When someone asks me what a software engineer's day is like I give them the following description:
"Remember back to high school math class. Now imagine 8 hours of high school math class back-to-back-to-back. Now add in "pop-quizes" (e.g. production issues) sometimes multiple times a day."
I think that once someone can relate to what day-to-day software engineering is like, they can empathize with why "walls" and "doors" are nice.
As someone who has only ever worked as a programmer in open offices or from home, I would be so much more willing to RTO if I had even a cubicle, so say nothing of an actual office with a door. Most of my friends who were dragged back into an office don't even have an assigned spot in an open office.
I don't need a door; a traditional cubical with high walls would be good enough. Open offices are absolute hell for ADHD. Even with medication, I can't be productive.
It really kills my productivity too, although I don’t suffer from ADHD. For me it’s about the feeling of being seen/watched that makes me feel anxious and self-conscious all the time. I’d also be happy with a cubicle as long as it blocks line of sight.
The same anxiety occurs with camera ON Zoom meetings. It's okay to have the camera on for a few minutes during a meeting but if folks aren't talking they shoul dbe allowed to turn it off. This is because you can't tell who's actually "watching" (looking at) you so you are in a heightened on edge state. This is the cause of Zoom metting fatigue. In an actual meeting room you can glance around a definitely tell who, when, and how often people are watching/looking at you.
I don’t have ADHD and it’s difficult for me too. I remember being in an office watching the marketing guys toss a football back and forth in my peripheral vision while trying to concentrate. It made me hate those guys and our management.
I have ADHD and cannot agree more. I have to go into the office for one week at a time 3-4x a year.
Almost nothing gets done.
Well, stuff gets done like helping folks out with their work by looking over their shoulder and helping walk them through things they are stuck on.
Our leadership is okay with this. They want us to have the face to face in-person time. And I will admit, it does help build relationships with those who do not excel at building relationships through a chat client.
I find that even working from home is too distracting, and I live alone.
My most productive programming hours are 11pm to 8am. I'll pull an all-nighter maybe once or twice a month and I get a stupendous amount of work done. That's the only time I feel a sense of complete solitude and focus. No Slack messages, no phone notifications, no noise from my neighbors, not a single disruption whatsoever for hours on end. And knowing I have a horizon that stretches on for so many hours somehow allows me to relax and tackle the most difficult tasks.
Yes! I am much more productive in those hours too despite being "a morning person" with regards to getting up early and being spry. On my own hobby work I find myself crushed by hitting flow at like 1 am, finally, and then oh crap I've gotta work in six hours, better get some sleep
> These two types of work were first identified, to my knowledge, by the programmer and essayist Paul Graham.
Repeating my comment from https://news.ycombinator.com/item?id=38693398 when this came up last month, DeMarco and Lister described flow state in their 1987 book "Peopleware: Productive projects and teams".
A totally different reason, I think, actually. If your job is to manage people, that sometimes might involve conversations where you provide one-on-one feedback. This is sometimes negative, and the recipient might not want to share that, so it is good to have the option to make it private.
As you say it yourself it happens sometimes. So they can have these one-on-ones in a conference room, can't they? And yet, somehow they need their private offices at all times. Strange.. :)
I can't speak for all managers but I did some analysis on this because, when we came back from COVID, we were trying to keep the number of used offices to a minimum as they took a lot longer to daily deep clean.
For me personally, over a period of about 6 weeks, I needed a private room for a sensitive conversations on at least 3 days in each of those weeks. 2 of the weeks were for all 5 days.
3-4 of the conversation were related to employee's health so probably more than a typical period. But the period didn't cover any of the usual high sensitivity events, like appraisals or comp, so it's a bit of a wash.
Now that doesn't mean that I need exclusive access to an office (and I don't) but it is frequent enough that it makes an important part of my job very hard without easy access.
But that all obviously depends on details as to how many people you manage, org's views on management responsibility etc.
The frequency is probably the key factor. Also being able to have their many meetings without adding to the general noise of the office.
I advocate for most employees having offices, but we can still acknowledge that the reasons you probably want your managers to have them are very different than the reasons for engineers.
I think it is easier to give everyone a door. I get that people treat offices as a special sign of prestige or something, it is dumb.
But really it does seem useful to have a door. Bad news might be rare, but unless you want to have the conference room imply bad news, you also have to schedule good news meeting in the conference room. Now all the meetings are in the conference room. Managers’ whole job is to go to meetings as far as I can tell, so we might as well give them the conference room I guess, haha.
True, but the answer to that is to use conference rooms — either existing ones or special, smaller ones intended specifically for two or three people to use. It’s what non-managers have to do whenever there’s a need to speak freely.
I think it would require a bit of “opsec” type thinking on the part of managers, maybe? If you always take meetings in your office, it is pretty easy to close the door discreetly.
If you book a conference room for one-on-one meetings, you have to do all your one-on-one meetings in the conference room. Otherwise, obviously, booking the conference room implies negative feedback. And if someone wants to do an impromptu talk that goes in a confidential direction, there’s some impedance added… none of this is to say it is impossible or unworkable, but it seems easier to just give everyone a door.
People can and do have their conferences and calls and 1:1s in conference rooms. We are definitely not inventing the wheel now, this kind of setup I'm sure can be seen everywhere, and long before covid. I won't even blame the NIH syndrome, but can't stress enough: there ARE solutions to everything, and they can be (re)applied once there's a will.
When I was a manager meetings were darn close to my whole job. Sure there was some quiet work, but when if I'm going to be spending 6 of my 8 hours in the conference room, you might as well save space and just make the conference room my office instead of giving me a cube.
Also I think the focus on private conversations is a bit of a red herring. I can assure you, none of the engineers or sysadmins that reported to me wanted me in the middle of the cubes, constantly on the phone or having a conversation.
To be clear I think most employees should have offices.
I sure feel this. I've worked at home for the past 30 years and I've asked and pleaded with my wife and children (all adults now) to let me work uninterrupted and they've ignored that the entire time.
I've tried to explain to them why I need to work without interruptions, much the same as the author of this has described, but they've never given it a thought.
I cringe every time the phone rings because generally the 1st thing they ask is "are you busy" and I always answer "yes, I'm working" and they blow right by that to tell me why I need to do something for them right away.
> These two types of work were first identified, to my knowledge, by the programmer and essayist Paul Graham
I would hope there's significant amount of neurological research predating paulg's 2009 essay. Does anyone know? Is this just a version of "don't interrupt someone in the state of flow"? Csikszentmihalyi spent decades researching that.
I have trouble believing that the idea of “don’t distract the people trying to concentrate” was ever really invented anyway. “Oog make good sharp rocks for arrow heads. Why you bother Oog with request for round smashing rock? Go bother Uug, Uug perfectly capable of picking round rock for smashing, or go to river and pick one yourself.”
>These two types of work were first identified, to my knowledge, by the programmer and essayist Paul Graham
my sweet summer children, private offices for every programmer was a thing at least as far back as around 1980, and at that time we didn't think "oh, we've just identified the need for this."
I was myself already noticing that I also liked working in a "bullpen" (big room, terminals all plugged into shared computer) and swapping jokes, syntax tips, and the like with my neighbors. Actually, when I read about people on here today talking about cli stuff, I think "it only seems a mess to you because you don't have more experienced folks sitting next to you in the bullpen"
I reckon there’s just as many programmers that want to work siloed away in a private office as there are those (like me) that prefer an open plan. I think a lot of this “abstract concepts stacked into a glorious philosophy” nonsense is just jerking oneself off about how smart we are.
If it's purely the question of cost per employee in favor of open office, why don't companies charge for private offices? I hope nobody actually believes that open office is measurably better in any other way.
I'd happily take a reasonable pay reduction for a private office but, as you can see, most people wouldn't. Rather shortsighted of them since the increased productivity pays off in bonuses and career growth.
> the increased productivity pays off in bonuses and career growth
Anecdotally, neither I nor my friends have seen this happen. Putting in more effort just leads to an increased workload along with a "Meets Expectations" during annual reviews. I have a great manager now but their hands are tied as far as bonuses, promotions, and raises go. It's a bad sign when the best benefit I get is that my workload isn't cranked up because I complete tasks more efficiently.
I'm still young, but I doubt that I'll see anything similar to a bonus or significant pay raise due to performance in my career. I still put in my best effort, but I don't see any reason to take a pay hit to get better working conditions if I'm not going to get rewarded later down the line like you assume will happen.
There is always an implicit tradeoff of working in company A with one tuple of salary / working environment, or working in company B with another tuple of salary / working environment. If we eliminate tuples that are dominated by others then we end with a curve of equipreferencial points, representing such tradeoffs.
If such tradeoffs are offered by different companies, there is no reason why they should not be also offered as by one company.
"Remember back to high school math class. Now imagine 8 hours of high school math class back-to-back-to-back. Now add in "pop-quizes" (e.g. production issues) sometimes multiple times a day."
I think that once someone can relate to what day-to-day software engineering is like, they can empathize with why "walls" and "doors" are nice.