Ask HN: Language choice when starting a new project | Hacker News Reader

Ask HN: Language choice when starting a new project

2 pointsdennisy3 comments

I have wanted to start a new project for weeks, but I cannot make a choice as to what backend language to go for!

I want to keep things simple and although the project in the long term will or could become complex, ideally now it will be a single service. If I want to make it a single service for as long as possible I should opt for Python as there will be ML components, at the same time I know that in the long run a single service will get bulky here and I should split them, this pushes me towards selecting Node and splitting out the ML services right away.

Anyway, I am not asking for help on actually selecting the language, but rather how do people overcome this non starting due to making a technology decision such as this?

View on HN

There's practically very few projects which won't be done via python. Even in cases where python won't be a winner like kernel modules. I'll be trying python first and see if I can make it work. Python does the job for 100% of the projects I will take on.

>rather how do people overcome this non starting due to making a technology decision such as this?

Every project has oodles of decisions to be made. Even if you choose python, now which libraries will you use? Will you do type hinting? How advanced will your code be? Lots of decisions to make.

Better yet, even if python isn't the right tool or some library isnt the right tool. You always have the ability to change decisions. In fact, for any project to complete, you must make a myriad of decisions and then redecision bad decisions. So evaluating good vs bad decisions isnt your problem. It's making the first decision.

you have to ask yourself what is causing you to make no decisions?

Laziness? Most likely not.

Anxiety/fear is probably a good possibility. Kind of going back to the good vs bad. Lets be realistic and not fool ourselves. All our decisions are bad. The new decision is to be constantly improving and fixing those bad decisions.

My big one that often paralyzes me is fear of failure or consequences.

For example I wrote a book. 74,000 words after lots of editing came with a ton of decisions. I can make the decisions but my fear of being sued or submitting my book to 100 agents and being denied by each one has paralyzed me into not submitting to the first agent and therefore it's not published.

Thanks! I see you are suggesting go for Python, which is a sound decision! Do you use async when working with Python?

On the more general part of the question, I agree all decisions will be bad so I should just make one!

Thanks!

Python is going to be extremely easy to hire for, too.

Reply on news.ycombinator.com