21–30 of 234 posts
The noise would drive me crazy. Earplugs would not even work with so many phones so close together. How can people program in that environment?
Coding from starbucks at the moment. It's definitely doable. Our office has an open floor plan and once in a while there are conversations I wish I could block out a little more effectively but it's not a daily frustration.
Despite everyone hating them, I actually think cubicles are the best arrangement. The cubicle walls offer separation and let people focus and get more work done, but not so much that they stifle collaboration and communication (which everyone having their own office does).
Open office is by far the worst, I actually really hate it. So distracting, noisy, and for sure the least productive.
As an interesting example, the people in charge at Facebook seem to be well aware of the downsides of open plan offices. They however consciously choose to lower engineer productivity in exchange for higher openness and visibility. It's actually refreshing how they don't try to pass it as "much better for the developer". It's better for the company as a whole, not so much for the individual: http://ecorner.stanford.e…
And good open plan offices will give you nooks and crannies to get away from the noise when you want to.
Does anyone know how much it costs to have private offices all round? My estimate was that a 1sqft/yr is $55 in SF right now, so the 100-150sqft for 1 open plan desk is $700, and you might need twice as much space to give people (small) private offices. Would love something more accurate though.
Earlier quoted context omitted.
The open plan is something I don't miss about working at Facebook. You look up over your monitor and see nothing but rows of harsh fluorescent lights. Opinion was definitely mixed, but I was in the camp that found the lack of quiet tiring after a while.
Let's be honest. If we wanted engineers to be most productive, we would give them quiet, private spaces to work. But that is expensive so we put them in open plan arrangements that cost less and pretend that the main reason it's done is higher productivity.
As a rough guess, optimal business value in an enterprise software company probably results from engineers spending 20-40% of their time with users/product managers/other stakeholders and 60-80% of their time actually programming/designing/working with other developers. These numbers are highly imprecise and depend on circumstances, but they're probably in the right order of magnitude.
The issue with private offices is that in my experience many engineers will prefer to spend 95% of their time engineering and just 5% of their time understanding the problem. As a result they build more features but the features are significantly less valuable. 2-3 person offices seems to mitigate that pretty well, while open plan offices seem to go way too far in the other direction.
I really struggle to get in the 'zone' where I can do the creative aspects of software engineering (e.g. researching graph databases like neo4j and determining if our startup can benefit from it) when I'm in an open environment. I have been diagnosed with ADD for the past 8 years, but am working towards a medication free approach to dealing with it. Open spaces really hurt my productivity because my mind naturally tr…
I really struggle to get in the 'zone' where I can do the creative aspects of software engineering (e.g. researching graph databases like neo4j and determining if our startup can benefit from it) when I'm in an open environment. I have been diagnosed with ADD for the past 8 years, but am working towards a medication free approach to dealing with it. Open spaces really hurt my productivity because my mind naturally tr…
Could it be, for example, that companies with worse management practices tend to adopt open-plan offices, but that it's actually the bad management practices causing stress, sickness, less productivity -- and that the open-plan offices have very little to do with it?
In environments as incredibly complex as office settings, it is particularly important to realize that correlation != causation.