Ask HN: What's your quarantine side project?

So, when you join the app, you will go through onboarding flow where we try to understand your movie genre interests, as well as, you can opt-in to give suggestions in the future (this way you become part of community). So, once you submit a request, app will send those request to selected ppl from the community asking for a good suggestion. Once they suggest movie/TV show/series, you will be notified and will be able to pick those that you like/dislike. App will try to understand who are ppl from the community which suggestions you liked the most and in the future will connect you profiles similar to those people. That way you will have your unique path. The main reason for this is when we did PoC most of the participants stressed that they take suggestions from people they know and share similar taste. But again, we are in stage of releasing our MVP, so I'm sure there will be more things to learn :) Hey, why not subscribe for "Early access" and help us crack that part :)

Great, thanks for the feedback!

> One thing that may make this easier is beveling the edges of the tiles slightly so you don't accidentally select letters on either side of the diagonal unintentionally.

Thanks, I'll have a play with that. If you use your web devtools to inspect the HTML, you should see over each letter tile, there's actually an invisible tile on top of each one at a 45 degree angle that's being used as the real touch/mouse target for selecting letters (as accidental selection is awful if you use the actual tile as the target). Maybe there's a more reliable way but playing with the target shapes and sizes will probably help.

Thanks, stream of consciousness feedback is ideal. :)

> After a few levels I found clusters of letters I was having trouble with building up over time (there was an “X” right in the middle, and like six “I”s clumped up). I found myself wishing there was some mechanic that could clear them out. Maybe that would make it too easy though.

Yep, this seems worthwhile looking at. Maybe smarter letter randomisation would help this e.g. don't allow a new "I" tile in a place that has two "I" tiles already, don't allow clusters of consonants.

> I really love the animation when you make it to the next level, very satisfying.

Great! I thought I needed something flashier so I'm happy this could be enough.

> On mobile safari, holding your finger on a tile for too long causes a text selection.

Ah, thought I caught that. Thanks!

> I’d love a “zen mode” with no timer!

Yep, I want to figure something out for this. It's pretty fun thinking of different scoring mechanics and how this impacts gameplay strategy.

Thanks! You mean "zen" was allowed but shouldn't be?

Thanks! Yep, mute is missing and I'll add it. I was trying to push myself to launch an MVP without over polishing it.

Great feedback, thanks! Sounds like I need to make the game rules clearer but hoping I can do that without explicit instructions somehow (it's a fun design challenge too). Let me know if you've got any ideas.

> 1. "?" is a wild card, right? (meaning it can stand in for any letter)

Yep! It'll autocomplete to a valid letter as you use it - there should be tons of valid 3 letter words so I was hoping people would figure it out with a quick bit of experimenting. Try just making up a long word to see if you can find one for big points (at the cost of the time it takes you to guess).

> 2. I once got a diphthong "qu" in a single tile - not sure if it was a glitch?

Yep, there's no "q" tile, only a "qu" tile to make it easier to spell something. Adding diphthong to my personal dictionary!

> 3. Why is a tile sometimes yellow?

Scoring explanation:

- Word scores directly depend on the length the word and goes up quick e.g. for a 3 letter word you get only 1 point (a deliberately lame amount), then 4 points for 4 letters, 9 points for 5 letters, 16 points for 6 letters, 25 points for 7 letters.

- The exception is yellow/bonus tiles are counted like they were two letters when calculating the word score, so one or more bonus tiles in a word grows the word score massively e.g. a 5 letter word using 2 bonus tiles will get you 25 points instead of the usual 9.

- You get rewarded a bonus tile if you spell a long word (this one for sure isn't obvious unless you play several games).

One of the core strategies is to earn a few bonus tiles with longer words, then make maximum use of them to level up by spelling words using multiple bonus tiles at once.

For the scoring, I was hoping for players to go through this thought process 1) "hmm, 3 letters words only give you a single point" 2) "4 and 5 letter words get way more points, and the score goes up rapidly by the length of the word" 3) "the bonus/yellow tiles boost the score up even faster". Sounds like this needs more work though so thanks.

Thanks, that's useful feedback! I agree some colour variations would be good for guiding the eye.

I partly moved from 5x5 originally to 5x6 because it looked better on mobile/portrait but then I found in play testing more letters gave people more room to manoeuvre when the orange/bonus tiles start appearing (where you want to delay using them).

Thanks, it's more work than it looks which you'll know. The swiping code was a bit of a pain to be honest. The web APIs for handling mice and touch events together aren't well unified.

Haha, we have the same confetti effect at the end. Nice work! Is the situation with iOS and PWAs any better since you last worked on it (I saw your note about a few issues on the GitHub page)?

Thanks! Glad the timing didn't feel stressful. I've had some people say they don't generally like timed games but when it was game over here they felt it was fair instead of blaming the game - seems a decent thing to aim for.

You actually get an extra second before your time runs out to increase the chance of "just in time" saves. :)

Yeah, planning to promote it on Reddit and ProductHunt so hopefully can gather more users and reach the critical mass.

Cross provider sync is on our roadmap but not available yet. If you want, send me an email with some more details and I'll look into the feasibility.

No. Right now it's just our two weekly quizzes (Tuesday and Friday) from Ireland, one quiz from California (Friday) and a couple corporate quizzes we host for companies, maybe one every two weeks.

As soon as it is working!

I've been following Caddy and what they're doing makes sense.

Certera aims to fill a gap in centralizing and managing LE certs and allowing those certs to be used in more places and scenarios.

Yep, exactly. I was originally pretty excited about Rust, but I feel like it biases toward large highly-structured projects like C++ does, making it a bit daunting for personal projects. Just a bit too professional for something I'm hacking together over a weekend, lol. Zig feels like it's easier to wrap my head around, and seems optimized for the "pet project" use case.

I also feel like learning Zig is helping me better understand C. I probably wouldn't have been able to contribute much to that Wine patch if Zig hadn't already gotten me more comfortable with pointers, statically-allocated variables, and such in a reasonably-safe way (having prior experience with Perl did help a little bit for pointers, since referencing and dereferencing variables is pretty common in Perl codebases, but it always felt a bit detached from what the machine was actually doing behind the scenes).

Email sent (expect one from northrup at (my HN username) dot us).

Reminds me that Zigler's yet another reason why I've been wanting to learn Zig; I usually go with ports over NIFs, but should that ever change I'd definitely rather be doing it in Zig than in C (or Rust, as similarly-neat as Rustler may be). So thanks, and keep up the good work!

Yep. Currently, I only parse their RSS feed. Later, I want to extract and organize information in a way that's easy to search, look at and save.

Thank you, great insight, I'll give it a thought. I want to have a focused, narrow scope for the MVP, but I think finding and digging (public!) financial information is a mess currently, even with all the apps, blogs, yahoo finance and the likes out there

I'm using Rocksmith and Yousician (the free version), but I think there's room to improve over what both of them provide for practice / exercises. For learning and performing songs they're definitely the best due to the sheer amount of content they have.

I'm very early in my learning journey so I'm not concerned that developing a prototype version I could use for myself will take too long (it could be that I'm underestimating the difficulty of the problem). I was already able to slap together most of the "glue" code in Unity in a couple hours, and given my signal processing background I have solid ideas for how to approach the difficult algorithmic parts.

NodeJS wrote a custom little api for cheddarflow and Robinhood so it can check unusual options activity and contract prices instantly(ish)

Always 100 shares

Yeh your right, It’s programmed to not mess with anything where the bid ask is >3%. Otherwise yeah, it would get eaten alive. It uses a combination of unusual options flow and some basic technical analysis to make trades and has been doing well so far but... it is easy to lose a lot quickly if ya blow it.