Hacker News new | past | comments | ask | show | jobs | submit login
Ask HN: Struggle with project ideas? Heres some novel ideas you can build now
40 points by purple-leafy 3 months ago | hide | past | favorite | 17 comments
Hey HN I love thinking of ideas, I do it almost all day everyday - and I'd love to share some with the community rather than hold them close to my chest. I'm trying to be more open-minded, and giving, rather than always taking.

-----BROWSER EXTENSIONS-----

1) Natural language browser commands:

Develop a browser extension that allows users to input natural text or voice commands, with immediate outcomes. The commands are passed to an LLM which determines what the user wants and runs the correct pre-configured browser APIS. I've already tested a prototype and it works really well with Google Gemini Flash model. You just need to pass to the prompt a JSON mapping of keywords/commands to APIs. For example:

User: "Clear history"

LLM: <Runs pre-made clear history API>

Stack: TypeScript/JavaScript/Google Gemini API

2) Declutter Mini-map -

A browser extension that when activated on a webpage, adds a mini-map of the webpage to the dedicated chrome extension side-panel (a dedicated 320px wide screen area where chrome extensions can host their ui permanently). The mini-map is special in that it only features webpage "landmarks". Landmarks are groups of functionally related components that are physically near one another. Examples are headers, footers, content sidebars, video players, etc. The mini-map features solid rectangles representing the size and position of these landmarks. The user can then turn landmarks "on" and "off" by clicking on the landmark on the mini-map. Its basically a generalist webpage element hider/declutter-er

Stack: JavaScript/React

3) UI Copilot -

A browser extension that allows users to capture a screenshot of any portion of a webpage that is automatically sent to an LLM for generic UI design feedback.

Stack: JavaScript/Svelte/ChatGPT API

-----WEB APPS-----

4) Pinterest for developers

5) Extension Sandbox -

An in-browser sandbox, that allows you to try browser extensions before ever installing them.

6) Peer-to-peer whiteboard -

Miro, without the server and database. Whiteboard only stays up as long as users are using it

7) Self-hosted internet archive and bookmark repository

In the world of AI bloat, capture, record, and keep content that matters to you. Create a Browser extension that can screenshot and capture static web content, and intercept bookmarking.

All this content is funnelled to a companion self-hosted website that hosts all your saved content

----

What will you build?

What would you like to see built?

What ideas are you sleeping on?

Wanting to collaborate? I'm currently learning GO and know frontend, but suck at UI design!




One thing that has worked for me for over 30 years (when you have a lack of a fresh idea) is to just recreate and try to improve on something you admire that someone else made. Pay special attention to products you pay for or use for “free” but are not solely owned by you (25 years ago I was amazed by all this free stuff people were giving to me, being perhaps too naive to think there were no strings attached or unwanted side effects).

Use some cloud based product? Build a locally hosted version. Paying for an app that is useful, but not quite worth the subscription? Try building it. Some may be harder than you think, perhaps not even worth it. Some may already be available in other free alternatives, but maybe it’s worth while just making it your own anyways. The worst thing that happens is you learn something exploring new problem domains and trying new technologies or techniques.


This is a great technique thanks for sharing. What are some services or projects you’ve recreated over time? I love the idea of self hosting as of late, but I’m yet to do it!


Also see Challenging Projects Every Programmer Should Try

https://news.ycombinator.com/item?id=38768678


While we're throwing ideas, someone please implement this:

An as simple as possible Android app that lists any discovered Chromecast devices and when you tap on one it attempts to cast your screen to that one.

Why? Because screen casting isn't an AOSP feature. It is really difficult to get it to work on GrapheneOS if it works at all.

But VLC can stream video to Chromecast on AOSP without issues.

So my thought is: Take a look at the VLC Android source. It's mostly a wrapper using the VLC bindings as an API. Look at how it uses VLC to discover Chromecasts[1] and how it uses them as output[2].

Then somehow pipe the createScreenCaptureIntent API into it.

GrapheneOS users will be thankful.

[1] https://code.videolan.org/videolan/vlc-android/-/blob/master...

[2] https://code.videolan.org/videolan/vlc-android/-/blob/c773c2...


I'm currently working on a .NET hosting solution where you zip your published .NET app and then drag the zip file onto the site and it's automatically hosted and a unique URL is created. Takes about 3 seconds once the zip file is uploaded.

Kind of like tiiny.host but for .NET... it's almost ready too :)

It was one of those "I wonder if this is possible" type of ideas I had in the shower and would you believe it... it bloody works :)


That’s sounds like a cool useful project!


Thanks. Just getting rid of the showstopping bugs at the moment... as always, works great on my local dev machine :)

It'll be running on Hetzner in the background so I'm just working on getting all the moving parts synchronized (and writing the docs!) and it will be ready for early access.

Hopefully in a month or so.


Currently working on a new way to enable low/no-vision people to hear what's on their screen in real time. As far as I can tell from my research my method is novel.

Once I can clear the decks of my main projects I want to reach out to potential user bases and get some beta testers.

I keep a list of project ideas in Obsidian. Whenever something pops in my head I write it down there. I make a habit of revisiting the list before I pursue something else.


Hello fellow builders, I just created a list of 60,000+ Manifest V2 extensions, their links and some details, because I see a massive opportunity for developers to step in and save popular extensions from extinction by building their ManifestV3 versions. Anyone interested in this list? Here's it https://abasiofonessien.gumroad.com/l/onpedu


I am currently building out a project that lets users store and send encrypted files through my service [1]. Note that the UI sucks as I, like OP, suck at UI design! Edit: If you would like to help with this project, contact me via email (which is in my bio). Note that I can't pay you anything as I am a lone, broke developer lol

[1] https://protected-docs.io


Cool idea, how deep into it are you? What’s your plans for the project?


Well, I've got sign up, plus encryption working with gpg. Sending files could be more intuitive and the UI needs (a lot!) of work lol. But it is coming along nicely imo


Well done! Is there any reason you wanted to pursue this project? Does it aim to differentiate itself in any way?


Somebody approached me with the idea, and, having more time than money, I decided to develop it and put it up online on hosting I already had. It only costs me about $30/month, and that includes my other sites, so I don't mind hosting it for a bit until it gets users.


To quote Edison:

  Having a vision for what you want is not enough...
  Vision without execution is hallucination
source: https://www.azquotes.com/quote/664713

Put another way, pick one of the ideas you fancy and pursue it to completion.

The journey will be far more rewarding than the destination.


I’m always executing - I’ve already built 9 of my own ideas in the past year, I have a huge list of ideas so sharing some with people rather than holding them tight to my bosom. See [0] for my most recent execution (finished yesterday)

I’m always executing, but I’m also always seeing people short of novel project ideas.

Perhaps this is also my call to the void for more collaborative efforts.

[0] - https://chromewebstore.google.com/detail/ui-copilot/hgaldpfd...


>> The journey will be far more rewarding than the destination.

> I’m always executing - I’ve already built 9 of my own ideas in the past year ...

No worries my friend. I only suggested what I did within the context of this thread :-).




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: