One of the times this came up[1] someone made a good point (wish I could save comments :) ) about the important distinction between sharing an open office with

One of the times this came up[1] someone made a good point (wish I could save comments :) ) about the important distinction between sharing an open office with engineers, versus random people walking by/loud sales cales/managers/etc. In the first case, people don't mind interruptions - from other engineers. In the second case, interruptions are terrible. So according to that comment, a group of expert engineers working in a shared space can really experience some synergy. sometimes if you can say, "hey do you have a sec" and ask a super-quick engineering question, it can really help. If a few people start discussing architecture, everyone can join. etc. Apparently it works. If it's other engineers. This makes me wonder. Reading our present article - what we're really missing is some data. Really, any data. All of those people who have beaten it into us that "correlation is not causation" have won a Pyrrhic victory. We're left without even an indication of what the correlation is. I have no idea whatsoever whether the data points plotting productivity (y) and space per worker (x) is a random scatter plot, clearly correlated positively, negatively, has a peak, has any shape at all, or whatever. I don't care what data you're showing. For the y axis you can use startup market cap/valuation per employee for all I care. You can use lines of code - a terrible metric. Or whatever other proxy for productivity you want. But whatever you use, anything is better than nothing at all. What I'm wondering is: in a software or startup context - or any context at all - what is the shape of the correlation at all between productivity and space per worker? This seems to be a central question, but the article only hints at an answer without presenting any data at all. Since it's clearly a choice all organizations and startups face - let's see it! - [1] Other discussions on open offices: https://hn.algolia.com/?query=open%20office&sort=byPopularity&prefix&page=0&dateRange=all&type=story https://hn.algolia.com/?query=open%20office&sort=byPopularit... (top stories) https://hn.algolia.com/?query=open%20office&sort=byPopularity&prefix&page=0&dateRange=all&type=comment https://hn.algolia.com/?query=open%20office&sort=byPopularit... (top comments)