59
Esk 🐌⚡💜 (@esk@hachyderm.io)
hachyderm.io@thisismissem @mekkaokereke @dma so stats, based on dec 2024 exit run rate (rounded for simplicity):
#hachyderm costs about $1600/mo to run. this is up somewhat, as we've started to add some infra as part of our resilience plan announced in nov.
we currently have about:
- 55000 users
- 9700 MAU
- 3.7M toots
yielding:
- $.03/user/mo
- $.16/active user/mo
- $.0004/toot
from a raw compute & storage perspective.
again, this is based on 100% volunteer work. today, our mods and infra folk graciously donate their time to keep this thing going.
hypothetically, if we paid them, say, $120k USD/yr (chose this to make the math cleaner), that would add $10k/person/mo to the cost.
if we go with a staff of eight (mix of mod & infra), that adds $80k/mo to the run rate, for a total of $81,600/mo, yielding:
- $1.48/user/mo
- $8.41/active user/mo
- the toot figure is silly, so i'm not calculating it again :blobfoxlaugh:
orders of magnitude of difference.
we could argue about the staff size - i went with roughly what we have today and assumed we made everyone full time so they could hachy for 32/hr/wk vs. calculating the number of hours we actually work. e.g. maybe we could it out at ~$4.50/user/mo, but still a multiple orders of magnitude bump from the raw infra cost.
In any case, what number do you think is reasonable? A quick search shows that Facebook employs about 65000 people to serve 3 billion users, 46k users per employee. Even if we were to ask the hachyderm team to be as productive as one of the largest corporations in the world, we would still need at least 2 FTEs.
But given that we are asking them to be as productive as a FB employee, it should be fair to pay them as much as Facebook does, so the real cost per employee goes easily to something like $250k/year (Base salary + bonus + overhead).
So, okay, let’s cut the number of people by 4 and multiply their cost by ~2. We are now talking about ~$50k/monthly cost. That’s still $0.91/user/month, $5.15/active user/month.
The point is, even if “the math” is skewed to make things look “expensive”, even a more conservative estimate has (a) costs per user in the same order of magnitude and (b) cost of labor absolutely dominating over cost of hardware/hosting.
Meta does a lot more than manage a Mastodon server. A single full-timer is likely all that is needed. Two to reduce burnout. Those costs are still high, but you shouldn’t discredit the notion that eight full timers is an exaggeration. The top comments on the toot you link are the volunteers saying exactly that.
In practice, you’d need some redundancy because the admins will also need time off, vacation, get sick.
So, I am not disputing that 8 FTE is too much. What I want to make clear is this: there is not a single instance out there that is getting enough money in donations to pay even one admin, which is a clear indication that the model is not sustainable.
18 months after the API debacle on Reddit, most of the instances are still around. If the model was not sustainable, wouldn’t have all closed?
Your question is as short-sighted as “If global warming is real, then why is it snowing in Southern Europe?”
No, a system that is not sustainable does not imply that all the ecosystem dies simultaneously. It just means that it relies on a continuous stream of idealistic people coming in, willing to help, only to collapse eventually later.
I don’t know that your comparison to Facebook holds water. Firstly, Meta’s employees are spread over three divisions: Apps, Platforms/Infrastructure, and Product Services (ads, strategy etc), where Facebook itself is just one part of the Apps division. Even assuming that Facebook occupies 50% of Meta’s total workforce (likely a massive overestimate), that brings us to around 30k employees for 3billion users, or 100k users per employee. That gives you about 0.5 FTE for your instance.
More importantly though, the job of administering a mastodon instance isn’t really comparable to the job of engineering a social network, so taking a Facebook’s salary or user numbers doesn’t really give us much actionable data. We don’t know how many Meta employees are directly involved in administration of Facebook, or how much they’re compensated.
Ultimately, it’s about what your users are willing to pay. If you can persuade all 10k of your MAUs that $9/month is worth the value they get from your instance, then go ahead. However, I suspect that you’ll be lucky to get even 1/10 of that.