Building the MacPaint for hardware | Hacker News Reader

Building the MacPaint for hardware

medium.com

33 pointsdivy22 comments

It's interesting how he talks about the raspberry pi:

> The first was listening to Eben Upton, the brains behind the Raspberry Pi, describe that the number of qualified applicants to study computer science is dropping. Rather than learning how computers work at a more fundamental level, students are applying only with knowledge of webpage construction and high-level scripting languages. He argued that this is emblematic of a larger generational problem and the Raspberry Pi was his solution.

>> “Our idea was to build something cheap, powerful and available for children’s bedrooms so they could have the same experience we had.”

My impression of the Pi is that it's too complex for this sort of application. It's cheap and therefore easy to procure, but it runs a desktop operating system that is arguably more difficult to work with than the one on your regular desktop PC, and it has a weird architecture consisting of a CPU as a coprocessor of a proprietary GPU, with an inscrutable binary blob managing everything underneath. That's just too much to deal with if you're aiming for the 80s bedroom computing experience. I think there's still room for something that fills that niche. A modern "Commodore 64" needs to provide the same unfettered access to the internals, with similar simplicity, but with modern performance and amenities (better editors, better languages, built-in documentation, flexible communications and control options, etc).

The closest I've gotten to that simple 80's computer experience is imaginary computers in Minecraft mods. Not exactly real-world applicable, but still a lot of fun. They're Lua scriptable with a simplified notion of networking and robots.

OpenComputers is a bit more interesting than ComputerCraft, but it also turns the initial construction process into a giant multi-stepped resource-intensive pain. Recommend playing in creative mode if anybody's curious to just take a quick look.

I was never a C64 hacker, my first computer was an Atari ST. It sparked my interest and lead me towards buying a 486 some years later.

I open with that because I could very well be wrong...... But surely at the time the C64 had a "weird architecture" and some proprietary blobs?

I do think that loading desktop linux onto an R-Pi results in a different kind of experience to a kid firing up a C64 and finding themselves at a basic prompt. However, it's pretty easy set up an R-Pi so that you're greeted by something similar.....

Note: I've got 2 original R-Pi's and my R-Pi 2 arrived today. I also have a 1 year old girl who I hope to one day get excited about the kinds of things that these wonderful devices enable me to do play with.

The C64 came with a manual including the full memory map, what addresses to PEEK and POKE to control the sound chip, etc. "Proprietary blob" may be technically true in that the ROMs weren't open source, but it was simple, accessible, and documented enough that that doesn't well describe my experience with it.

(Though it's not a machine I ever got deep into, myself.)

Yes, exactly. Strange as it is to say, possibly the closest thing running on modern hardware would be TempleOS.

Agreed, the system isn't the "C64 of the 2010s" that it's been proclaimed to be. Unless there's a wave of success stores out there that I haven't seen. Mostly it's people using them as beefier Arduinos for their projects.

But I wouldn't go as far to say all the Broadcom blob-nonsense is the hindering factor. That's completely a non-issue if we're talking about education. The reason the C64 and Apple and BBC Acorn were they way they were was because they booted right into a programming environment.

Where's the autostart environment for the RPi?

It's not just booting into a programming environment, but being supplied with documentation that had memory maps in and taught people how to link assembly language routines into their BASIC programmes or how to POKE or PEEK.

I would say in RISC OS Pi.

One of the early prototypes of the Pi was built around a microcontroller that bit-banged video output. It was a lot closer to the ultra-low-level experience that you describe. He decided it was just too alien an experience and too limited compared to a modern desktop PC - there's no way it could ever run a web browser, for example.

A fundamental problem with the "modern C64" ideal is that the software stack is much deeper these days, especially in the browser.

The nearest thing to your hardware-orientated version of simplicity would be something like this: http://hackaday.io/project/3877-layerone-demoscene-board which outputs raw VGA from a PIC24.

Isn't this more like "Lego" for hardware?

In MacPaint I can paint anything. In your toolbox I can only assemble something with the parts that you give me.

The MacPaint analogy would hold water if you'd let me draw any shape that I want (to be 3d printed), and merely assist by enforcing the basic constraints required for my device to be manufactured and functional (e.g. enough space to put the electronics in, etc.).

To be fair, there were many constraints in MacPaint - color being the most obvious example. It was a limited drawing program. Put into context, I do believe the analogy works - you start simple (box, few components, limited layout) and evolve to include malleable 3D printed materials and generated electronics.

Disclosure - I wrote the post.

Quick question, perhaps: what's the maximum computing power, especially RAM, you're thinking of offering?

If it's just microcontrollers, my interest in somewhat limited. Something I reasonably put a Lisp on (code as data wants RAM, even if a lot can live on flash) ... well, looking at Wikipedia's list of Arduino CPUs, 96KiB is a very qualified maybe, 64MiB is a lot more like it.

I think it's a bit looser of analogy, where MacPaint was one of the first desktop computing apps people saw and immediately realized how much better than the old command line tools something could be. Word processors, spreadsheets, etc. worked fine with all text but MacPaint was truly special and couldn't be replicated without a full desktop environment. From what I gather in the article they want to do something similar where they build a product that makes people realize 'oh so THAT's what I can do with all this Arduino, Raspberry Pi, 3D printing, etc. tech!'

The mouse was a must for MacPaint, but it didn't run in a 'full desktop environment' at all. It blitted something looking like a window onto the screen. If it used the OS for doing that, that only was because it took up less memory that way, not because it wanted to behave like it ran on a desktop environment.

Did it have movable windows? No.

Did it have resizable windows? No.

Could you move the tool palette around? No.

Could you open multiple windows? No (no surprise, given that a full-page painting took a whopping 28kB or so and the OS didn't really have enough RAM to store that.)

I think whats needed is a revival of the Denshi Blocks paradigm, but with a lot of rPi, 3d-printing, eurorack-modular-synth, and a whole lot of open source mixed in:

http://en.wikipedia.org/wiki/Gakken_EX-System

Imagine a new attempt at making something similar in the modern era ..

It sounds like your basically describing what littlebits are doing (http://littlebits.cc/) - modular electronic blocks that can be arranged in a multitude of combinations

Sure, thats pretty close. But .. take it up a notch. Make it an open spec that anyone can manufacture. Make the block stacking standard really work .. littlebits has its virtues, but the whole thing is quite delicate and really only suitable in the lab/workbench. A 21st Century Denshi Block system would have to be a lot more resilient to spills and bumps.

> Retro-computing is a weird hobby of mine.

I hate to criticize the very first sentence, but I want retro-computing to be mainstream, and calling it "weird" seems to cover for the author's lack of confidence in his hobby. I'm trying to find a rationale for calling one's own hobby weird--I can't find one.

I have technical hobbies--resurrecting a Lisp Machine, collecting 80s and earlier computer keyboards, duplicating websites in Emacs--that I call "hobbies", not "weird hobbies".

Related examples (especially to my "covers for lack of self-confidence" point): https://www.reddit.com/r/AskReddit/comments/1mspk9/reddit_wh...

https://www.reddit.com/r/teenagers/comments/1e6pzh/rteenager...

i think you are taking it too literally. What I normally mean when i say it in this context is fetish. Something other people might find weird but i enjoy immensely.

> i think you are taking it too literally.

If you're making a spectrum between "taking something literally" and "accepting something as a throwaway comment", and saying I went too far in taking something seriously or literally:

I'm looking at the /effects/ that word has on its readers. The author (are you the author? hi) doesn't /intend/ those things I said in my grandparent post. But the effect is (in my estimation) the author looks unconfident in his hobby. And the secondary effects are: people are /less/ likely to feel invited to join in. I would prefer the author invite me into his hobby by confidently showing me how cool it is.

(I'm stating my preference in order to make my point understood; I understand the author didn't ask for it.)

I think your annotated exception and interpretation of this phrase is something of an overcorrection here. Leaving aside how the author may feel about his own hobby and what justification you have for speculation on the point, the word "weird" is a versatile one and could very easily be a way of saying "uncommon" or "highly specific."

"I don't think retro computing is a weird hobby" would seem to say all you want to say, but in a sympathetic and positive way that doesn't turn things around on the author.

littlebits.com as software and more professional.

Reply on news.ycombinator.com