Comment by dfee | Hacker News Reader

Comment by dfee | Hacker News Reader

I bought one, and of course it hasn’t arrived yet (though I think it will in Q3).

The best parts for me are the following:

1. It’s indie.

2. It’s accessible.

3. It’s got a completely different engagement model.

I have a 7yo who I bought it for (really) and with the web drag-and-draw IDE, I can’t think of a better way to get him involved in the exploration of technology. Maybe it can be fun and provide an outlet for deep exploration. I sure as hell am not going to sit down with him and explain the principles of enterprise architecture.

For engagement models / indie-ism, we have a Switch and just got a PS5. They’re neat. I guess. But they’re also kinda stale and boring (outside a few winner games). Kinda like watching Disney these days with the Avengers series. Diminishing returns on every title and no real way to engage beyond consumption.

I’m really hoping this is good for him (and good for me as a reset against my day to day as an engineer indoctrinated in the Bay Area ways of growth). To that end, the other comments I’m seeing don’t seem to understand the appeal that has attracted me.

If your 7y/o is into the switch, you might give Nintendo's Game Builder Garage[0] a shot; it's a pretty fun engaging node-based visual programming environment for building games on the switch.

My 7y/o has really enjoyed it, and she talks to me a lot about the stuff she's building.

[0]: https://www.nintendo.com/store/products/game-builder-garage-...

Appreciate the tip! Will check it out!

and Dreams for the ps5! https://www.youtube.com/watch?v=Mw0tW5FNEWQ

And also Labo. It has a visual programming environment.

And it matches pretty closely to visual programming, like Unreal Engine Blueprints or building shaders in 3D modeling software or Unity, for example (saw a lot of node types that almost mapped 1-to-1 to some in Blueprints, for example).

But it gives each type of node a fun and unique personality that kind of matches its function, and has a very 20+ hour long (and entertaining) tutorial series, where the elements chat back and forth to each other as they handhold you through the process of making simple but playable games in multiple genres. Put a smile on my face multiple times, and I'm an experienced developer that just tried it for a lark.

I had the same hopes for my kids with PICO-8, but it didn't really pan out— the things they were excited to do (make a platformer!!) all require a ton of boilerplate around movement, gravity, collisions, sprite animation, and so on. The high profile projects on the platform are all super polished and based on endless hacks and tricks that push the boundaries of the system, and it can be hard to get away from those expectations.

GameMaker has gotten pretty professionalized at this point, but back in the day I think it was a good entry point, with decent non-code workflows to get the basics going and a smooth on-ramp to adding behaviour code as needed.

I don't know what fills the hole that used to be platforms like ZZT and MegaZeux, though. Maybe it's Minecraft and Roblox that are the modern outlets for that kind of creativity.

Pico-8 isn't super accessible as a game making platform, although it is a very accessible programming introduction platform.

HN tends to over use "boilerplate" as Pico-8 has zero boilerplate and is super minimal. Your post is no exception. Core gameplay code is not boilerplate and Pico-8 is not a game engine in that sense.

Playdate has Pulp, which does include built in collision, dialog systems etc and is inspired by Bitsy. Pulp has it's own (optional) scripting language and is specifically designed as an accessible game making tool. Even the Playdate Lua and C SDKs have more gameplay functionality than pico-8 though, including a simple game object model, collisions, animation, etc

I think he meant that Pico-8 is so stripped down you have to spend a lot time writing lower level game engine stuff before you get to the fun, moving things around, stuff. Not boilerplate in that you are repeating code within your Pico-8 game, but that you have to rewrite basic functions that come built in to typical game frameworks.

Exactly. And it's not that PICO-8 is bad for it; in fact PICO-8 is marvelously flexible since it imposes so little on you in terms of structure, so the games are a lot more variable than, say, something like rpgmaker, where the results can end up feeling a bit samey (until you have something really exceptional on the art and writing side, like Rakuen).

But yeah, as far an introductory programming platform, it's much better of think of PICO-8 as an upgrade from QBasic than an upgrade from ZZT.

I would always start with Scratch

https://scratch.mit.edu/

It has build-in collision detection etc. but to me turle graphics is the way to properly understand programming. Most other environments help with getting people interested in tech (and perhaps game development), but visual configuring event handlers is not programming. When I taught programming I often got people from other courses with an visual environment. They knew how to build forms and save data, but they could not programm: (to me) Think about a problem, make the problem into smaller problems until you can write code for them. Many were happy when we did switch to turle graphics and said "We should have started with that".

Scratch helps transition from the event handler configuration (which is fine if you want to write games, many environments don't make you write code) to coding (write code in a programming language) via turtle graphics.

Or Snap!

https://snap.berkeley.edu/

My daughter uses a couple of block-programming browsers environments. One is Frozen-themed where you direct the characters to draw by skating as directed. Modern turtle graphics.

> 1. It’s indie.

Shamelessly, this is what I'm excited about most. There's some big little names attached to the project, and I'm excited how the limitations will stretch their creativity.

The limitations also make it accessible to many more people too. A few months ago me and two friends got together and spent a weekend making a little game in that format. (the sdk wasn't out yet but we just used love configured for its resolution and 1bit color).

None of us had ever made a game or even a serious attempt at one before but we had a playable proof of concept of a mechanic we could iterate on by the end of the first day. Similar thing with the art. Making truly great pixel art under those constraints is very hard, but making good enough to play with is much more in reach than it is with high fidelity engines.

Reply on news.ycombinator.com