5 ms·

Previously, I also used various so-called tiling window managers and configuring lots of shortcuts to improve my typing efficiency. Now, I mainly focus on resea

by west0n 3y ago

Previously, I also used various so-called tiling window managers and configuring lots of shortcuts to improve my typing efficiency. Now, I mainly focus on researching how to use GPT to generate code .. Perhaps, in the future, using GPT effectively to generate code and documentation could create many 10x or even 20x programmers.

dailykoder 3y ago
Maybe I just haven't used those "enough" yet, but I don't feel like chatGPT and others increase my productivity. Yes, they are nice to get some quick sketches/outlines, but that's about it. I rarely use them. Most building blocks are stored in my head by now and I don't even have to think about them.
- tazu 3y ago
  I've found it most useful for generating code I don't have to maintain (one-off scripts) and lots of regex stuff.
anbotero 3y ago
It may not be the same for everybody, but I once tracked 6 months with normal mouse usage, 6 months without but a lot of shortcuts... Do you know how much time I saved? Roughly 2 hours. I still use them and I do it without mouse whenever I can, mostly for everyday ergonomics, but I use the mouse when I really need to without thinking how I could do it without it. It’s important to take breaks too, longer-Pomodoro helps. As with GenAI, I’m already doing documentation with it. I pass it some YAML file or my own previous documentation, ask it to generate an organized version, I review it and edit it, and that’s it. For writing unit tests it’s been really helpful in environments where there is no QuickCheck or something that generates tests. Yes, it’s about using them efficiently, but we have to make sure it’s actually providing the positive impact we expected it to create.
- alchemist1e9 3y ago
  > but I once tracked 6 months with normal mouse usage, 6 months without but a lot of shortcuts... Do you know how much time I saved? Roughly 2 hours. I’d be interested in techniques/tools you used to measure the time saved. Your point is extremely important and a discussion around measuring the positive impact would be helpful.

by west0n 3y ago

dailykoder 3y ago

Maybe I just haven't used those "enough" yet, but I don't feel like chatGPT and others increase my productivity. Yes, they are nice to get some quick sketches/outlines, but that's about it. I rarely use them. Most building blocks are stored in my head by now and I don't even have to think about them.

tazu 3y ago
I've found it most useful for generating code I don't have to maintain (one-off scripts) and lots of regex stuff.

anbotero 3y ago

It may not be the same for everybody, but I once tracked 6 months with normal mouse usage, 6 months without but a lot of shortcuts... Do you know how much time I saved? Roughly 2 hours. I still use them and I do it without mouse whenever I can, mostly for everyday ergonomics, but I use the mouse when I really need to without thinking how I could do it without it. It’s important to take breaks too, longer-Pomodoro helps. As with GenAI, I’m already doing documentation with it. I pass it some YAML file or my own previous documentation, ask it to generate an organized version, I review it and edit it, and that’s it. For writing unit tests it’s been really helpful in environments where there is no QuickCheck or something that generates tests. Yes, it’s about using them efficiently, but we have to make sure it’s actually providing the positive impact we expected it to create.

alchemist1e9 3y ago
> but I once tracked 6 months with normal mouse usage, 6 months without but a lot of shortcuts... Do you know how much time I saved? Roughly 2 hours. I’d be interested in techniques/tools you used to measure the time saved. Your point is extremely important and a discussion around measuring the positive impact would be helpful.