Ask HN: How to find joy in writing/learning about tech in this AI world?

7 points by sriram_malhar 96 days ago | 4 comments

Looking to hear from fellow HN'ers who have found a way out of this downspell.

I've written code practically every day for 40 years, some of it for livelihood, but mostly because it gave me immense joy. I don't have much public codebases to show for it; I wrote code like an artist doodles in their spare time.

But lately, I am feeling lost. I find that this impulse to learn new things and write code has completely vanished with the new AI LLM regime. Things that I strove to learn and build slowly can be accomplished with ease. It is very possible that my aims were very modest and that my skills were ripe for getting automated.

I'd like to get out of this lull, but I simply can't find the motivation to dig into agentic AI and churn out stuff, like an old-school woodworker told to learn CAD and let the machine handle the nitty gritty.

Of course, I can continue to do what I used to do earlier, since I am neither interested in money nor fame. But one thing that I _think_ I had at the back of my mind in my earlier life was to internalize tiny 'katas' (patterns) and form insights that I imagined I could teach to someone. I find that I can no longer imagine that "someone", since everyone I meet is more interested in AI delivering the end product rather than going through the process and paying their dues.

Apologies for the rambling, and grateful in advance for suggestions.

PaulHoule 96 days ago |

Write about something else!

To me the combination of "I don't have much public codebases to show for it; I wrote code like an artist doodles in their spare time" and "things that I strove to learn and build slowly can be accomplished with ease" is telling.

From the viewpoint of somebody who makes a living at it and is only proud when I can put something in front of customers I don't think there is anything "easy" about it today. I mean, it is so irritating that HN is flooded with posts by people who are somewhere between delighted that they can make stuff that almost works with A.I. (e.g. no insight into the gap between "works" and "almost works") and who are crying that they don't know the secret sauce that influencers are using to launch 15 new perfectly polished products a day (e.g. no insight into anything.)

A.I. is the coding buddy I never had. It doesn't always give the right answers but neither does the programmer in the seat next to you or the crowd on Stack Overflow.

sriram_malhar 96 days ago | |

I'm curious about what you mean by "telling". What is the tell that you perceive? I would like to understand whether I misrepresented myself.

I agree with you, that none of it is easy. It is precisely why I used to doodle, to craft small projects to understand the core essence of what is going on: building an entire TCP/IP stack, writing a compiler, writing a database, an editor etc. That practice has allowed me to deploy into production a fair amount of efficient code.

But now, I find myself in the role of a project manager telling my highly capable coding buddy what to do, a role that I do not relish.

PaulHoule 96 days ago | | |

I think your motivation is wrong.

I don't know if it came from my work getting a PhD or from my work in startups [1] or earlier than that but I think any side project that "hasn't been done before" is not worth doing. For me any side project has to be something I can demo to an audience that, with a dash of showmanship, will knock their socks off.

For instance I knew a machine-learning based RSS reader was possible in 2004 and almost 20 years later it hurt that nobody else had made one, so I made one. I got interested in heart-rate variability and couldn't understand why I couldn't find any web-based HRV apps that used the BTLE API so I made

https://gen5.info/demo/biofeedback/

I wrote the prototype of that using Junie, the agent built into IntelliJ IDEA. I had a lot of anxiety because how do if I know if I coded it wrong or if the Windows Bluetooth stack is just being the Windows Bluetooth stack? The fact that I couldn't find public examples that could connect to a heart rate monitor made me wonder if there was a showstopper problem; what if I invest hours in study the documentation and "it just doesn't work?"

With Junie I had something up and running in 20 minutes that I understood and was ready to continue the development of. Now I can study the documentation and experiment with things and not have the fear I'm going to get stuck.

If you're making things that make no different like another TCP/IP stack and another compiler and another database and another editor no wonder you have been working on it for decades and have nothing public to show for it. You could have made an implementation of any of those things that was unique and different and shipped it which requires and entirely different kind of craftsmanship (if you use AI or not) and leaves you with a very different kind of feeling in the end.

[1] like oil and water in most people's mind, but like peanut butter and jelly in my mind.