3 ms·

It's not a reach to suggest that if you've used software written in the past 2-3 years, you're enjoying cool stuff. Moreover, all of the tools that the people

It's not a reach to suggest that if you've used software written in the past 2-3 years, you're enjoying cool stuff. Moreover, all of the tools that the people who build software use are also cool stuff. It's also not just code and software that is benefitting from these new tools. Use of LLMs in engineering tasks is blowing up right now.

jayd16 5mo ago
I'm not sure that extrapolating the last 2 to 3 years as a sign of things to come is as enticing an argument as you seem to think it is . If you exclude AI for ai's sake, the feature lists of the last 2 years have been incredibly anemic. If you include AI companies bootstrapping themselves with AI, the cash flow has been a nice change but I can't say it's felt fully baked, or flooded with stable software and well-crafted workflows. I'm really not trying to be a hater but when people tell me that we're already in the AI Nirvana it gives me pause.
- peteforde 5mo ago
  I don't know how to argue with this without making the same kind of sweeping generalizations that you are. I don't know that you or I are qualified to set the goalposts or come up with any kind of objective measure for the progress of all recent software and all recent hardware that is convenient to analyze or debate in this text box. While there's a loud minority who love to debate the topic, LLM use has become status quo on most products and projects pretty much across the board, and most people are happy to acknowledge 1:1 that their personal productivity is some multiple of [all time before they started using LLMs]. At the same time, you can surely appreciate why many are quiet about their personal usage because there's no upside to discussing it but there's lots of people just hanging out waiting to tell you that you're imagining the whole productivity thing when you do. At some point, the path of least resistance is to let the loud minority be loud while you get an extraordinary amount of work done.
  - jayd16 5mo ago
    So we've abandoned the idea that coolness is even perceivable. Even if the argument was about pure productivity, or any hard measurement, the numbers don't seem to add up. Can we point to any hard metric that has improved in the industry in the last 2-3 years? It explains why work hours are short, everything is cheaper and non-AI companies are experiencing cost reductions. Services are more reliable. Except, where is all that? You could argue that it's yet to come but to argue that it's already here... how do you justify that?

jayd16 5mo ago

I'm not sure that extrapolating the last 2 to 3 years as a sign of things to come is as enticing an argument as you seem to think it is . If you exclude AI for ai's sake, the feature lists of the last 2 years have been incredibly anemic. If you include AI companies bootstrapping themselves with AI, the cash flow has been a nice change but I can't say it's felt fully baked, or flooded with stable software and well-crafted workflows. I'm really not trying to be a hater but when people tell me that we're already in the AI Nirvana it gives me pause.

peteforde 5mo ago
I don't know how to argue with this without making the same kind of sweeping generalizations that you are. I don't know that you or I are qualified to set the goalposts or come up with any kind of objective measure for the progress of all recent software and all recent hardware that is convenient to analyze or debate in this text box. While there's a loud minority who love to debate the topic, LLM use has become status quo on most products and projects pretty much across the board, and most people are happy to acknowledge 1:1 that their personal productivity is some multiple of [all time before they started using LLMs]. At the same time, you can surely appreciate why many are quiet about their personal usage because there's no upside to discussing it but there's lots of people just hanging out waiting to tell you that you're imagining the whole productivity thing when you do. At some point, the path of least resistance is to let the loud minority be loud while you get an extraordinary amount of work done.
- jayd16 5mo ago
  So we've abandoned the idea that coolness is even perceivable. Even if the argument was about pure productivity, or any hard measurement, the numbers don't seem to add up. Can we point to any hard metric that has improved in the industry in the last 2-3 years? It explains why work hours are short, everything is cheaper and non-AI companies are experiencing cost reductions. Services are more reliable. Except, where is all that? You could argue that it's yet to come but to argue that it's already here... how do you justify that?