7 comments

[ 2.0 ms ] story [ 25.5 ms ] thread
I've been thinking with the advancements of agents being more capable of being left alone and just treated like a co-worker, is anyone actually finding themselves doing the assingment tasks from their phone, away from their laptop.
This logic seems reversed though. If someone is primarily vibe coding, why wouldn't a phone be just fine?

Either way, there are still completely legitimate reasons why one would want to code on their phone, with or without AI.

Many decades ago I had this vision that the quality of ones ideas scales with the distance from civilization and specially from computers. If true pocket computers would rule out good ideas.

I use to plan out entire projects in my head. Today I'm indeed guilty of chatting with llms rather than gaze at the horizon.

Me too, I find myself gaining or loosing motivation for projects based on social sentiment online. When you're in the real world, it's so different almost as if it's completely disconnected from what we see online. Especially in tech waves, we tend to focus on the fringe instead of making basic progress because we're afraid we're already too late. But when you step outside and talk to real people, you realise it's really not too late. So I agree, ideas get better the further away you are from a computer :)
There is this abundance of real things in the real world (haha) you cant help but have thoughts attacking real problems. I will come up with software idea in areas where the state of the art is pathetic. Browsing online you won't run into things that didn't get enough love.

It does seem to be possible to flip out your phone in the field and write a solution on the spot but the main drawback I think is that you cant properly pronounce something like for(i=1;i<42;i++){ and typing it on an onscreen keyboard might even be worse. I remember writing code in a textarea one time then pulling the screen down refreshing the page. I've been wondering eversince if blanking input areas on reload is ever what the user wants. I put so much work into those 10 lines.

The medium seems more fit for a "language" of dragging around elements in flowcharts. Then again, how would you begin to pronounce that?

Maybe a simplistic DSL without symbols to [say] only make crud apps?

Perhaps start out with something like:

    HAVE A FORM WITH USERNAME AND PASSWORD. 
    IF USERNAME AND PASSWORD MATCH A DATABASE ENTRY
    REMEMBER THE USERNAME AND SET A SESSION COOKIE.
And work towards:

    HAVE A REGISTRATION FORM. 
With some default database structure. If there is no DB make it, if there is no table make it. Have 3rd party account providers baked into the language.

It seems LLM's would be really good at strictly defined English.

Me.

  Irssi Connectbot (archived version/modified).
 Hackerkeyboard (modified)
  Ssh + GNU Screen + emacs
  Pixel Fold
Interesting! I've been wanting to build something in this space. Any learnings you can share about your experience so far?