I hate subscription services. And I hate money-grubbing corps. Especially when they try to profit off of your own data.
That said, this is not that as infeasible as it sounds. The dev for Tessie reportedly has 400k users. That’s roughly $12.50/month per user. Modestly speaking, if the dev charged their users $13/mo, he’d profit $2.4 million per year. For $15/mo, he would profit $12 million per year.
That’s probably what Tesla is hoping their devs would do. And I’m sure a lot of Tesla owners could afford the fee.
Yes, I didn’t account for the transaction fees. But I believe my point still stands. If people find enough value in it, they’d probably pay for it; and that’s why Tesla is charging what they are.
I do agree with you about it being batshit crazy. If it were me, I sure as hell wouldn’t want to pay $15/mo for that. But I try to be a cheapskate where I can.
We can probably look at the reddit paid API change as a real world approximation of free to paid conversions versus users just dropping usage. Not reddit usage overall, but a third party app that stayed and needed to start charging users because of the paid API. Not sure if there are any apps that stuck around and released that sort of information to reference.
RedReader is still going, and is still free. I've been using it since essentially its first release over a decade ago. I was the first to suggest the dev accept donations, around 4 years ago, as it's a one-man job and it used to get regular updates. And the dev has stated he wants to add lemmy functionality alongside the existing reddit code, so users can switch between the two like an account switch, and maintain the UI they are fond of.
(I'm using Thunder currently while I want for RR lemmy support, but obviously I quite like the app, have recommended it to many)
Per user costs for a website is on the number of pennies a month and most of that is for electricity.
I can plug in a $750 second-hand server with a xeon processor, 40 TB of storage and 128gb of ram and easily serve all of the needs of several thousand users on essentially any website type for $1.50 a day.
Sure, if you throw in video and a lot of bandwidth then the number would go up, but for pictures and text and website interaction on the par of bluesky or twitter or mbin sans hosted video it would work very well.
If I reached the point where I needed to expand for the raw processing I can just throw another $1,000 and $45/month in electricity at it and double how much I can handle.
Computers are stupid cheap. Internet services are stupid cheap. Asking for more than a dollar a person per month for anything that doesn't have licensing fees on it (like tv/movies) or very high bandwidth usage (like YouTube) is a greedy rip off.
That being said, at those prices I would not make anything for running the service, and that also would not cover additional development costs for any new features that needed to be added, but even so, unless your goal is to disenfranchise users you should not charge more than a buck a month or hell, $10 a year per person for all of their access to your service.
Selling literal shit at a restaurant also isn't unfeasible if the customer doesn't care about eating shit. But nobody is going to eat shit and nobody (normal) is going to pay $10+ a month to get mostly gimmick features. At a glance there's barely anything useful in the API.