HN user
I believe there are only 43 because they require "review" before being posted officially.
It is likely that they do not approve the majority of comments, I submitted one but it has not yet been approved.
This seems like a great way to get an app running in production. Seems like Vercel and Supabase combine to make a pretty solid modern heroku replacement.
I think a lot of people who have never run a company view business this way.
If the discussion was purely about cost (and I assume it's at least partially), then you'd be right that laying off 20 engineers would provide the same impact to the cost bucket.
However, how much revenue is generated on average by each facebook engineer? I can't find an exact amount, [disclaimer: fuzzy math ahead] but looking at their q2 2022 earnings (https://s21.q4cdn.com/399680738/files/doc_financials/2022/q2...) they employed 83,553 people and had a total revenue of $28,822,000,000 for q2. Presumably each of those employees contribute in some way to the revenue, on average resulting in $344,954 in revenue per employee per quarter assuming that's static (it's not) that's $1.379mm in revenue per employee per year.
How much revenue do janitors cleaning empty offices generate for Meta? Roughly $0, there might be some extrapolated value to a clean office you can indirectly attribute. But regardless, they're an $8mm/year cost with no revenue directly attributed.
So laying off 20 engineers you'd net $8mm/year in cost savings, but lose $27.58mm/year in revenue (hypothetically), leaving you with a $19.58mm net loss compared to laying off 200 janitors with $0 revenue lost. Obviously it doesn't work out 100% this way, but hopefully you this helps you to understand what math the accountants at Meta are doing.
I'm not suggesting that this is the right way to run a company, but it is the way public companies are run. I'm surprised no one in the comment threads so far has considered this, so I felt it necessary to at least mention it.