It's the incentives and everything is a trade off. Time to market, performance, features: none of these choices are made in a vacuum. Oh, and people like to go home and see their families once in a while.
As a developer of over 35 years, I feel like I hear the same arguments over and over again. "Programmers used to care about performance!" No they didn't, they just had no choice because computers sucked and you had to work on performance or your application would barely run. "Progammers used to care about the quality of their code!" Really. You apparently never worked on legacy systems with years of hacks and spaghetti code that took an afternoon to trace through just to figure out what it was doing.
People haven't changed. Kids aren't lazier these days. The incentives are always just to ship as fast as possible. Performance will be dealt with when and if it is so bad that the customer complains and not a moment sooner.
When I was much younger I fancied myself a "craftsman" of software. But any "craft" I was able to bestow on my software was in spite of the surrounding incentives not because of them. Software is closer to assembly line work than craftsmanship and LLMs are just driving that point home faster and harder than ever.
I still love software development after all these years but it's entirely because I love solving problems and computers still fascinate me the same as they did when I got my first TRS-80 Color Computer at age seven. Nobody that's not a programmer cares as long as the software does what they need it to and does fast enough that they don't start wonder why they have to use this piece of crap software in the first place.
Correct. More to the point, when it appeared many on the left attempted to wave this away by claiming that it was, as one put it, a "gentle satire" <https://en.wikipedia.org/wiki/Animal_Farm#Reception> when anyone who has read Animal Farm knows that it is in no way gentle in its satire.
What nl is attempting to do above is the latest iteration of what Animal Farm and 1984 both received from those who could not stand the spotlight of their scrutiny: Claim that the target is something else, and/or that Orwell's attacks are so pedestrianly obvious (since "everyone knows" that Stalinism is bad) as to be pointless.
I'm a strong supporter of the "I did it because I wanted to see if I could do it" ethos. So this isn't a criticism of the project itself, but I'm pretty sure a snap gun will beat this almost every time.
Disco Elysium is truly a wonderful game for adventure/rpg fans. I have a small fraction of the time I had as a younger man to play games so I have to be very selective with my choices and Disco Elysium has taken up a large portion of that time for the past few months.
It's the incentives and everything is a trade off. Time to market, performance, features: none of these choices are made in a vacuum. Oh, and people like to go home and see their families once in a while.
As a developer of over 35 years, I feel like I hear the same arguments over and over again. "Programmers used to care about performance!" No they didn't, they just had no choice because computers sucked and you had to work on performance or your application would barely run. "Progammers used to care about the quality of their code!" Really. You apparently never worked on legacy systems with years of hacks and spaghetti code that took an afternoon to trace through just to figure out what it was doing.
People haven't changed. Kids aren't lazier these days. The incentives are always just to ship as fast as possible. Performance will be dealt with when and if it is so bad that the customer complains and not a moment sooner.
When I was much younger I fancied myself a "craftsman" of software. But any "craft" I was able to bestow on my software was in spite of the surrounding incentives not because of them. Software is closer to assembly line work than craftsmanship and LLMs are just driving that point home faster and harder than ever.
I still love software development after all these years but it's entirely because I love solving problems and computers still fascinate me the same as they did when I got my first TRS-80 Color Computer at age seven. Nobody that's not a programmer cares as long as the software does what they need it to and does fast enough that they don't start wonder why they have to use this piece of crap software in the first place.