HACKER Q&A
📣 surrTurr

What do you want to build but don't have the time for?


What do you want to build but don't have the time for?


  👤 surrTurr Accepted Answer ✓
I probably won't have the time to build it, so I post the idea here.

Pitch:

  I've been thinking a lot about the current state of social media. It feels like we're all trying to curate this perfect image of our lives for the world to see, but deep down, we all know it's not real. I've noticed a lot of us, especially the younger generation, are yearning for something more genuine ^1^2. Furthermore, I am in a long distance relationship and wish there was an app to help us connect more.

  \*Introducing "For 2"\*

  It's a simple app designed for couples or best friends. The core idea? You can connect with just one person. It's not about broadcasting to the world; it's about sharing moments with that one person who means the world to you.

  \*Core Features:\*

  - \*Daily Image Share:\* Every day, send one image to your partner. It's a way to share a moment, no matter how mundane or extraordinary.
    - \*Home Screen Widget:\* Display today's image right on your home screen (akin to Loket^2).
    - \*Timed Image:\* Maybe there's a specific time both of you are free. Set a time, and you both get a notification to take a picture. If you miss it, you can't view images from your partner that day. (Still mulling over this one. Thoughts?)
    - \*Live Sync:\* Imagine both of you taking a picture at the same time and seeing each other's live view. Cool, right?
    - \*Filters & Frames:\* Play around with fun filters or split frames to make paired poses.
    - \*Streak Indicator:\* Keep the connection going and see how long you can keep the streak alive.

  - \*Chat:\* Because sometimes, an image isn't enough.
    - \*Reply to Images:\* Share your thoughts on that daily image.
    - \*Snap-style Media:\* Send media that's visible just once for those fleeting moments.
    - \*Regular Media & Voice:\* For everything else.

  \*Thinking About Adding:\*

  - \*Map:\* A way to see where your partner is.
  - \*Stories:\* Click on their profile pic to see their day's story. No notifications, unless you haven't checked it out by day's end.
  - \*Sound Message:\* Once a day, send a sound notification to your partner. A fun way to say, "Hey, thinking of you."
  - \*Daily Question:\* Once a day, you and your partner answer an interesting question and you can read each others answers
  - \*Calendar:\* Shared calendar

  [^1]: https://bereal.com/en/
  [^2]: https://apps.apple.com/us/app/locket-widget/id1600525061

👤 mlhpdx
Malloc for the web.

It’s famously what S3 was supposed to be, but isn’t. Remote memory as an API would enable some cool new serverless architectures. I have neither the time nor know-how to build it, but would love doing so.


👤 ilaksh
I don't know exactly what it is, but there have got to be ways to combine an LLM with a text adventure/D&D/RPG engine or something similar.

The engine will enforce some structure on the world so that it stays consistent, rather than having to rely on just the LLM context which will run out. Info about this structure and world state gets fed in with each prompt.

The LLM can call functions to roll dice or look up monsters etc. It will ensure that user commands are understood without requiring the user to hunt for the right keywords. The LLM can provide variations on descriptions based on random inputs, time or season, world state, etc.

There are surely some ways to combine that with image generation.


👤 coreyp_1
A lot of 100% self-hosted home automation. (I absolutely hate how every device wants to install another app on your phone and you have to create an account, etc... No, I will no download your app in order to be able to integrate with Home Assistant! If I can't use the product after you've either gone out of business or chosen to no longer support the device on your "cloud" offering, then I'm not interested in buying it.)

For context, I already have a Home Assistant setup. I have replaced ~35 switches with either Zigbee or Z-wave controlled switches, and have a dozen other Zigbee devices. I've gone all-in, so to speak. I hate the Z-wave fan controllers, but they were the only option for what I needed. Z-wave is completely unreliable in my system, but I love the Zigbee stuff! I have automated everything from my pool pump to my garage doors, to my exterior lighting, and I absolutely love it!

As such, I have 6 things that I am planning to build, if I get the time:

1. A digital mixer for my pre-wired whole house home audio. (When I bought my house it had a lot of cool stuff, but it was all older. Oh, and the previous owners took the sound control system, so none of my speakers work.) I want to build a digital mixer that can control 16 channels in, 16 channels out, capable of sending any audio source to any channel, wholly controlled over my Zigbee network. I've already been working on a circuit design on EasyEDA, and sourced the ICs that I want to use, but I have to have time to test and iterate on the design. I'm anticipating spending $500 just to build my own, but I think I can build 2 for that cost, so I'll take one to my church and put it in line with the existing mixer so that I can control it from the front (I'm the pianist) for the weeks that we don't have a sound guy.

2. A self-hosted control for my sprinkler system. There's already a few on the market, but you have to download their app. My last system lasted 15 years... If I buy one of these "you can control it with the app!" systems, then how long will they continue to update their app and maintain their servers?

3. Tying into my current hard-wired security system. I'm sick of paying for subscriptions. I want to be able to turn on lights (Zigbee controlled switches) automatically when my motion sensors (which are already wired into the house) pick up movement. I want my home assistant to be able to remind me when I leave and forget to turn on the alarm, and then be able to turn it on from my phone.

4. Video surveillance of my property. No Ring, Nest, SimpliSafe, Wyze, or Arlo. I don't trust any of them. I want my system that is 100% under my control. I'll pay for the cameras. I'll pay for the hardware... I don't care. It's just a pain trying to figure out how to build one that will integrate with my existing Home Assistant server.

5. A self-hosted digital media server. Yes, they already exist, but I have to figure out how to incorporate it into my intended setup. In short, if/when I move, the home automation system would stay with the house, but I would want to take the media server system with me.

6. Electricity use monitoring. Every DIY option that I've seen requires you to use their app. Why is this so difficult? My data should never have to leave my own house! That, and most monitoring solutions are very limited in their inputs. I have a lot of circuits (like maybe 70?) in 3 electrical panels. Most monitoring solutions cost several hundred for 6-12 monitors. My house isn't exactly that big, either... it was just built and added on to at different times. For this one, I already have parts on order to start building my own. We'll see how far I get.