5 ms·
(As transaction costs (search costs, costs to contract, costs to outsource, etc.) go down, optimal firm size goes down. Optimal size will still depend on the i
by bdunbar 15y ago
(As transaction costs (search costs, costs to contract, costs to outsource, etc.) go down, optimal firm size goes down.
Optimal size will still depend on the industry.
Seventy guys can operate a dropbox.
You'll need a few thousand guys, minimum, to run a shipyard, assemble tractors, extract oil.
Doesn't mean they are inefficient - just that some tasks require a lot of capitol and manpower.
- ciex 15y agoThe point is that a shipyard could also be run by a number of smaller companies, each doing part of the job, if only they can collaborate effectively. Capital and manpower can be divided between firms.
- rdl 15y agoThere are a lot of big construction projects (and even factories) where you get exactly that. Some of the staff on-site at the same facility work for sub-contractors (they may be small companies, or may be a detachment from a huge company). e.g. the guys running IT are a legacy company are probably employees of a big enterprise IT vendor (ITT, IBM, ...), not Big Steel Company. The guys maintaining the radiation camera are probably with a specialist industrial radiography company, not the big company.
- bdunbar 15y agoRDL already pointed out that shipyards (and other large manufacturing sites) do operate this way. There is one problem with a y-combinator approach to capital intensive work: someone has to be in charge. There is 'a' party responsible for delivering the product to the customer, and that drives everything else. Put ycombinator in charge of a shipyard and a lot of cool stuff will happen, but delivering a finished boat on time and on-cost is not one of them.
- 3pt14159 15y agoI agree with you, but not in the longer term. Robots + AI = 80 person Standard Oil.
- mattadams 15y agoYeah, that is probably never going to happen or if it does it's so far out that it's not worth speculating on.
- 3pt14159 15y agoThink about the amount of data the 70 people at Dropbox move. A couple hundred years ago that would have taken every single literate person.
- jey 15y agoHow do you justify that very strong assertion? What's your definition of "far out"? The past 100 years are a great example of how a lot of change can happen in a short amount of time.
- Qworg 15y agoThere already are unmanned oil platforms. As communications and remote systems get better, I expect to see even more automation. http://en.wikipedia.org/wiki/Normally_unmanned_installation http://en.wikipedia.org/wiki/Normally_unmanned_installation http://www.shell.com/home/content/innovation/meeting_demand/monotower/ http://www.shell.com/home/content/innovation/meeting_demand/...
- mattadams 15y agoI'm aware that there are unmanned stations -- not only oil platforms but regular leases. I live in oil & natural gas country (Northern Alberta). What you don't see are the thousands and thousands of guys running around keeping it all ticking. There are some things that you can automate -- but a lot of things in the oil & gas industry cannot be automated. Specifically, the leg work required to get things started & keep them maintained. Automating those sorts of things requires a pretty advanced AI and I don't see that coming around the corner anytime soon.
- rdl 15y agoYou can make things simpler for dumber AIs (or dumber workers) by designing field-replaceable-units with that in mind. It is going to cost more/be less efficient, so as long as a human costs $300k/yr or less, no need, but when that isn't the case, you see "forklift upgrades", etc.