Sounds cool, it could find videos already hosted various sources on Nostr as well as an option to upload a video (using either a designated host or hosting partner like Nostr.Build) and a percent of zaps could be given to devs and the host of the videos with the majority going to the user.
Discussion
You could also have an option to only watch from your follows and a fully open one where it shows videos all across the Nostr protocol, you could have the ability to follow hashtags and block hashtags as well. Just spewing ideas here.
# nostr-shorts
### Functions
* Upload video
* Host a link to video
* Post video link to blastr with:
* Tag npub of uploader
* Include any message from uploader
* Track zaps of note in a database
* Hosting consumes XX sats/minute; first 120 minutes are free
* Share zap dividends to users
* Allow creators to sell ad space on their uploads
I think having a central account is sort of useful as people can just follow all the videos or you could segregate videos to subject accounts (screened, unscreened, art, dance, bitcoin, etc.). I think you could also have the user sign a re-post event right in the website with nip-07 or nip-46
I only I could build this 😢
Currently asking Deepseek-Coder: 33b running on my start-os to do this 😄
Assuming each video got at least 20 views, then a single video would have a hosting/data/processing cost of 280 sats. Given that most posts don't get 280 sats, the users would need to pay some fee so that you don't rack up a lot of money-losing videos. Let's say you charged 500 sats per upload, then you would break even on every un-zapped video that didn't get watched more than 20 times; there is also the option to sell ad space in the video notes.
Videos that got watched a lot would need to be zapped or rapidly decay. An image linked within the post could be dynamically linked server-side to appear to change, indicating the relative zap-health of the video.
The cost feasibility of this concerns me as people expect video / image hosting to be free. People are so used to being the product that the transition to micropayments for services will be a difficult bridge to cross.
Hmm, I don't think it would cost that much, mabye it could use Nostr build 😂
Just extrapolating from here: https://aws.amazon.com/cloudfront/pricing/
I am sure there are cheaper options, but this is what it would look like running at enterprise scale. The free tier would make the service profitable for early stages with few DAU, but eventually it would be a money losing enterprise. Hard to find investments that have a better return than holding bitcoin.