Hacker News Reader

Comment by quatrefoil | Hacker News Reader

I don't think it scales linearly anywhere. A significant part of the workforce is just there to provide services to the workforce itself. You have full-time engineers just maintaining tools for performance management, or for booking conference rooms, or requesting new laptops, you have legal teams for basically every country you operate in, etc. And then once these bureaucracies grow large enough, you hire people whose job is just to help others navigate the internal processes...

I mean, LinkedIn has what, 20,000 employees? If you had spherical developers in a vacuum, I doubt you'd need more than 40-50 to maintain their products, but there's more to running a tech company.

View on HN

> A significant part of the workforce is just there to provide services to the workforce itself.

The bureaucracy is expanding to meet the needs of the expanding bureaucracy.

Is there a rocket equation for bureaucracy? Something to model the exponentially growing efforts of a company that are spent self-support and bureaucracy as the amount of productive work scales up?

No rocket equation but, if you want to go to the source, see Parkinson's Law (the original essay, not the shorthand aphorism). Which was actually even worse--productive work was not scaling up in the British Navy.

I'm sure there are papers that have studied what happens specifically in business though.

Yes, the original Parkinson’s Law paper contains an equation:

https://en.wikipedia.org/wiki/Parkinson's_Law > The growth was presented mathematically with the formula x = (2k^m + P)/n, in which k was the number of officials wanting subordinates, m was the hours they spent writing minutes to each other.

(check the original paper for details since this obviously doesn’t explain what n is)

> I mean, LinkedIn has what, 20,000 employees? If you had spherical developers in a vacuum, I doubt you'd need more than 40-50 to maintain their products, but there's more to running a tech company.

This is a ridiculous take. You would need way more than that just to maintain Kafka and run ops for it, that LinkedIn heavily relies on. You are trivializing the complexity of running a platform like LinkedIn.

Perhaps if reducing the complexity was considered an urgent necessity, over hiring more, there would be less complexity and new opportunities would be easier to consider and jump on.

Complexity compounds, so the tradeoff is, continually tame it, or continually hire with less and less impact per employee.

The latter seems to happen a lot when money flows.

Organizations can't scale linearly. You need more middle managers as you add more people and organizational layers, who don't contribute any "output" so this is trivially true.

Alternatively, you could have thousands of people all reporting to a single manager, and if you think that could work, I have a bridge to sell you.

>I doubt you'd need more than 40-50 to maintain their products

Only if you're certain they will never quit their jobs. In tech where the average tenure is supposedly 2-3 years, a large number of employees are hired to make key persons redundant, groom for succession, and to make sure key persons can offload some bandwidth to work on emergent issues. It's irrational to have billions of dollars of revenue (or even tens of millions) riding on a single person who can decide to walk away with 2 weeks notice.

> spherical developers

This could be a physics model, or an imperative to scale back the free food.

Reply on news.ycombinator.com