Sendbot - the easy way to share & recommend websites, videos, books, and music to your friends.
A tiny side project, built to solve a frustration I have always had trying to share things with friends.
I've always struggled with how to best recommend music, articles, and other pieces of content to my friends. Typically, I will send a text or WhatsApp message, but those usually get lost in the chaos of the day very quickly - and for things like music or video links that are not vitally important, it means they are sometimes forgotten and never seen.
The onus, usually, is put on the 'receiver' to open the link and read it then, or bookmark it to come back to later. This feels unfair in a way, I'd prefer the organization to be put on the 'sender', making it as easy as possible for the receiver to browse and ingest the links/articles/videos whenever they can.
I've been toying with this idea in my head for years now, and this Sunday decided to make a basic implementation and start testing it this week by recommending my friends various albums and blog posts that I loved in 2024.
How it works
Right now, the input of the Sendbot app all works via a Telegram bot. I chose Telegram as I wanted to make the input - the recommendation - as easy as possible, whether I was on desktop or on my phone. I'll be adding a way to add links via the website soon - but I think this is a great little MVP (and my first time working with the Telegram API, which was fun).
- Sign up and message the Telegram bot. Add your name via the /setname command, and you are good to go!
- Add your friends (via the /add command), which creates a new URL and database entry for them, connected with your account.
- Then, simply share a URL into the Sendbot chat in Telegram. It will quickly which of your friends it is for, and if you want to add a note. That's all!
- Your friend will then be able to check their individual page (eg. sendbot.xyz/patrick/nick will be the individual URL for links sent from Patrick to Nick) to see all the recommendations you've sent them.
Little things
The app will automatically categorize your link based on the URL and data which is scraped when you share it to the bot.
Your link will be categorized into:
- Music (eg. Spotify, Soundcloud Links)
- Movies/TV Shows (IMDB, Netflix Links)
- Articles (Medium, General Web Links)
- Videos (Youtube, Vimeo Links)
I will soon add Google Maps support and a few other meta categories.
Beyond this, your friend will also have an RSS feed they could subscribe to, which will update every time you recommend a new link to them. I personally find this super useful as I prefer to read updates via my RSS reader app rather than checking a website every day/week to check for updates.
In the near future, I'll also add the option to send an email notification to your friend when you recommend something to them. I think this will be good, but I need to find a way to not spam the friend when you suddenly add 15 new songs to your recommendation queue for them - some sort of debounce / delayed queue is likely the answer here (or simply an email at the end of the day after recommendations have been added).
The plan
This is still very new, and I don't know where it will end up, but at the very least I will surely use this myself and see what my friends think of it.
A feature I really want will be individual lists (maybe need a different word for that) - say I want to send my friend Jace a list of my favourite 2024 albums that I think he'd like. I should be able to create a specific page / list that allows the same recommendation links, but pigeonholed into this one specific project/list. I could see this being useful for music playlists, as well as trip planning.
I may try to release the above list/categorization feature as a 'premium' feature, and see if I can monetize this and ask for a few dollars a month / year for it. We will see about that - I am trying to think of this from the perspective of my non-tech friends and I struggle to see how any of them could see value in paying for something like this, but I could be wrong.
Will likely post it to Hackernews when it's more polished, I found a link from 8+ years ago of someone who had a similar concept, and it got a lot of praise on HN - but their app is no longer available.
Comments ()