Launching the Handmade Software Foundation | Hacker News Reader

Launching the Handmade Software Foundation

handmade.network

63 pointsDeathArrow39 comments

I never understood the Handmade Network. AFAIU, it came from people who watched Casey Muratori's Handmade Hero, a game developed on video over several years from scratch. But Casey, as far as I know, didn't start the Network and it never seemed to align with the intent of Handmade Hero.

The purpose of Handmade Hero was to show people that they are capable of making a game themselves and to learn things which have a reputation for being too hard. There was, of course, an emphasis on the hard things being hard because of complexity introduced by things like OOP, C++, etc. But the main purpose always felt like education and enablement. Casey's a great teacher and the videos are very informative.

The Network, on the other hand, was some weird "we want to make stuff by hand", whatever that means. That's fine. But that's not what Casey spent like 7 years doing. He didn't do it "just cuz". Instead, it was to teach and share. That seemed lost on the Network.

As a result, it seemed just like a less toxic Suckless project without the focus on making a new ecosystem. It was just a forum to say, "Hey I made this thing", all the while co-oping the feel-goods from Casey's Handmade Hero.

To be fair, Handmade Hero also seems like a project designed to co-opt unearned feel-goods. Maybe one of the goals was to teach, but another goal was to actually ship a game, and he took pre-orders for it before eventually abandoning development. It turns out it's a lot easier to talk about making good programs than it is to actually make them. I do think it is possible to make high-quality handmade software, but being performative about doing so rather than just doing the thing is probably counterwise to ever actually just doing the thing.

I for one feel like my 15$ spent on Handmade Hero were well served by having access to the source code and the breadth of video that annotates every line of code. I think anyone that looked at he proposition that Casey made as something more than a way to support him, was naive.

The goal was always, first and foremost, to teach. This is super obvious from the announcement trailer alone[1], where he says the point of the project is to pass on a way of life that inspired him. If Casey wanted to make money from a game he wouldn't have bothered with thousands of hours of Twitch streams.

[1]: https://www.youtube.com/watch?v=A2dxjOjWHxQ

It's a large community of programmers whose values align with those demonstrated in Handmade Hero. Naturally not all of those programmers are going to do carbon copies of Handmade Hero. Some make and publish their own software that is developed with similar values in mind (e.g. File Pilot, 4coder, Odin). Others just enjoy programming that way for fun and like to discuss their hobby projects with other like-minded people. This shouldn't really be surprising; this is to be expected of any online community.

>What will the Foundation do?

>The number one goal of the Handmade Software Foundation is to support, promote, and sustain the development of Handmade software.

I would love some concrete ideas what this means. My primary concern is that if any money is involved, like stipends for handmade software it will be gamed and there is no way to monitor if LLMs are used or not. In current world it is hard for me not to be cynical when I heard about new good thing, but also they are asking for money and they don’t really tell what they will do with it

They don't mention the possibility of providing funding for developers anywhere in the article.

> Basically, the thinking goes that Handmade programmers have the technical chops to make amazing software, but don’t always have the aptitude or desire for the many, many other tasks that go into shipping. Payments, licensing, emails, support, design, marketing, testing, the list goes on.

Instead, it sounds like they want to take on the role of a publisher. Perhaps providing publisher-like services without any money changing hands between developer and publisher, with the latter being funded by donations and the former still having to make their own bread, just with some of the development-adjacent work offloaded.

It’s going to be next to impossible to vet people, but I guess if they take it really slow and get to know them it could work

Unless of course “handmade” doesn’t mean what I think it does

Why do you think it would be impossible to vet people? Publishers vet the developers they work with all the time.

Edit: Upon re-reading, I actually missed this paragraph the first time.

> Membership will grant you access to a private Discord channel with other members, access to the aforementioned business resources, and possibly more benefits down the line. We have other ideas for Foundation activities, but we don’t want to distract ourselves from our primary goal as we get the Foundation off the ground.

It sounds like you get the publisher-like services in exchange for paying them. So in that sense they probably don't care if you're vetted because you're giving them money? Assuming the membership fee is sufficiently large. They are also talking a lot about "community", in-person meetings, etc. so I assume a close relationship is expected, though.

I really thought that “handmade” meant “no AI” which feels borderline impossible to verify, but now I am not sure what it even means

The "Handmade" brand goes back over 10 years, predating LLMs. The central premise is understanding the code you write at the machine level. Not that the self-professed "Handmade" coders always write in assembly, but rather, they can if they need to, and when they use high-level abstractions, they understand exactly what they're abstracting over (usually having built most of the abstractions themselves rather than pulling in 100 dependencies as is the norm nowadays).

Okay. I guess it is cool, but personally feel like the sweet spot is somewhere in the middle, but you do you

No, we're still going to vet people. There's a slice of benefits we can offer to everyone who pays to become a member, but then on top of that we'll be able to provide tangible support for some number of projects as well, funded by the membership dues. Which projects we choose to support in this way will be at our discretion, and vetting the developer and the project is an obvious part of that process.

The thing we can offer developers, which is rather unique, is bespoke "Handmade-style" handling of many dull aspects of professional software: payments, licensing, email lists, websites, support, etc. The goal is to do these things without locking the project authors into a third-party platform; that is against our Handmade values and there are plenty of such products on the market today.

Yes, it does sound like an AGPL (hopefully) version of Stripe + maybe the software stores like F-Droid. At least that's what I'd want this to be.

They mention funding projects like 'File Pilot' but it doesn't seem like file pilot will be the first project they'll fund. They didn't mention what % will be going to events and I personally don't want to fund those

There doesn’t seem to be any relationship between Handmade and “not using LLMs” (besides one that can be argued if you believe LLM-aided software is worse in size or performance).

I would imagine, or rather hope, that "hand made" is not entirely just a metaphor for the people supporting Handmade network.

Okay.. so what’s the point?

The Handmade community has been around for a decade now, well predating LLMs. Our values derive from Handmade Hero, which promoted a love for the craft of programming and an engineering mindset informed by deep understanding of how the computer and the platform actually work. The hallmark of the Handmade mindset is enthusiastically "reinventing the wheel", because building software from scratch, in the context of a specific problem, with a deep understanding of how computers work, can produce shockingly high-quality software.

Many people in or around the community have made great software over the years (4coder, File Pilot, RemedyBG, the RAD Debugger, Essence OS, Odin, raylib, kb_text_shape.h, Spall, etc.) but because it's a community of programmers, expertise in many "soft skills" is often lacking. This often prevents Handmade software from being a bigger success and sustaining its own development. Our goal is to fill in a lot of those gaps and help this amazing software achieve widespread success and redefine expectations for software quality across the industry.

We'll have a lot more details about what we do with the money once we actually start taking people's money.

Do they define "Handmade"? I couldn't find a definition.

https://handmade.network/manifesto

I read that but it doesn't define handmade. It gripes about large frameworks and rewriting in different languages but doesn't say what handmade is or how it addresses anything.

I agree it's vague.

That page seems like it's trying to define what Handmade is through a bunch of complaints and what it is not

Still no idea what they actually do, other than maybe this is just some random site about building a community to "make better software".

Software isn't bad because engineers don't care. It's bad because eventually people need to eat food, so they need to get paid, which means you have to build something people will pay for, this involves tradeoffs and deadlines, so we take shortcuts and software is imperfect.

> Software isn't bad because engineers don't care.

Caring is certainly a wide spectrum. I see the Handmade stuff being proudly on far end of it.

I don't disagree, the field has become lucrative enough it has attracted people who are interested in the money and not the craft

I'd use unrealistic to describe Handmade, proud is also accurate and works too

> the field has become lucrative enough it has attracted people who are interested in the money and not the craft

Yup, exactly.

> I'd use unrealistic to describe Handmade, proud is also accurate and works too

In certain settings definitely. But even in those corporate settings where it's unrealistic I'd rather work with one than not. If not applied dogmatically, that corner of the corp has a good chance of being an oasis. But a fleeting one perhaps.

My immediate assumption was that this was a reaction against LLM–assisted or –written software, but I couldn’t see any mention of this in the front page.

So maybe ‘handmade’ refers to artisanal, high quality, made with care, etc.

I assume you have absolutely no clue what it refers to.

Handmade Hero is a long running yt series by Casey Muratori. He builds a game engine from scratch, no cheats, no shortcuts, straight to the metal (from C-ish perspective). So you learn how to deal with computers to achieve things, fast and efficient, by understanding computers.

At some point Casey thought it was a failure and a waste of time. But to his surprise quite a fanbase evolved around it and it turned that it really helped people to go from zero to "hero". The handmade "movement" relates to this timeline and the aftermath of people thriving from it. My rough definition of "Handmade" dev mentality would be: Ignore the things that seem to make things "easy" (high level software) and learn the actual thing. So you learn what a framebuffer is instead of looking for a drawing api, applicable to different contexts.

That being said is that this foundation doesn't seem to be endorsed by Casey. Their mission goals seem quite shallow, if at all.

> no cheats, no shortcuts, straight to the metal (from C-ish perspective)

Not the person you replied to but even when I stumbled over this (the network, not the game) for the first time, I was left wondering where the line is drawn.

> You can learn how computers actually work, so you can unleash the full potential of modern systems. You can dig deep into the tech stack and learn what others take for granted.

Just.. no libraries? Are modern languages with batteries included ok? What makes a library for C worse than using Python? Is using Python too bloated already? Why is C ok and I don't have to bootstrap a compiler first? (E.g. building with Rust is a terrible experience from a performance perspective, the resulting software can be really nice and small)

I'm not even trying to be antagonistic, I simply don't understand. I'm just not willing to accept "you'll notice when you see it" as an example.

"The 501(c)(6) differs from the more familiar 501(c)(3) designation in that we are not a charity. The 501(c)(3) is explicitly designed for charitable organizations, and confers the additional benefit of donations being tax-deductible. Over time, though, the definition of a 501(c)(3) has become extremely distorted, especially in the software space, since companies were able to convince the IRS that making open-source software is a charitable/scientific activity. The result is that large companies were able to fund their own development by creating a “charity”, open-sourcing some of their core technology, and then building their extremely lucrative closed-source software on top. That way they get to deduct the core tech expenses from their taxes! What a deal!"

I get that, but I don't understand why it supports a 501(6) in this case[1].

Just because others have abused it doesn’t mean you should give up on it. Even if it's only about sending the right signal, that still matters.

Or is this about brutal honesty and they are saying bluntly: We're not a charity, so don't expect us to act like one in the first place.

If it is that, then why would anyone support them apart from their sponsoring organizations?

EDIT: Reading the whole thing carefully, I think they are going for an exclusive club. I genuinely wish them well, but to me it looks like a quite quixotic endeavour.

[1] There are many cases where a 501(6) makes sense. I'm strictly arguing the "Handmade Software Foundation" case here. Otherwise it gets complicated quickly.

The point is that as a 501(c)(6) we are directly allowed to act in the interests of the software industry, without having to invent a tortured explanation for why benefiting a very lucrative industry is Charity, Actually.

The hope is that people will sponsor us because we directly boost the creation and publishing of high-quality software, and give some measure of benefits to our paying members, which is typical 501(c)(6) stuff.

Really think they should support PanGUI (https://www.pangui.io/) first - they're making a native GUI framework that doesn't suck, and have very similar philosophies.

Is the software better, though?

On a tangent, still waiting for File Pilot to have even the bare minimum of CJK support... I do like that the software is fast, but without support for other languages than English it's useless to most of the people in the world.

Wow, no Unicode support is a pretty glaring flaw for a paid solution being sold as the superior alternative to Explorer.

I don't see why you'd ever want to use File Pilot over Directory Opus on Windows.

I'd like something better than Dolphin for Linux though, because Dolphin is absolute garbage. I'm going to have to write my own there.

Better can be argued. More performant though? Yes, massively so.

Turns out spending some time understanding what your CPU and GPU are actually doing when running your app, and how to make them do less work, leads to pretty speedy software.

Then it also turns out that this does not seem to impede most of the features of the software it is competing with, meaning that software is by definition wasteful.

It can't even be argued that the other software made better use of human resources since it's a large team vs one guy who is often not even getting paid, and the guy is the one with the fast software.

Needs an LLM to explain what it all means.

Reply on news.ycombinator.com