121–130 of 800 posts
The kid who showed his work in detail in math class is doing better in life 9/10 times than the kids that only knew how to use a calculator. Now consider how well the people who think you just need to know how to yell at the calculator are going to do? When Maps apps came around, people totally lost the brain muscle for being able to navigate. Using LLMs is no different, people over reliant on these tools are simply…
>what's the point of "getting in early"? You're trying to make the point using BitCoin, but in the early 2000s I had just over 14,000 of them, so I can quite clearly see a point in getting in early.
This underscores a main counter point: dipping your toes and casting a wide net often has a low cost, since back then you could mine (and even purchase) Bitcoin relatively inexpensively. If it hadn't worked out, then it wouldn't have hurt much at all.
Guy who's ok with being left behind (crypto, AI) did a MSc on the metaverse. Sounds like he tried to go with the hype once, got burned.
You're welcome to read my MSc and see if I was "going with the hype" https://arxiv.org/abs/2304.10542 The answer was quite the opposite. I wanted to see if the technology lived up to the hype. The answer, unsurprisingly was no. If only Zuck had listened to me :-)
I was highly skeptical of this happening not that long ago, but I have to say that it seems increasingly likely. LLMs are still quite mediocre at esoteric stuff, but most software development work isn't esoteric. There's the viable argument that software development largely isn't about writing code, but the ability to write code is what justifies software developer salaries, because there's a large barrier to entry there that most just can't overcome. The 80/20 law seems to apply to everything, certainly here - 80% of your salary is justified from 20% of what you spend your time doing.
It's quite impossible to imagine what this will do to the overall market, because while this sounds highly negative for software developers, we're also talking about a future where going independent will be way easier than ever before, because one of the main barriers for fully independent development is gaps in your skillset. Those gaps may not be especially difficult, but they're just outside your domain. And LLMs do a terrific job of passably filling them in.
It'd be interesting if the entire domain of internet and software tech plummets in overall value due to excessive and trivialized competition. That'd probably be a highly disruptive but ultimately positive direction for society.
In general, we as a society have not adjusted to technology. We've gone through to much change to have any stable base lines. So we're going to float in insanity for a while until things finally settle down. Probably 2 wars, a famine, and several periods of resource scarcity away still, but we'll get there one day...
Not going to lie, I’d rather be poor. Not destitute - I’ve been poor but not destitute and I’d rather not go desperate - but poor? As in (because “poor” is very imprecise and can imply anything between utter poverty to “not owning three homes”) like having a low paying job but still enough to pay rent?
I’d rather be that than do AI assisted software development. Genuinely the only thing stopping me now is that there’s actually way more skill and qualifications in most low-paying jobs than a typical software developer imagines, and acquiring those takes time and money itself. But by now I know multiple people who made the jump even before the latest madness, and they’re all happier. Some still code, but don’t even publish. Some are like “I haven’t used a proper computer in _months_ this is great.” All work hard jobs at odd hours. None regret.
>I didn't use Git when it first came out. This really hinges on what you mean by "didn't use git". If you were using bzr or svn, that's one thing. If you were saving multiple copies of files ("foo.old.didntwork" and the like), then I'd submit that you're making the point for the AI supporters. I consulted with a couple developers at the local university as recently as a couple years ago who were still doing the copy…
I don't understand how this, at all, makes "the point" for anyone.
As someone further down the road in my career, I would argue that waiting is your prerogative but you do so at your own peril. I made these kind of mistakes early in my career, stuck it out with PHP for far too long ignoring all the changes with frontend design trends, react, etc. I was using jQuery far too late in my career and it really hurt me during interviews. What I was doing was seen as dated and it made ageis…
It is easy NOW to look back and see the optimal path for a web developer, but was that obvious from the start? How many killer technologies lie unused today?