I use llms exactly and exclusively for the first two cases - just write comments like:
// map this object array to extract data, and use reduce to update the hasher
And let llms do the rest. I rarely find my self back to the browser - 80% of the time they spit out a completely acceptable solution, and for the rest 20% at least the function/method is correct. Saved me much time from context switching.
For me the quick refresh is better as I only need to do it once (until I don't use the language/library again) and that can be done without internet (local documentation) or high power consumption (if you were using local models). And with a good editor (or IDEs) all of these can be automated (snippets, bindings to the doc browser,...) and for me, it's a better flow state than waiting for a LLM to produce output.
P.S.I type fast. So as soon as I got a solution in my head, I can write it quickly and if I got a good REPL or Edit-Compile-Run setup, I can test just as fast. Writing the specs, then waiting for the LLM's code and then review it to check feel more like being a supervisor than a creator and that's not my kind of enjoyable moment.
I agree with you, creating something just feels better than reviewing code from a LLM intern ;D
That's why I almost never use the 'chat' panel in those AI-powered extensions, for I have to wait for the output and that will slow me down/kick me out of the flow.
However, I still strongly recommend that you have a try at *LLM auto completion* from Copilot(GitHub) or Copilot++(Cursor). From my experience it works just like context aware, intelligent snippets and heck, it's super fast - the response time is 0.5 ~ 1s on average behind a corporate proxy, sometimes even fast enough to predict what I'm currently typing.
I personally think that's where the AI coding hype is going to bear fruit - faster, smarter, context+documentation aware small snippets completion to eliminate the need for doc lookups. Multi file editing or full autonomous agent coding is too hyped.
> It seems so inferior to have the same thing but lose access to the browser's suite of tools/capabilities.
Explicitly removing the browser capabilities/tools is a feature imo. i.e to hide the moving parts.
Within a browser, the average low tech users may:
- ...install extensions(ad blockers? dark reader?) that interfere with the app in some way.
- ...mess with the back/forward buttons ,corrupting the router/ui state, or just get very curious why the back button kicks them back to the last page instead of closing a fullscreen modal.
- ...bookmark SPA js-driven pages where the ui state is not fully saved in the address bar(via query params, hashes, etc) and surprisingly find them broken/not working as intended when accessing the bookmark again.
- ...try to copy and send the "localhost" link to their friends and complain.(Hold on to your papers - I've seen this behavior at my workplace!)
All of those will generate complains, useless bug reports and sometimes negative reviews. While some (if not most) of the problems can be solved with great software design and extra care, I think solving these problems comes at a cost that can be very well avoided by simply removing these capabilities.
All but the last list user addordances that are good for the user and that a well designed webapp should accommodate for. The browser has some solid tools to for all of these. I don't see these as a burden, I see them as positive abilities of the web.
They have a million ways of conveying the message of "compress", yet they delibrately choose the most violent and destructive one.
Think of cinematic or cartoon-ish effects like turning the objects into colorful fluid on compress and magically forge an ipad.
Instead of this they choose to show every detail of instruments being painfully crushed and destroyed.
My laptop wakes itself at midnight (somewhere around 12:30pm) and tries to install an update, but it fails every time and just hangs at the lock screen. Because the update is never properly installed, the device will boot up every night until I manually run "Update and restart". Ended up me killing windows update completely.
I've tried to do this a number of times but no matter what I disable in the task scheduler it just ends up reenabled the next time I update the damn thing. Frustrating!
