levmiseri

A note to show real-time collaboration feature: https://kvak.io/hn

Ha! As someone who has been to Naples a few times, I fully support the 'walking hostile' sentiment! The amount of wild traffic through the narrow streets and having to constantly look over your shoulder makes it a horrendous walking experience compared to other European cities.

Content for larger cities is generally GPT4. For smaller ones I went with GPT3.5. If I were to do another pass and get fresh data, I'd probably use GPT4 for everything as the difference in content quality is noticeable.

Added the prices to the article: ~$50 for generating content for 5000 cities ~$90 for all images (Midjourney subscription)

The internet situation content comes directly from GPT4 with a decent accuracy. For something like the 'economy' content I'm using combination of GPT and some 3rd party things like the BicMac index.

I wrote about how Meoweler was made here: https://kvak.io/meoweler

The code for the website is open source: https://github.com/riesvile/meoweler-fe

This is a style issue. Not at all on the same level. The ISO standard defines a space, but many, many style guides suggest otherwise. E.g. Very often you will find advice to omit space for one-letter abbreviations (i.e. '100g' not '100 g') — I believe this specific one is in the Chicago Manual of Style.

Thanks! The site itself is built with SvelteKit. For content it's all GPT4 and Midjourney.

One other use case is — generating large amount of data on a scale not really possible before. I built https://meoweler.com - a travel site with all of the content generated by AI.

I'm a big fan of svelte vs. other frameworks. The closeness to vanilla js and natural coexistence with html+css makes it so smooth to work with and you feel productive right away. Recently built this with it: https://meoweler.com

At least for small projects like this, it's perfect. For something larger/complex I'd be a bit more weary.

Thanks! 99% of the content and all of the images. The website itself is not generated, though I did get a bit of old-fashioned help from gpt4 on some simple scripting stuff.

I have been working on a programming game https://yare.io but kinda burnt out on it. The last 5% of polish/onboarding/marketing is so hard to get through. Would happily join the group to hopefully rise my spirit a bit!

The motivation behind my side projects has been mostly having an entertaining outlet for learning new things. One of these is https://yare.io that taught me plenty about building something end-to-end. From concept, design, front end, back end and failed marketing efforts. Despite 0 revenue, the experience itself is greatly rewarding.

This doesn't really fit there, but I tried to create a short sci-fi story written purely by chatGPT: https://kvak.io/?n=chatgptstory1

(the chain of prompts is at the end of the story)

Thanks! I'm working on a small album that I hope to put as some background music + sound effects into the game.

To shed some light on the somewhat unusual controls – there are some additional considerations I had to make that don't necessarily exist in other similar RTS.

1. The fact that it's all in the browser doesn't allow for panning the map with mouse-over-edge.

2. People often play the game with mouse and keyboard AND typing code at the same time, so the controls had to be tailored to minimize conflicts

3. The whole idea behind the current manual controls UI is to highlight what is possible to program for the players themselves. It's not 'technically' part of the game, it's just an injected html/css/js so the players can edit the controls to whatever would feel more natural (and add new buttons or completely change the mechanics).

But you are not the only one highlighting these issues, so I'll think about how to make it more intuitive and abide to conventions as much as I can.

There is a bug for some users where the manual controls aren't properly activated, sorry about that!

If you have an account, click on the name in the top-right corner and activate the 'manual controls interface' module.

I'm somewhat embarrassed by the code quality (dev is only a hobby, my day-job is interaction design), but as soon as I clean it up a bit, I'll open source it.

I tried to combine both of these approaches. The link takes you to a page that explains the game with screenshots and a trailer - no need to play it right away.

Thanks! Found the bug causing most of the crashes, so it should work better now.

I think that if you give it a chance, you will find that there is a lot of strategic complexity even with only 1 unit type. Here is an example of a game between 2 humans: https://yare.io/replay/77cg1k232811g0

Thanks! The first run experience is something I'm definitely struggling with and will improve. Your issues might have been caused by the sudden traffic increase - I made some changes on the servers - it should be working a bit better now.

Not a full-time developer, but I have been struggling with developing my first game a lot as a side project next to a regular day-job. It's already in a kind-of beta state (https://yare.io), but still far from finished and would love to be part of such group!

This couldn't be further away from what I really wanted. Way too much friction, lots of features and noisy UI... I'm sure it has its target audience, but it isn't for me.

Agreed! I'm planning to update it in a few weeks with additional features including real-time 'multiplayer'.

One such simple project for me was https://kvak.io. A weekend crafting for a need I had and couldn't find anyone offering - a simple online notepad where I can share the note by just copying the url and the 'sharee' continues writing.

The process itself is indeed very fulfilling and joyful.

Consistency is overrated. Context is what matters. And context often requires customized controls.

We will never live in a world where an abstract concept like a checkbox can have an 'assigned' specific visual affordance that doesn't allow for adjustments. Such thinking is stuck in the past and won't allow for new/better UI paradigms.

For people that want to learn/improve in javascript in a more 'entertaining' way — I made an educational javascript RTS game: https://yare.io

For the fellow .txt filists — I made a minimal online notepad https://kvak.io that makes it easier to access your notes anywhere. (With my wife we are also using it as an always-up-to-date shopping list).

I really enjoy an article "Learning to See" on a similar topic from designer's point of view - from the founder of iA: https://ia.net/topics/learning-to-see

A small excerpt:

> Personally (dis-)liking a color, form, or image is not a matter of design, it is a question of personal taste. And as we all know, when it comes to personal taste there is not much to talk about there. But in addition to personal taste there is something that we can call “trained taste” or “sophistication”. Let me recapitulate:

- Whether I like pink or not, sugar in my coffee, red or white wine, these things are a matter of personal taste. These are personal preferences, and both designers and non-designers have them. This is the taste we shouldn’t bother discussing.

- Whether I set a text’s line height to 100% or 150% is not a matter of taste, it is a matter of knowing the principles of typography.

- However, whether I set a text’s line height at 150% or 145% is a matter of Fingerspitzengefühl; wisdom in craft, or sophistication.

There is one last major piece to be done for the game (now I’m no longer working alone on it) after which I want to write something a bit more detailed about the journey.

But on the high-level — my full-time job is interaction design. Throughout my career I used JavaScript a lot for building interactive design prototypes and it’s the only language I know, so when I had the idea for Yare.io (heavily inspired by MIT’s Battle Code), vanilla JavaScript (and Node for server) was the only thing I could use (didn’t know any libraries or frameworks)

The project was really just a “problem” to be solved. Use JavaScript to move basic geometric shapes on a canvas in a 1 versus 1 battle. It needed to have a UI, rendering of a game state, authentication, event queue, basic ruleset, … None of this really required any tracker or rigorous process. I know what needs to be done, because I’m literally sitting in front it, seeing what needs to be done. It didn’t need a “plan”, because it didn’t matter when each piece of the puzzle was made. Just, whatever I was in the mood for that day.

I think the principle of simplicity (as cliche as it sounds) – trying to keep everything (especially the foundations) as basic as possible – was really the main thing that allowed me to finish the game.

I don't think I could enjoy it as much as I did with some scrum method, brainstorming bullshit, or anything reminding me of work.