Splitting from Handmade Network | Hacker News Reader

Splitting from Handmade Network

handmadecities.com

21 pointssurprisetalk9 comments

I've been following Casey since his podcast with Jeff Roberts. He had the most influence on me as a programmer. I loved the Handmade Hero series, and I reread his blogs occasionally.

That being said the whole network idea, and the fetishisation of his particular coding philosophy was always cringe. I've never seen anybody in the network approach Casey critically, or try to explore beyond what he said. I've only seen emulation. Same with Jonathan Blow fans. It took him and his team 8 years of dedicated work to make The Witness, and it's just not a very technically impressive game. Casey couldn't even finish Handmade Hero, and I haven't heard of him creating anything else recently.

There's clearly flaws in making software this way.

This whole drama seems like fanboys becoming community managers and arguing with other fanboys. No original thought in sight.

I feel the same about Casey and his work, and like another commenter said, his simple, practical and honest approach to his work is refreshing in an era of languages, environments and tools that feel bloated, slow and ineffective (IMHO).

> That being said the whole network idea, and the fetishisation of his particular coding philosophy was always cringe.

And I agree with this too. The point of Handmade Hero to me was always to show that the complex software systems that you interact with can be understood despite their size and complexity, and that it's still possible (though perhaps in a lot of cases impractical) to recreate these from the ground up. Also that there's tremendous value in understanding how something works from first principles (see also Casey's 'How to Open a Black Box' lecture, https://www.youtube.com/watch?v=SYomOZIfeoU).

Having said that, while I feel it's important to aspire to some level of simplicity and solving-the-actual-problem that Casey's coding style embodies, it also feels like an approach that works best only when a) the number of people working on a single program is relatively small and b) the expertise is deep and so the need for abstractions is low or non-existent.

With Casey's background in game tools it makes a lot of sense, but this approach doesn't seem like it scales up, and he's essentially the only contributor to Handmade Hero too.

> It took him and his team 8 years of dedicated work to make The Witness, and it's just not a very technically impressive game.

I'm a big fan of the design of The Witness and Blow's work in general and I don't feel like being "technically impressive" was ever the goal, as it wasn't with Braid either. Jon is as much of a solo designer on his projects, and it's not clear how much of that total development time went into their custom rasteriser and navigation (Killing the Walk Monster, https://www.youtube.com/watch?v=YE8MVNMzpbo) versus puzzle design.

At the end of the day I believe the end result was fantastic and I don't think too much about the development time, but it's true that Jon was in a unique position to be able to self-fund almost the entire project which a lot of other studios can't do.

> There's clearly flaws in making software this way.

As is always the case, being somewhere in the middle feels like the ideal place and extremes at either end have their own problems.

My point about The Witness was that not many people live in a reality where you can spend 8 years developing a piece of software that doesn't really solve any new technical problems for the end user. Obviously a lot of development time didn't involve programming, but this also wasn't a solo project; multiple highly-experienced engineers worked on The Witness.

The point is to have perspective when digesting strong opinions about software development.

So Handmade Hero turned into a whole network of projects?

I remember the announcement and waiting for the first episode. I had asked Casey whether he'd be using what is considered "modern" C++ (he doesn't), and whether he'd be developing in a test driven development style (he doesn't). His response was practical and honest. Like his programming. Highly recommend watching one of his videos.

Unfortunately, Handmade Hero the project itself seems to have died or at least to have been put on indefinite and unannounced hiatus: it’s been a bit over two years since the last video.

He finished the game engine. It came to a natural end. He started a new series on performance-aware programming[0].

[0]: https://www.computerenhance.com

No, he didn't.

The final video available is titled "Handmade Hero Day 667 - Simplified Tile Occupancy Checking", and ends with Casey explaining his plan for next week's session.

Visiting handmadehero.org gives a message saying:

> Handmade Hero is currently on hiatus. The original set of daily source code archives can still be purchased here for those that are trying to follow along with the original series. You can also follow Casey's current programming courses at computerenhance.com. > We hope to have more information in the near future about whether we will resume the series or replace it with a newly-designed one that better provides the educational benefits people ended up wanting.

There’s a bit more colour on the content of the argument from the other side: https://handmade.network/blog/p/8989-separating_from_handmad...

Neither of these statements really say what the disagreement was about. It seems like the Handmade Network people were shocked by the content of some of the talks at Handmade Seattle? Looking at the list of talks I don’t see what would be objectionable.

Frankly the whole “handmade” community seems prone to tedious arguments over what is/isn’t “handmade”.

Reply on news.ycombinator.com