorer.org/ website.
Also, in my free time during work I'm trying to improve my writing by creating technical and leadership blog posts...it's difficult!
This is not learning a skill, but it is an enabler for learning: first, using the correct sources for studying which will be identified, and first, because it will be projectedly a time-saver (processing bad information is very time consuming, a waste).
So, not sure if this fits the question - spent a lot of time learning new tech, focusing on non-tech for a short period of time. I feel that I have been overlooking that side of skill-set.
fwiw, two books helped me massively in this area:
1. The Field Guide to Human Error. This book I read on a whim just because I thought it would be useful for software development. The first few chapter's ended up being kind of life altering. It felt like along personal attack on my character. In short, it was about perspectives we take when dealing with other people, and how viewing from our vantage point is not only frequently wrong, but it's lazy.
Even with things like CR comments. I now ask myself "_why_ do they think that's the right approach?"
2. Never Split the Difference. While it's about negotiations, it deals a lot with how humans think, and despite what we tell ourselves or want to believe about we being rational creatures, emotion dominates almost all interactions. It gives all kinds of useful advice for framing conversations and using language which avoids being confrontational or accusatory.
This was another huge one for me, as it shifted just about all of my conversations from starting with "you're dumb and here's why" to "let's make sure we agree on what the problem is" and then making finding the solution a collaborative effort, rather than a top down directive.
https://www.clearerthinking.org/post/2019/03/06/want-to-impr...
I found it through HN and for me, personally, it has helped.
learning quantum-resistant crypto - Learning With Errors and variants. Because they interest me, however, little direct value to work.
Generally how to provision a server and deploy an app on it. Mostly to migrate slow-and-steady workloads off my laptop, but also as a skill in its own right.
Language learning: Julia. I tried porting some models to Rust, it has its strengths, but if you need matrices as a first-class feature - just don't.
Otherwise: Japanese for 2-3 hours a day (will hopefully pass the N2 exam in December), and have also been reading a lot of pedagogical theory lately, and will implement some of it in a curriculum that I'm proposing soon (My day job is teaching).
I have wanted to produce some useful tutorials on coding and robots for kids. I taught some lessons at my daughter's school. This gave me some great insight, but I want to reach a bigger audience online.
I am finding that thumbnail, title, and how well you structure your cuts into a story makes a big difference. I am mostly trying things out and seeing how an audience responds to a video in terms of average view duration.
Feels like there is potential for something interesting to emerge in the knowledge + social graph layer/squad/studio operating models.
Particularly with everyone moving around now and the increasingly fine lines between legal corporate entities, co-living, creative work, and intellectual property.
Open to any recommended readings on the topic, or just to chat about tangential spaces!
I've been developing https://news.ycombinator.com/item?id=27627381 using Django and its templates for UI. This became unmanageable fast. These days there are wonderful volunteers, who help me. We've chosen React for the UI and I learn it to be able to contribute to and maintain the new UI.
Will hit the books on how Data Clean Rooms work once the fall comes.
Deno’s locked-down permission system and wide support of browser APIs is really cool, and they’ve been cooking up some really cool stuff (see their new Deploy cloud service).
hope to release in a few months
On a more relaxed note, also learning more techniques about how to do proper gardening to grow more veggies and flowers.
After that getting Grafana, Prometheus and Influxdb going.
Machine learning - how it works behind the scenes was really insightful.
(Currently working as a PHP backend developer, Drupal.)
Cooking- trying to make 25 recipes for each of the Eight Great Traditions of Chinese cuisine. Just finished Fujian, doing Cantonese now.
Biking- exploring different markets and bakeries in the city I live in.
GeoGuessr.
I'm an SRE and my company has had these as pain points for years... Might need to learn how to advocate for adoption next XD.
Also looking into ML for time series data (metrics). Greykite and Kats seem to be interesting candidates.
Front crawl really seems to be one of the most beautiful & rewarding processes to master.
I've been skateboarding more and reading my books about statistics in a no pressure, leisurely manner.
Given the decline of European culture, law is on really shaky philosophical foundations. I’d seriously like to figure out if Roman law concepts can be refounded upon something like Vedanta.
I'm also smoking ribs for the first time today, that should be a fun learning experience.
It's actually quite similar to React meets Ruby or something, but without the nested dependency breakages you get with node in React. Anyway, client wanted so client gets.
I’m building up the side hustle gradually to escape the soul-sucking grind of the day job, and being a good manager of an outsourced team will allow me to take things to the next level.
Just taking it slower this time, doing ALL the examples and trying stuff for myself.
It's going way better than my first attempt about a year ago.
I have a strong hypothesis that working on my side projects full time would make me happier. These days I focus on noticing my own problems and exploring whether others have them too.
I'm 31 and for whatever reason (financial, time etc), kept putting off need to learn and acquire this basic skill set until I felt helpless in certain situations.
But I have wanted to go through the whole of SICP for a while. I realized that that is a better thing to do (grow core, fundamental skills) during the pandemic where I can get uninterrupted time.
I can learn a language any time. But I was more focused towards learning the functional paradigm rather than the language itself. It's a fun language to code in, though.
- Exploring Manifestation and subconscious mind.
Would also like to learn knapping and more about blacksmithing, but hard to find someone to learn from around here.
In most cases, it's just a big and elaborate grid search done manually.
- how to fix my xaomi m365 battery
- Japanese
- basics of koi ponds, for a gift to my mother in law
- cooking with a instant pot
- ayurveda
Computer org and design (risc-v) with Patterson
Digital signal processing
For that PDF parser I wrote a Packrat parsing engine, which I think might end up as a useful way to explore some of the possible optimizations that could be applied to Packrat to reduce its dismaying constant factors in both memory and time. So I've been learning a lot about Packrat too.
I've been learning about the Imp language, which is a sort of followup to Eve, and it has a lot of really interesting concepts in it. But I wouldn't say I'm "learning Imp" yet.
This weekend I learned to use an old Arduino with Sigrok as a logic analyzer and get Sigrok to decode the PS/2 keyboard protocol.
I've been learning about the history of political philosophy and social movements, much of which is very unsavory.
I've been learning about economics and current events. Did you know that China now produces more than half the world's cement? Or that their electrical power generation has doubled over the last decade? Or that in tropical and subtropical countries photovoltaic power plants are not only cheaper than building coal or nuclear plants, they're cheaper than keeping existing ones running? Or that the Federal Reserve permanently stopped publishing M2 monetary supply data in February, after updating it monthly for 41 years?
Last night I melted glass with a torch for the first time, but I don't think I can reasonably say I'm "learning glassblowing" yet. I didn't cool it slowly enough and it cracked as it cooled. I did learn vermiculite will stick to glass if the glass is soft enough for long enough.
This month I learned about Melisa Orta Martinez's brilliantly simple "Haplink" design for a two-degree-of-freedom mechanical actuator and encoder, originally designed for haptic user interface research, but in my view much more broadly applicable.
I've been learning about planetary roller screws, which with modern advanced digital fabrication technology could plausibly replace a lot of existing linear actuators with significant improvements in precision and achievable reduction ratios. Also I think you can use them as a worm gear to get these benefits and more for rotary motion. I haven't built one yet.
I've been learning about the RISC-V instruction set and some of the issues that go into designing and implementing such a thing. But I haven't written a RISC-V simulator yet.
I've been learning about exotic mineral cements like "geopolymers", aluminum borate, and aluminum phosphate, which can be precipitated hydrothermally as well as with high-temperature reactions. But I haven't synthesized them yet, just calcium phosphate.
I've been folding origami from strange materials. Aluminum foil, aluminum window screen, corrugated cardboard (must precompress your crease pattern), aluminum cans, plastic coke bottle walls.
I've been learning how to keep my potted plants alive and deal with insect pests.
I've been learning more about abstract algebra (the conventional kind, with rings, lattices, and semigroups, not category theory) and how it relates to algorithm design.
I've been learning about food-product rheology, and how thixotropic flow isn't quite the same thing I thought it was, nor is it caused by the same causes. Thixotropy is super important for digital fabrication.
I've been reading about filled polymer systems, especially the kinds of surface treatments used to adjust the adhesion between the matrix and the fillers. I wouldn't say I'm learning it yet because I haven't been able to get much of anything to work. But I will, if I can stay alive a bit longer.
I'm learning how widespread outright fraud is on MercadoLibre. The last thing I bought there was a "1600x1200" USB microscope which turned out to be 640x480 (from DUAITEK). I also got a "600 watt" immersion blender that turned out to be 300 watts. I'd like to take "Origins of Persistent National Poverty" for $800, Alex.
I'm learning that HN doesn't value people like me, and I'd be better off spending my time elsewhere.
But that said, I have another question that I hope you will indulge. When you said that:
>"...people like me, who do learn things quite a bit. ..."<
I had to nod: you obviously know and learn a great deal in any period of time. But I must ask: do you ever forget? In a way you remind me of Jorge Luis Borges' character Funes in the short story "Funes the Memorious". But of course you are not as limited as Funes (e.g., he is incapable of abstract thought). Also, to me, you seem to be studying so many different areas. What would happen if you focused on one, two or even a handful of projects? In any case, it seems that you are intellectually indefatigable.
I am learning more about Transformers architecture and modern NLP using the Hugging Face API.
Building a computer from scratch using nand2tetris on Sundays.
Learning to meditate following a detailed meditation guide. This has been working better than I expected.
Going on and off with learning classical music online (with a piano).
Learning German.
I am primarily focused on learning deploying AI to embedded devices right now. Working with the TinyML book and the course on edX.
As someone who does Computer Vision for a living, and always loved Electronics in college (no microcontrollers experience), I am finding it right at home with this. I wish I had started earlier. Really liking the experience.