[lovich]

>They're afraid they'll "break something".

Right? I feel like my non technical social connections have an even greater learned helplessness from interacting with any sort of open to customization technology because they've learned that everytime they touch something the tool stops working in a way they don't understand. Most people are not going to spend days to weeks, and definitely not months to years learning how to tinker with and expertly maintain their technology.

I feel like technologists in this forum are acting like blacksmiths who would scoff at any of us for having purchased a hammer, rather than smelting the ore, forging the head, and carving the handle so we could have one that fit our needs perfectly.

[saurik]

I dunno... the way it feels on the other side is that y'all think people are too dumb to not hurt themselves with hammers--which is true!!--and so, rather than trust that people who are afraid of hammers will simply avoid using a hammer they should be actively prevented from even owning a hammer, or even letting their friend or a hired carpenter use a hammer to help them, which is kind of overkill.

[bzzzt]

Relying on people not being dumb doesn't work out in practice.

There's lots of stuff you're not allowed to do because we figured out the average person can't do it without endangering others.

[bombolo]

> which is kind of overkill.

But a profitable one

[klibertp]

Well, it's a fact that all that technology is incredibly brittle. Systems lack resilience, error recovery, and accessible debuggability, and when something breaks, there's a high chance it'll have disastrous effects. It's objectively safer to stay within the "works for me" happy paths that authors are likely to be actually testing/using themselves. Even this sometimes fails, sometimes seemingly without reason, only to later (maybe) start working again. It's a nightmare, a constant source of stress and another thing that people feel they have no control of at all. It's not strange users flock to authoritarian-style environments, managed by someone who do have the capability to control that chaos to some extent - even if they sell users' PII data to Sunday and back.

There are complex reasons for this, but the end result is simply that IT is not ready for mass adoption. Software is still in its infancy - I suspect that the broader the possible implications of technology, the longer it will take it to be ready to be mass adopted. We gave up all hopes of ever proving program correctness in the 80s, then in the last decade we've given up all pretenses that we know what we're doing... and nobody saw a difference. By all rights, software should be confined to research labs and garages of nerds for quite a few more decades.

The problem is that this technology is too useful. It has too far-reaching applications in almost all spheres of human activity. When the software (and all layers below it) actually works, it brings small miracles to its users, enough that they're willing to pay a lot for a product obviously unfinished, rushed, that'll probably get killed after few years. They think that, yeah, it breaks all the time and I'm afraid to breathe in its direction, but it's ok, I'm strong, I can deal with it if I'm able to do X or Y.

Tl;DR? I dunno. Maybe developers should put more effort into professionalizing the field, but this kind of thing is impossible to rush. Or maybe the users should get a grip and accept that it's not developers who force them to use their products. The massive amounts of money involved, along with the life-changing potential of IT products, skew incentives so much that, currently, both developers and users pretend that it's all fine, even though it obviously isn't, and then both complain. Users are stupid, developers are lazy, but neither can live without the other any longer...