I'd be up for a meet, do you have a preferred email or messaging app?

Nit: I think that when an exercise was just attempted, it should not come up immediately again (soon, yes, but w/ some enforced delay, maybe) — otherwise it feels a bit tedious (even though it's probably decent for internalizing it)

- stack A needs lots of tedious, not-quite boilerplate not-quite-copy-paste work whenever you implement something, but reduces cross-OS bugs, UX and accessibility issues down the line. Lots of dev frustration, slows down iteration (when humans actually have to do this).

- stack B has awesome DX and affords smooth operation of the design/UX/dev teams, very fast. More issues down the line, though. May be worth it if unaided humans need to work on it.

If you're looking for feedback: It's easy to get started (pleasant that you are not forcing a login), but not so easy to understand what to actually do — what's the main flow? Which buttons am I supposed to press to do what? Some onboarding may help.

Since I'm not really in the market to learn English: Are you planning to offer this for other language pairs?

Hello HN,

...I keep making little web apps to practice geography, and since some people here enjoyed the previous version, I thought this one too. It's a simple game loop where you have to either tab or name countries, w/ some varying conditions like zoom levels.

Following feedback from last time, I reconsidered almost all of the game systems, fixing quite a few issues (too hard, too easy, too many boring islands, countries recurring often, confusing UX, ...). The stack is ts/vue and geojson, with some python for data crunching; learning algo built on top of FSRS.

It's all free, no-ads, no-signup and open source, in case you were wondering.

Feedback and ideas welcome!

Hope you get something out of this, cheers :)

Comments URL: https://news.ycombinator.com/item?id=49495618

Points: 3

# Comments: 0

yeah if it's no effort, please share, otherwise I reckon I'll rebuild it myself some time soon :)

I'm a bit too beginner to effectively use the app, sadly, but it's quite an interesting flow.

As an aside, a while ago, I sort of prototyped the reverse: Learn vocab to understand a natural language sentence, translate the sentence (https://infi.koljasam.com)

Happy to talk and exchange notes, if you want :)

I think this has been my biggest hurdle with LLM-based learning, even though I think the concept may be quite powerful...

Happy to hear more about your approach :)

website: https://lectures.koljasam.com/

repo: github.com/koljapluemer/awesome-lectures

Hi everyone,

this submission is my attempt to contribute something to the question of how we will find, share and evaluate information in the future, in the face of the many problems that I think HN is well-aware of (slob, search engines returning SEO garbage or corporate stuff, spam, trust problems, big companies eating everything,...).

I was specifically inspired by recent threads here on HN like https://news.ycombinator.com/item?id=47625952 , so I hope some of what I'm saying here resonates, even if it's very much not comprehensive (nor that coherent).

To get specific, I decided to build a little platform (if you want to call it that) to submit, rate and find long-form recordings of talks/lectures; both because this a personal itch of mine, but also because I consider it a good low-stakes environment to test some (possibly larger) ideas, such as:

- having a community-information-curation platform where both code and data are managed via an open git repo, but where non-technical people can also contribute, without interacting with GitHub or code

- allowing submissions and contribution without any gatekeeping (=no sign-up), keeping in mind that spam exists

- running such a platform on close-to-zero-budget

- allowing to (at least eventually) democratize the platform completely, so that it doesn't die with its maintainer's motivation

The current tech stack is as follows:

- data source of truth: JSON files in an open repo

- main interaction point: Classic HTML-based static site, works without js if you want to, generated with a custom SSG

- rating; submitting lectures; proposing changes: JS-based features on the static site, talking to a Flask backend via an API, writing change proposals temporarily to a DB

- Flask backend: Allows moderators (currently just me because no one else knows about this project yet) to accept or deny submissions (=spam prevention). Accepted data is written to the JSON-source-of-truth (using Bayesian updates for ratings and some semi-fancy merge algos)

Crucial idea here:

1) The data source+the viewing modality can be replicated, modified and viewed very easily, even if I get hit by a bus or get bored of the project

2) When the backend gets spammed (likely with no-auth submission affordances), hammered, hacked or neglected, (view) access to the data is not affected

So while I think that it in fact would be great if there was a great community-driven list of lectures, I also think this concept may as a (dis)prove or concept for far more important platforms, e.g. for discussing scientific findings, finding truth, storing community knowledge or stuff like that.

Just in case this wasn't clear yet, everything is still very experimental and work-in-progress (but feel free to give brutal feedback anyways)

Anyways, truly any kind of feedback or contribution is really welcome, be it in the form of lecture submissions, ratings, technical help, philosophy, or whatever else :)

Thanks and let's try to keep some kind of human internet alive.

post edit: formatting

Comments URL: https://news.ycombinator.com/item?id=47720295

Points: 5

# Comments: 0

It seems to be simply a great idea...like...should we bring it back? Could we?

Comments URL: https://news.ycombinator.com/item?id=47466588

Points: 3

# Comments: 0

Comments URL: https://news.ycombinator.com/item?id=47451372

Points: 1

# Comments: 0

- Doing anything usually involves prep work. Want to take a step? First put on your shoes (literally or figuratively, depending). If your attempted habit is 70% prep, your brain will somewhat rightfully conclude "this is stupid" fairly quickly.

- "Just do X every day for [long time period]" has an inherent falsification problem: You aren't "allowed" to argue against it until you tried it. Stopped after 2 years because you saw no change (and 5 was recommended)? You are still not allowed to argue against the strategy!

- You can actually make steps so small that they're useless. I once set out to have (at least) one github commit online per day (going for that green tile!). This led to my brain finding hacks like rephrasing one sentence of an old blog post. Doing that for 20 days is way less effective than one single coding session, at 20 times the emotional cost.

- Doing something daily for a long time is extremely hard to achieve, especially if it's not the main thing you're doing. It's rare in the wild. You will find piano virtuosos who play piano daily, but not piano virtuosos who also go to the gym daily.

Code is an abstract way of soldering cables in the correct way so the machine does a thing.

Art eludes definition while asking questions about what it means to be human.