[ravenstine]

For better or worse, in a lot of ways, I think the last decade of developers have done a lot to kick the ladder out from underneath them.

One of the reasons I'm a software engineer today is because I could easily examine webpages to see how they work, tinker with the memory of programs, open up hardware to see what's inside (relatively low risk of destroying the device), and so forth. And yeah, back in the day of Halo PC, modding the levels taught me a lot about what goes into the game. Knowing what a "BSP" is can be a pretty useless piece of trivia, but it made me feel smart and capable of understanding more.

It's still possible to get into tech and learn things today, but I have a hard time seeing how this can be accomplished by genuine tinkering. Software is way harder to crack/debug today; certainly not impossible, but the barrier of entry is much higher. Plus there's a ton of moving parts that go into getting software to work securely on not just mobile but modern desktops, adding another layer of hassle. You can still examine what a webpage is doing, but even that has changed significantly; so many websites today are div soup (yes, worse than in the early 2000s) to support JavaScript monstrocities that are also minified and obfuscated. When it comes to games, you can almost forget it in some cases because they're so heavily dependent on content streaming from servers. As far as hardware goes, you've either got to face various security hoops that can brick devices, use a heat gun to unglue the bezels on certain things, and face a lot more risk in permanent damage.

All of these changes were made for a reason, but we've also taken the fun out of everything. Even game modding really wasn't what it used to be in spite of some of the tools available today technically being better than those in the past.

[somenameforme]

I agree with you in general, but I think it has to be said that we also live in the era of things like the source for Unreal Engine 5 being completely and wholly available, along with a zillion tutorials and extremely well documented source. This is unlike anything we had, and better in basically every single way imaginable. In the hardware world there's Pi's, dirt cheap PCBs that you can have delivered in < 24 hours, and so on endlessly. It's like instead of breaking into a trailer, we've all been granted carte blanche access to a mansion.

But maybe there was something about how counter-culture and esoteric stuff was itself attractive precisely because of that. There was also a lot more reward for a lot less work, largely because so few people were doing it. Now if somebody wants to go learn Unreal then it's just a pretty mundane and common thing, and you'll also be largely incompetent unless you're willing to dedicate years to it. By contrast when I was a kid changing the text in a shareware installer was enough to wow my friends with my leet skills, and that's something that took about 5 minutes to do, and not that much longer to learn. Oh and then creating secret directories by naming them alt+255, and so on. Dumb stuff, but it soon enough led me to much more than parlor tricks.

[ravenstine]

> But maybe there was something about how counter-culture and esoteric stuff was itself attractive precisely because of that.

There's definitely that element, but I also think something missed by compartmentalizing hardware "tinkering" to devices designed specifically for the task. Nothing about a Raspberry Pi, for instance, is mysterious. If a person is going to buy one, they already have a significant level of interest and base knowledge. A kid's not gonna have one lying around and get curious about it unless their parent is a geek who owns those things, and even then said kid may have no good reason to even bother with one. Practically nobody today is opening up their laptop or their phone to mod it or even just see what's inside. I'm not saying the modern situation is bad, but a significant amount of it is artificial in a way that wasn't when the devices one would play with were the devices actually being used, and it's not clear to me whether what we have now is actually better in regard to inspiring future generations. Engaging with one's everyday hardware is an exercise in the power process that fewer and fewer generations are experiencing.

[thomastjeffery]

The big difference is that people used to mod existing games. That allowed modders to leverage the game's engine, gameplay design, art, and even its existing playerbase.

Having a full-featured open-source game engine is great, but starting with that means starting at square one.

By obsessively enforcing copyright and "anti-cheat", we effectively bury the game-making process 6 feet underground. Every game is declared dead at the very moment of its release. Every decision its creators made up to then is set in stone. The game studio itself must exist in isolation, ignorant of the very world it is creating its games for.

It's no wonder that AAA studios are so out of touch with the people who play their games. Gamers are explicitly excluded from the creative process.

[BlueTemplar]

That's their loss. Meanwhile Minecraft is the most sold game ever, and Roblox is probably not far behind. (That one comes with its own issues, but hopefully children are sick of those by the time they become teens and are ready to move on to something more open ?)

[xkcd-sucks]

One consideration is maybe stuff that's "hackable", i.e. immediately accessible, can be incrementally and reversibly modified, is more accessible to "play" while stuff that's extremely capable but has a high activation cost (setting up the environment, learning all the stuff to make something basic etc.) is more accessible only to "focused / goal oriented study" and this has all kinds of implications on who does it, who succeeds, under which circumstances etc.

e.g. my friend tells me you can open game.exe in notepad + change this value to walk on lava, then I fiddle with it and tl;dr make a map of my school, then get frustrated with limitations and start learning a game engine with some practical background on what these concepts are etc.

vs. I decide want to make a game because that's cool, I buy a book on game programming, it depends on these libraries, I install them, I install a compiler, the libraries don't work with the compiler/each other, ......... and give up because the grit in my life is reserved for stuff other than video games.

... like really, what is the overlap of people that are really mind blowingly creative as artists, and the people that are super type A driven to go through all this frustration up front? less friction, more better art

[coldpie]

I agree with your concern, but I'm quite happy to see what's going on in the retro emulation & decompilation/reverse-engineering scenes[1]. A lot of that is being done and is driven by "the kids". It's an appealing, easy, and low-risk entry point for newer developers who want to dive into low level stuff, and it even has a bit of a "fuck the man" bent to it, which is fantastic. You're right that the environment is different from what we grew up with, it was always going to be. But I think the kids will find their own way.

[1] If you haven't been tuned in, check this out, it's the goddamn Super Mario 64 source code in C, reverse-engineered from the game ROM: https://github.com/n64decomp/sm64/blob/master/src/game/hud.c Similar projects exist for a ton of other classic games. It's jaw dropping what they're doing out there.

[LocalH]

There are also efforts for games that aren't quite so old too. There is a Rock Band 3 Wii decompilation project underway, for example.

[ravenstine]

Yeah, I've heard of that decompilation, and have been more closely following the complete decompilation and PC port of Perfect Dark, which is pretty amazing.

[CYR1X]

To be fair you have to realize how much more dependent the global economy is on software nowadays. Halo was a Bungie product, and when Bungie left as Microsoft took it over yeah, the product had loftier business goals that needed to be protected. Smaller devs can't really reach critical market saturation anymore like they used to.

You can criticize where things has gone with MTX but I don't think that was a choice by the game developers. I also think it's a generational thing as you and I are now too old to spend all day fucking around with some game we played a ton. I'm sure the kids of today are tinkering with minecraft/fortnite/whatever replaced those games.

Also to me it's a bit of wanting to try and put the tooth paste back in the tube. When we were teenagers modding Halo we maybe didn't fully understand the impacts it had on the game's community and the overall experience. As an adult I just see how games like Call of Duty appear to be constantly losing the fight against hackers, and using over engineered matchmaking algos with SSBM to try and maximize the typical user's gameplay experience. But I can't regain my naivety and be one of the many younger adults who probably don't even really notice these issues, the way I didn't when I played Halo 3 on Xbox Live which had boosting, standby, and stealth servers.

[radicalbyte]

I started coding with an Action Replay cartridge. Press a button and you get dumped into a new context where can fully inspect the machine and running program, and can modify any of it.

That's easier it was later on Windows (which for 15 years was how 99% of people used computers) and anything after.

[mrguyorama]

CheatEngine exists on modern systems and can still do a lot of this, though modern game engines are less friendly to the simple "scan for the health variable and set it to 1000" workflow of yesteryear. However, newer versions of cheat engine include C#/monogame decompilers that let you screw around with some unity games

[pipes]

Action reply on a snes? I use to mess around with this too as a ten year old

[dtech]

I think you're making the classic mistake of thinking that because your on-ramp isn't available anymore there are no on-ramps anymore.

I read an extremely similar comment two decades or so ago about a dev saying that no-one would learn computers anymore because everyone now used GUIs instead of CLIs...

Yes modding might be harder nowadays, but you have things like Scratch and Hedy, or the freely available Unreal/Unity dev tools with asset stores.

[SmellTheGlove]

I agree with you, but would suggest that the thing to focus on is the next layer of abstraction. The thing that barely works. Our generation screwed around with the PSX and Xbox, the Wild West era of the internet, etc. Now the most obvious thing I can think of that is the same level of not done is AI. Yeah there’s a higher barrier to entry, but look at that guy the other day that posted about improving performance by dynamically pruning a model. Or half the crazy shit on huggingface. Generative video and image AI seems to have a lower entry barrier too.

Again I’m not saying you’re wrong. It just seems like the thing that’s always the most prime for screwing around with is the thing at the bleeding edge. There’s still some fun to have.

[poisonborz]

I would look elsewhere. SBCs like Pi Zero costing peanuts, myriads of few dollar sensors enabling all kinds of easy to hop projects, virtualisation, cheap (if non gaming) x86 hardware, much better OSS landscape, much more learning resources, GPT to help with questions... it can be easily seen as paradise as well.

[mrandish]

> It's still possible to get into tech and learn things today, but I have a hard time seeing how this can be accomplished by genuine tinkering.

I agree and have the same feeling we've lost something as so much computer-centric tech has become relatively inaccessible in the name of 'security' (although much seems like efforts 'secure' business models or IP). As you observed, the tech has also evolved in ways which make it relatively undiscoverable through casual tinkering. Although perhaps we suffer from a generational perspective bias, I think there really was a 'golden era' of computer tech hobbyist accessibility and discoverability.

When Did the 'Golden Era' Begin?

I'd peg it as starting around the late Usenet era. Interestingly, there was definitely a time period in consumer computer tech when it was too early. I know because I started "too early" and missed being a teenager in the golden era because I got my first computer as a teenager in 1981. It was a Radio Shack Color Computer with 4K of memory, a 0.9 Mhz 8-bit 6809 processor and storage via an external cassette tape player. Fortunately, the ROM BASIC on that model was perhaps the most evolved 8-bit ROM BASIC Microsoft ever made (vs the early Commodore & Atari flavors) and Radio Shack did a fantastic job on the large, well-illustrated color manuals.

Unfortunately, 1981 was too early because no one in my family's extended social network had ever touched a computer. So, beyond the BASIC manual in the box, I was on my own with my new computer. While there were a few big magazines like BYTE on news stands, very little in them applied to my computer. I eventually discovered a couple of zine-style publications at a distant big news stand. Although they were essentially overgrown stapled newsletters printed in B&W, they became my lifeline because they had articles written by hobbyists more advanced than I, as well as mail-order ads for cassette tape-based software. This was the key that unlocked the mysterious realm of assembly language for me when I ordered a $12 homebrew monitor program written by some random guy who took out a classified ad. The local library didn't have any books relevant to my new microcomputer, local colleges only offered computer courses under the math dept and those were focused on mainframes and COBOL (I think back then 'real' CompSci was limited to Ivy League and top tier tech unis). Even large bookstores had nothing useful to me I could order other than Osborne's 6809 CPU book which was really an architecture and instruction set reference manual mostly incomprehensible to an isolated teenage hobbyist starting out.

A few years later 300 baud modems became cheap enough for hobbyists to acquire but it took another year or so for BBSes to emerge which were targeted at my computer (most BBSes prior to that were run on CP/M hardware and focused on one platform (not mine)). So dialing BBSes focused on my platform involved long-distance charges which meant short calls. Another year later FidoNet connected larger BBSes and national-level info began to circulate and my local hobby scene stayed pretty much like this for a few years. New info centered around zines, local computer club meetings, mailed tapes & diskettes and short BBS calls. Info was available but it was scarce and you had to work at getting it.

That's why I think the true golden era truly took off in the late Usenet period. That's when anyone could subscribe to a ~$10/mo service providing 1200 baud access to Usenet feeds in their local area code. Before that, unless you were at a university studying CompSci or worked at a uni or large tech company, Usenet was a magical land you only heard about on BBSes or at user group meetings. When random home hobbyists got direct access to the firehose of high-quality, global Nerdverse content that was the Usenet CompSci feeds it felt like the Enlightenment dawning. From there the transition to the early web was pretty natural since a lot of early tech-centric websites were much like a BBS ring. We didn't need search because they mostly linked to each other and people were running them as a hobby so few had ads other than maybe a sponsorship from an ISP or modem company (usually just paid in free service or hardware). Fortunately, the tech hobbyist web wasn't impacted much by the 2001 dot com crash since it was never about revenue. Up until the slow decline gradually started in the mid-2000s, it was pretty great - flashing BLINK tags and all. Honestly, we didn't even realize how good we had it, or imagine that it might someday end.

When Did the 'Golden Era' End?

Having lived through the pre-Golden Era, the early days and through the end, I think the seeds of the Golden Era's slow decline were planted when the modern web business began to emerge from the ashes of the dot com crash. Although things were still pretty hobbyist-discoverable in desktop OSes and the web through 2010-ish, troubling signs were on the horizon. For those paying attention, the rapid dominance of iOS in the late 2000s was ominous. Apple's business model required a walled garden app store and their concept of users was not as active explorers but as purely passive eyeballs for media and app-snacking. Even though a few app developers did well in the early app store, the fundamental model relegated them to the role of sharecroppers working Apple's farm with Apple's tools and selling only to Apple's store (with no access to their app's end-users).

In all, entry-level, home-based tech hobbyists got almost 20 really amazing years in the 'Golden Era' from roughly the late 80s to the late 2000s. It would be wonderful if in the distant future that period is known as "The First Golden Era" but right now it's hard to be optimistic. While there is still an enormous amount of hobbyist info available online and more emerging, it's in a context of equally increasing locked down areas and ever decreasing discoverability (though open source and Github-like sites are notable exceptions).

Maybe this is why retro computing and retro gaming are booming now with new people who never experienced it the first time. It's a place where that unique Golden Era ethos, vibe and community still exists. Last year I went to a local user group meeting for Amiga computers, which is what I had mid-80s to early 90s. I met a bunch of enthusiastic Amiga users who hadn't been born when I bought my first Amiga. It was strange to feel both "old" and "OG Cool" at the same moment but also heartening to feel that same open community vibe still beating. :-)

[bee_rider]

They’ve got, with wildly varying levels of skill and investment: unity, unreal, godot, or Roblox.

Not quite as fun as modding though.

[anonymousab]

I think adding your own touch to something you're more intimately familiar with helps ease people into it. Touching an asset file to modify a texture and instantly see the results in your favorite game is a lot more approachable (and appreciate-able) for someone starting out, I think.

[bee_rider]

100% agree

[anal_reactor]

I think it's also the fact that back in the day shit usually didn't work, so it was expected to need to tinker with it.