Across the last 18 years I've been working on many different projects with different material systems; my journey looked like wood framing → steel → reinforced concrete → precast → mass timber → structural steel → composite materials → modular construction + all possible foundation types (from shallow footings to deep pile systems with geotechnical analysis).

I have a lot of projects I want to design/prototype quickly, but every time starting a new structure I feel stuck and paralyzed with many choices, end up reading engineering journals and CE forums for 3 days on all possible material and structural system options and find myself exhausted even before breaking ground.

I know I should just pick one system and stick to it, but it's very hard. I spent most of my career working exclusively on residential concrete construction and I know that system inside out. I was efficient and could estimate quantities quickly. With the rest I feel like everything is at the same level of unknowns and I have zero engineering judgment.

Do you happen to experience the same and how do you fight it?

> I know I should just pick one and stick to it, but its very hard.

You already know the answer. It is hard, but only because sticking with something means making a choice, investing time, and taking accountability to learn and do something with your new skillset.

> I have a lot of ideas I want to build/prototype quickly, but every time starting a project I feel stuck and paralyzed with many choices, end up reading internet and HN for 3 days on all possible tech choices and find myself tired even before starting project.

A few things. You mention you want to build/prototype quickly, but the tech stacks you mention aren't really rapid development environments. You might be getting frustrated because you're playing in objective-c when you should be on a whiteboard working out an idea.

The situation you describe isn't easy to get out of until you address whatever fear is causing you to over-worry about every step.

Overthinking is an escape. As you're learning, it leads nowhere. You will have to break this cycle by actually creating something and making a choice and sticking with it.

The comfort of knowing what you're doing the way you feel with obj-c is likely years, not days, away in any new language. If you actually want to feel like a deep expert in something, go back to what you know and freshen up. Anything else needs more than "3 days reading HN" level of devotion and energy. There's no shame in not having the energy, interest, or whatever, to learn something brand new.

I generally think people who are in this mindset of feeling paralyzed when they're fully equipped to succeed need a real wake up call.

Be honest with yourself about what your day looks like: how many hours are you spending in your head or looking at a screen vs. in the world interacting with your community?

I am just guessing, but based on reading this, you probably need to immediately change up your routine to get out of this rut of thinking or you will wake up old with every opportunity behind you and no chance to take it.

> The situation you describe isn't easy to get out of until you address whatever fear is causing you to over-worry about every step.

Thanks for the reply, I think this is exactly what I needed to hear; will try to reflect on this.

From the very top of my head: I know exactly what I need to do but I feel like with every tech choice I need to either learn "how" or refresh my previous knowledge.

Wearing multiple hats in rapidly popping up and fading out tech startups didn't really help to go deep with one thing (as I managed to with objc/ios)

Don't let my blunt way of writing make you feel like it should have been easy to see or will be easy to overcome.

It sounds like you have been through the transition from a career rewarding complete knowledge and understanding to today... where a "complete understanding" and depth of knowledge is not possible in the same way.

Rolling releases, CI, and cloud dev cycles with dependencies have completely changed what is expected of you - it is OK not to know every edge case in most situations today. The answer is a google or prompt away if a problem appears.

This wasn't really true 18 years ago... even though it doesn't feel like tech has changed a ton, it has.

If you decide to keep doing a rotation of hats and languages, you have to become very comfortable with identifying the value you add as a person involved in the process, even when you don't have every answer in your head.

In other words, being able to pop in and out of various startups tells me you have the ability to learn quickly. Why not just truly sit back and ask yourself what you want? Even if it doesn't fit the HN front page expectations, doing what you want with a lucrative skillset is as close to a good work/life balance as I have ever found. Holler if you wanna talk more, but otherwise I hope others will chime in.

There is not really one solution - you need to answer some specific needs to help determine the best fit.

What do you want as output?

Do you want it web based?

Do you want it to run on which platforms? (important if it's iOS and Android)

Are there any language specific features you want to use? (i.e. mapping may want to use PostreSQL)

What devices do you want it to run on? and are those devices disconnected from the internet?

Game oriented (high-end gaming?)?

User-Friendly to install.update?

Integrates with say Microsoft AD?

Open source with a large potential community?

These answers will help whittle down the choices. One language might do most of that stuff but not as good as picking the right tools for the job. (which also includes the ones you really like to work with - which is a major point)

Then pick one and start - If it doesn't work out, switch. One of my favorite mantras: "stop shopping - start doing".

With any ecosystem that allows lots of customizability means that there are millions of alternatives to choose from.

Web development is exactly like this.

Apple gives you a limited choice of supported languages to build iOS and macOS apps - i.e. Swift (default) or Objective-C which means less focus on seeking alternatives, context switching on the stack and is why you are productive with the technology and know it inside out.

If you haven’t got a really good reason just https://boringtechnology.club/ and https://mccue.dev/pages/8-16-24-just-use-postgres

Thanks! Postgres is the easiest choice ever haha

My analysis paralysis was solved with a Github Copilot account for $20/month getting me access* to ChatGPT5 (preview) agent inside Microsoft Visual Studio Code.

At least I think that's how I'm paying for it

I've just given it a folder and a description of what I want, and iterated until I got what I needed.

It's great for trying out ideas.

It is kind of tautological but, the only thing that will make you stick with one, is letting go. Let go all the new possibilities, and use the same stack you used for the preceding project.

> Let go all the new possibilities, and use the same stack you used for the preceding project.

I think this is very good perspective, thank you for the advice

A brain recognises futile effort with minuscule success rate and hints with procrastination.

Interesting point