There is a reason we use React today - if you want dynamic and complex site, HTML is PITA to work with. Either you reload everything on any change (not good), or end up with a maze of partial reloads in event handlers (not good either). It was done, it was bad, React (and similar frameworks) was the solution. If you don't need complexity of React, fine, use HTML. But if you do, who can guarantee that 5 min later someone will not start requesting site not to reload here and there, but update this bit here. So I don't blame people for just using React. And in this case the issue wasn't that react was used, but that it was used poorly.
I like to pause movies when some code is shown and see what it is. Apparently you can break into pentagon by knowing basic sql and high-level employees have alternate life writing tcp implementations and graphics libraries.
I am not sure RAM/fuel prices have much impact here. I have a phone that's about to get 3 generations behind latest one and comparing them side by side, I have 0 reasons to upgrade. Some specs are better, but not enough to notice in daily use.
> A good piece of software would make it hard to do this by accident, and probably should default to having the same behaviour with or without trailing slash.
Django redirects one version to another by default, which achieves that.
Those employees cannot (and are not expected to) code, so dynamic elements are not an option. So what you'll get are mostly either reports or messages. Reports could be created by software (which is how an engineer would do that), messages (like responding to client or manager requests) require more human-facing skills.
I assume other platforms will prioritize paying customers over agents (of not paying ones) because they do not have resources to attempt anything else.
Github ignored platform issues (not just performance) because it was busy migrating to Azure, and it wants to bear the cost of ai using its infrastructure now to benefit from that somehow later. Other providers do not have this problem and live mostly from paying customers whose priority is to have stable platform. So yes, it is easy to find someone else whose priorities align with mine and doesn't have GH issues.
> AI gets tossed around as excuse for things, but it really is finding some obscure vulnerabilities humans wouldn't have.
Which is fine, as long those using it do not do so at expense of others. As a paying Github customer, I do not wish to pay for a service that doesn't work because someone else is throwing agents at it. This is largely GH issue, not general problem.
> threat actors are going to start heading that way as well
bad actors are not the problem, Github priorities and attitude are, so switching solves that. Will other providers have outages? Sure, sometimes. But you'll be able to find one that manages that better.
> Ever seen most high profile apps in the last 10-15 years?
No, but the ones I am paying for are few and with rather solid development record. Looking at top android apps, we have some that have very high bar of engineering (firefox, google translate), some just useful dataset (cooking recipes), but again, even the simple ones outcompeted tons of alternatives.
> most enterprise software
I am not saying enterprise software is super complex, but it is very competitive domain.
> software development becomes a commodity and the job becomes something like a fast food job where practically any adult who wants it can do it
That will never happen. Sure, anyone can program something, but to make it professionally there is bar of quality and competition will push most people out. Similarly anyone can write, draw or sing but only a few do it well enough to be paid for it.
And I am old enough to see many tools that allow "anyone to program". They pop up whenever certain standards (like web) become popular, then programming goes in different direction and they vanish in irrelevance. Soon there will be a large set of skill on top of "using AI" and "chat, make me an app" will go out of the window as viable way to make something others want to use and pay for.
> What if we built things that are meant to last? Would the world be better for it?
You'd have a better bridge, at the expense of other things, like hospitals or roads. If people choose good-enough bridges, that shows there is something else they value more.