People’s obsession with Slack and chat apps (applies to open offices too) seems overly done in Silicon Valley. Career switching from the military to development has been interesting in this regards. The military uses chat, but only when needed—realtime communication is critical and people are distributed geographically. On duty on the carrier keeping track of planes and other issues happening “in country”, chat is great. On the ground in Iraq scheduling airborne assets and keeping track of SF missions, chat is great. But for most things, chat is more of a distraction and slows down the real work.
I understand chat if you are working in operations, but for pure development, it feels like a college dorm, more about the community and less about the work, where people still confuse time at the office with getting your work done. After a decade of 70+ hour work weeks, you really appreciate what is work and what isn’t. Building camaraderie is important, but chat probably isn’t the best way. And putting everyones communications in a single channel isn’t helping communication, it’s increasing distraction.
In my ideal dev team, you have chat, use it for one-on-one communication, and for anything that is time sensitive which goes into group channels. Anything that not everyone needs to know or isn’t time sensitive should go in an email or a message or a shoulder tap. Cluttering a few group channels with a ton of individual conversations, meaningless alerts, and random banter is downright juvenile and a detriment to productivity. I recognizing that promoting free flowing communication is important, but dorm room chat isn’t the solution.
> People’s obsession with Slack and chat apps (applies to open offices too) seems overly done in Silicon Valley.
> After a decade of 70+ hour work weeks, you really appreciate what is work and what isn’t.
> I recognizing that promoting free flowing communication is important, but dorm room chat isn’t the solution.
"Dorm room chat" is a great way of thinking and talking about it, and I think you can extend the metaphor really far. A lot of young people working in Silicon Valley and startups (not all, of course) see it partially as a lifestyle choice of sorts.
Where it gets interesting and messy is– if your employees value that sort of thing, should you encourage it? Discourage it? I guess it depends on what you're trying to achieve.
I agree completely with the philosophy that communications in general should be lean and tidy. It's interesting to observe that different people have different attitudes about this, and that every organization comes to its own sort of compromise.
> On duty on the carrier keeping track of planes and other issues happening “in country”, chat is great.
Very similar observations from 10 years ago and now. And in the hospital, chat has no role. It's curious watching the Dev community obsess about how many things could theoretically do without getting off their butts, and then swear the latest standing desks.
The collaboration excuse is a red herring. I'm sure some of the managers believe it, but it's not really the point.
Office space is leased by the square foot. $/sf/ year. Building a new building is usually budgeted in terms of $/sf. (Aside from some base unavoidable costs, it more or less scales linearly with floor area in terms of square feet.)
The fewer square feet you need, the less your lease is, which immediately and directly decreases your overhead and increases your profit margin.
Open offices fit more people in fewer square feet. There's a specific dollar amount a manager can point to and say "I'm saving that much money".
Maybe the productivity gains of private offices would more than offset the extra cost of needing more space, maybe not. It's not as easily quantifiable. Keep in mind that office leases are often signed for 10+ years and if you're building your own building you're stuck with it for even longer. Plus once you sign a lease you have to pay to build out the office. This can easily cost several million dollars in itself.
If you're a manager making this mutli-million dollar decision, are you going to go with the guaranteed overhead savings of minimizing square footage or are you going to risk millions of dollars on a vague promise of maybe increased productivity? Unless your company has the money to spare I don't see any rational person opting for the latter, much as I would rather have a private office myself.
As I understand it though, open office is a trend. Your mechanism doesn't explain why there was ever a time without open offices. So what's changed to make office managers more penny-wise, or have there always been "open collaborative offices" just under a different name?
I don't think people are saying everyone should have private offices, which I agree is not feasible from an economic standpoint. Nor have I seen any studies saying that private offices justify the costs of themselves in increased productivity gained.
But there are ways that offices can be designed to be less likely to disrupt your work. Maybe grouping a few people in a private office instead? Or set "quiet hours"? Have designated "quiet rooms" where talking is prohibited? I feel like there are a lot of things that can be experimented with or try that are not a huge expenditure.
And I know putting a dollar amount saved on productivity with innovations in workplace design are harder to show, but I think it's a widely known fact that interruptions affect knowledge work and peoples ability to focus. So if someone who should be working on focus work is frequently being interrupted every 30 minutes by noises and people asking questions and it takes that person 15 minutes to get back into the groove each time, it's easy to see just how many wasted hours there will be.
I used to think that cost was the main incentive too.
But it doesn't explain why so many A-List silicon valley companies use open plan. Google, Facebook, LinkedIn, etc., all have intense competition for engineers, and they spend a lot on perks for their employees. And devote lots of footage to amenities such as gyms, game rooms, etc. It doesn't seem that cost would be the driving factor for these companies.
why do people fall in this trap over and over again?
if you can do basic math you understand that the cost of people is >>>> cost of the space. For small startups, one person can cost you more, per month, than all your rent.
open space can work great sometimes depending on the type of work done, but I've yet to see a place where people are not simply tired of the noise and the constant interruptions.
I've worked in open offices, they made it impossible for me to program, so I spent all day "collaborating" and doing scutwork until everyone went home at 5, then I'd stay til 9 to get the stuff I needed to get done done.
I'd never do that again, if a company puts me in a position where I can't work during work hours I simply won't, will raise concerns and if not resolved leave.
Life is too short to do 70 hour weeks when you are only getting 30 hours work done.
Indeed. One gig I was at consisted of spending 9-5 doing nothing but meetings/groupwork/collaborating/email and dealing with bosses who kept asking why the coding wasn't getting done and behind schedule in one breath while dragging me into a 3hr+ meeting the next. Getting work done meant getting in before the boss, coding as much as possible until the boss showed up, waiting for the boss to leave, then coding again for another 2-3 hours to keep up. It does get ridiculous.
The problem is we've turned asynchronous communication and decision making into a near-real time short loop without considering the cognitive costs. Switching from my editor to email, parsing the 30 new messages, identifying the content to reply to and fabricating the correct reply costs the mental software model being used to guide the coding and it takes time to get it back into place. Switch in/out all day and your only bit of peace comes when the email writers go home.
We have a hybrid space. Cubes surrounded by white boards. Probably 20feet of whiteboard / associate and a few associates with offices with doors and 10 feet of whiteboard. Doors are seldom closed. We use chat for both my team and with another similar team across campus. We are very aware of the penalty of context switching so the spaces tend to be quiet except when there are operational issues. We use a scrum lite approach (daily stand ups but very lightweight).
We agree with the comments here that chat should be used for operational or time and sequence sensitive task coordination. We call quick meetings to collaborate on issues and opportunities right after standup and they tend to last about 10 to 20 minutes at most. These don't occur very often (1 or 2 a week). All of this works very well.
We have tuned the space a lot over the years. We also have a small table in the space so we can do demos (big monitor on the wall) or have meetings when we want and not be tied to formal meeting space availability. We maybe have 2 to 3 meetings a week in the space and usually they involve everyone (there are 3 of us).
My last gig was at a bank and I had no desk. I had a file cabinet and a shared admin (across maybe 80 people). Was expected to find space to work. I carried my office on my back. At times it was great. I could hide from my associates when I needed to get work done. I had 8 people working for me. But when we needed to collaborate or meet we could hardly ever find quiet space to talk through operational issues. We were never allowed to use chat so when operational problems arose people had to assume I was looking at my email.
Net net was I hated this approach. I love what I have now.
That resonates a lot with me. A few years back my manager in a big-co with a lofty motto would not allow me to work remotely for a week or two to stay closer to my dying parent. As she pointed out, "a lot of innovation happens at the watercooler".
Innovation == Talking about what we did that weekend? :)
This was cruel and there is no excuse why in 2016 this arrangement couldn't work for a week or two.
As a remote member of a mostly onsite team there is some validity that you miss out on the SOCIAL aspects of work (grabbing lunch, hanging out after work from time to time, etc) but the "innovation" on our team is just fine.
You should've quit on the spot. "Jobs come and go, but I only have one/two parents who deserve better, etc. Bye." She would've relented and let you go. I've called this bluff several times in my career and I've never lost.
But of course, you don't want to work for someone like that in the first place, if you can help it.
I work in an open office, and would consider myself a knowledge worker. The easiest way to manage it is to work from home, however I also positioned my desk in a corner facing the window (so my back to the room) I also by default put wvery meeting invite to tenative and ubless its absolutely necessary i dont show up.
All that has put me at odds with management to a degree. However, im given all the hard problems and churn out solutions. Based on projects completed, im doing the work of 4 employees at the same level. I should also add that I still answer 5 - 20 questions from coworkers a day, and am the go-to guy to fix many issues (so I dont completely shut myself off). This has put me in a position where my manager has hinted that im going to get promoted if I keep it up.
Of course the company I work for has "non-management management" i.e. you can become a VP without having to manage people. This lets the anti-social hard-core workers still move up in rank and pay.
You will not get promoted. As long as you're super valuable in your current role, they only have incentive to keep you put. I got strung along with that bullshit for years at my first job before I learned this lesson.
It's not a total loss though. If they do find you valuable, play hard on your salary.
Don't believe this line of crap. Employment is a popularity contest. You need to be popular to climb the ladder. Literally turning your back on the room and refusing to go to meetings makes you unpopular. It's true that your popularity among your superiors is the most important popularity, but popularity among your peers is also important; if you can't get the respect of your subordinates you won't be an effective manager. People will work to undercut you. In 99% of cases, you won't get their respect if they think you don't like them.
In aggregate, peoples' behaviors are very predictable. Before every action, ask if this kind of thing would endear a toddler to you or not. If it would, it's a good thing to do. If it wouldn't, it's going to be an uphill battle and you'll have to do many toddler-approved things to undo the damage.
This is why you should bring donuts for the office every week; toddlers love getting edible treats, and so do grownups. Make sure everyone knows you're the one who brought them, without looking like you're making sure everyone knows that you're the one who brought them. This will make people like you more.
Turn your desk around and start going to all their meetings. Say supportive and encouraging things. Make irrelevant broad, philosophical statements, literally stuff you'd find on motivation posters. You'll get popular fast.
If you care about your career, you must realize that your actual job is only ~20% of what you're paid for. You're really paid to be likeable and charming, someone that everyone wants to keep around the office. Do whatever you have to do to keep your code limping along and then put the rest of your energy into becoming popular. You'll be promoted and tasked to hire a version of yourself that hasn't learned these lessons yet, and the cycle will continue.
It seems to me that depending on where you're working, being the person that always saves the day and that does the "work of 4 employees at the same level" is a sure way to not get promoted.
Sadly, there really is a lot of collaboration cargo culting these days.
But it's interesting because in the large majority of fields the best outcomes really are in some ways collaborative efforts. The lone genius is actually extremely rare.
I'm inclined to say the key to effective collaboration is that it has to be driven entirely by the collaborators, and it is only actively happening for a very small percentage of the time spent on a project.
A cargo culting middle manager might think she's helping by setting up hours of meetings with different people to get their "perspective". Those people probably can add value. But the best way to do it is to let the project lead go to them with questions or asking for feedback when they think it makes sense.
Management can help by fostering a culture where people are willing and eager to talk to their colleagues when they need help, or to get new ideas. To some extent communication tools can help, but there are other much bigger factors: Don't make people feel like they'll be penalized in performance reviews for not knowing things or needing help. Cultivate a sense of shared ownership of projects and outcomes. Give people maxuimum autonomy over their schedules.
I really want to figure out which plan is better. I honestly feel that people with bad experiences with open plan get on here and gripe, and it makes open plan sound horrible. But there are bad experiences with offices too. Like rarely interacting with colleagues outside of scheduled meetings. And there's a natural bias towards one's own productivity vs helping the team.
Maybe it's just because open plan is more the norm nowadays, now we hear all the downsides of that. But it's worth considering it was adopted in response to the downsides of offices and cubes (shudder), not some shortsighted irrational plot to cut costs at the expense of productivity.
One thing I know: If anyone says rarely interacting "sounds great", they shouldn't be trusted. Yes you will absolutely maximize your coding focus. Yes that is an invalid argument. Team collaboration is important too. I suspect that to make either plan work you've got to pair it with a solid culture that shores up its weaknesses: minimize distractions in open plan, fight isolation in offices.
My straw man ideal is an open plan that is relatively quiet, but let's you look at each other and leverage body language and enhanced awareness to gague when it's a good moment to interrupt. That's the only way that a whole class of "quick questions" and "crazy ideas" get fielded that would otherwise not cross the interrupt bar of a door-knock or even an IM-ping. And that stuff can be super valuable esp. to a startup team. And it actually can get those handled quicker!
But I suspect it has to be paired with a culture of minimizing extended subcritical conversations at your desk and taking them over to separate meeting areas that are less disruptive.
At my new job, I've really started pushing back against meetings that seem superfluous. Often times I'll find myself saying the same exact thing at 3 different meetings, and it almost never affects anyone else in the room.
What form does this pushing take? How is it working? I ask because I've failed at this before; managers have great power to defend their habits and processes.
Is it possible that there's no one-size-fits-all solution to this problem? I get the feeling that if cubicles were still in fashion people would be complaining just as loudly about them as others are complaining about open offices. I think a good balance between the two extremes combined with allowing people to work from home on days they need to focus is the best answer we have to this problem today.
We can keep writing these articles and reading them and knowing we are right and get nowhere.
1) Say I'm an engineer with a manager who doesn't get it. How can I empathetically (that is, without spontaneously developing a "political performance problem" on my next review) help my manager improve?
2) Say I'm that manager, and I want to improve but meetings and spreadsheets are all I know. What can I do?
... Most people getting work done are typically on a Maker schedule. People scheduling and benefiting from the meetings are typically on a Manager schedule. This article can help those on a Manager schedule understand why large blocks of work time should be reserved for those doing the work. Managers need to strive to make meetings short, sweet and at the beginning or end of a day and concentrated on a specific day. Those are very specific items that can improve the Maker's productivity. The Manager should be protecting Makers from meetings, not subjecting them to meetings and this article makes that clear.
> 2) Say I'm that manager, and I want to improve but meetings and spreadsheets are all I know. What can I do?
Rethink what it means to be a manager. Your job is to eliminate barriers for your team and protect them from distractions. You shouldn't be one of the distractions.
Well one option is that if meetings are shorter and we waste less time on performance-evaluating BS, then the surplus of manager's time will result in less demand for managers.
There was a post today about Stanley Kubrick, which got me thinking: programming is like making a movie in many ways. So you can tell your boss: "if I were a film director and you knew I'm busy at the film set making a feature film for you, with some Hollywood stars, would you be calling me out for your shitty meetings as often as you do otherwise?"
Or: "If you were Stanley Kubrick's boss, would you interrupt him as often?"
1) Phrase it as a help-me-help-you arrangement. You're getting gauged on how much stuff your team is getting done. Let me help you improve your performance metrics by reducing my meeting load and letting me focus more on coding. Our output metrics will look a lot better and you'll come out smelling like a rose. Sr. Management will love you and your results. Sell them on what gets them ahead (and it's also in your interest).
2) I liken a good manager to an old school point guard in basketball or a good coach. Your job isn't to score but to setup your teammates for success and create a culture of excellence they can thrive in on your team.
Be ruthless about your meeting schedule and filtering input that actually needs to make it to your team. You should be the first POC for anyone needing your team or people on your team to do things to filter it for appropriateness rather than letting your engineers handle things. I'm not talking about their buddy in IT asking them a question, but people like Joe in Accounting who thinks it'd be slick if they could add FeatureX to the software because their boss thinks it's cool and it'd be sweet if it could be in this release. "Ya Know ;)". As a manager your first job is to firewall your people and help handle task requests, analyze them, and prioritize them.
Also as a manager you help handle tasking and assignments. While you're building out your current sprint--and you're keeping up with what's happening so you can keep bosses apprised and off your team's back--you're also preparing for the next one, organizing requirements, incorporating feedback, etc. so when the team is ready for the next batch of tasking, it's already there and ready for them to process.
WRT meetings, make your meetings targeted, on point, and ruthless enforce them. You're the boss. If someone's running their mouth without reason and taking up time, politely cut them off and get back on track. Have a clear objective for the meeting and stick with it. Once you've hit it, you're done. End the meeting everyone get back t work. Schedule your meetings for either mornings or late afternoons (preferably mornings) to give people plenty of time to get "deep work" done without having to worry about only having an hour to code before a meeting. If other people are trying to draw your team into meetings, figure out whether it's something you can handle and if so, you go. Let your team do their job. Again, you're the firewall.
This also means getting to know your team personally and being able to read how they're doing. Does someone look really run down and out of it after a long week? Send them home early. Someone having a hard time getting things done on time? Sit down and talk with them, not adversarially, ask them what's going on, and what you can do to help. Politely point out what you've noticed and why it concerns you. Maybe their estimating wrong or their skill level isn't up to the work. You can help fix that. You're their coach. Know your team and help them succeed.
My read on pushing collaboration is that it's a hedge against doing the wrong thing. In startups (say, <100 people and in grow-or-die mode), I think you have to focus on producing revenue on a 3-9 month timeframe. If you have a bunch of people working on their own, a lot of them will do things that are valuable, but won't show value to the company for 2-5 years. That not really OK, unless everybody agrees it's OK.
So, everybody collaborates constantly, so when you do something like run a pen-test before you have a customer with a CISO, there's agreement that it's OK to do it early. Otherwise, you might end up with a bunch of interesting work that doesn't have any coherence, and with only 25 people, that's not enough to sustain a company.
Of course, if everybody is working on the right things, and you have manager drive-bys via Slack/email/daily standups 3 times a day, that's a separate issue. It's the fear of working on the wrong thing that's motivating that pattern.
I'm shocked that SV ingenuity has ended up at such a horrific solution for office environments. The "open" plan hardly delivers any collaboration either, just a continuous stream of distractions or creepy silence with a room full of headphones. Surely we can do better?
There's a cultural aspect to collaboration, and an office layout aspect.
Culturally, collaboration can be great. Getting someone from sales involved in a product meeting could be revolutionary, or getting a product person on a sales call etc. etc.
It's possible to design an open floor plan with plenty of private spaces, but most companies don't do this because the people making the decisions are not introverted enough or their jobs don't involve achieving and maintaining "flow state".
Ideas like pair programming have created the misunderstanding that programming should always be done in a social way. There are many times when a problem demands focused thinking, and the ideal office space offers places to do that.
Fittingly, I'm having trouble focusing on reading this article because the team next to mine is having a conversation. They aren't speaking overly loudly, but there's nothing between me and them so I can hear everything.
I work in what might as well be an open plan: I share an office with three other people, and the desks are arranged so none of us have our backs to the wall (at least I get a really nice view of a highway out of the deal, though). The whole company is set up like this, with every 3-4 employees grouped into large offices.
It could be worse because I get along with my officemates, but it's easily the worst office arrangement I've been part of. The vulnerability of not having my back to the wall seriously gets to me, and it can get really, really annoying when multiple conversations are taking place at the same time. One of my officemates wears headphones 90% of the time specifically because of this.
Before this job, I've worked in cubes, I've had my own office, and I've shared an office with one other.
Of all those, I preferred sharing an office with one other, with desks arranged so our backs are to the wall. I had some degree of privacy, but I'm not totally isolated from other people.
I agree with the article and many of the comments here. I'd add that there's another thing: a one-size-fits-all design. Open plan offices are very good for workers for whom collaboration is constant; I've seen sales, marketing, some ops, and a few other departments who would constantly interact no matter what kind of office layout they had. For work that requires lots of concentration and little collaboration, though, that layout is counterproductive. Companies should start thinking about role needs and the space required to support maximum productivity, rather than just having a single layout because it's easier/cheaper. How about a ring of private offices for those that need it surrounding a shared collaboration space? I worked at one place like with that layout and it worked very well.
The Harvard Business Review reports "an overlap of only 50% between 'the top collaborative contributors in any organisation and those individuals deemed to be the top performers.'"
That's a pretty high percentage, if you ask me. I wonder what other traits you could identify in the set of top performers that would be equally or more prevalent. The article seems to equivocate more collaboration with more distraction. These two are perhaps related when policy aims to encourage collaboration through shared office space, Slack, and meetings which interrupt deep work. But let's be clear; collaboration is a good thing and we shouldn't act like it's wrong to encourage it, just that perhaps there are better, less distracting ways to do it.
It's one thing to put multiple people on the same task or question and have a focused, active collaboration. This would include things like brainstorming and pair programming.
It's another thing to give people their own tasks but put them in an environment of ambient chatter and real time notifications. This would include things like open floor plans and Slack. (I'm not arguing that these things are bad, just that they're not genuine collaboration.)
It's pretty easy to make an office look and feel like a collaborative space. It's quite a bit harder to find a few tasks that would benefit from real collaboration and invest double or triple the "hourly rate" for that one thing. Then again, maybe it's totally worth the cost?
First of all, forcing collaboration onto people by making them sit together isn't proven to be effective. There is no evidence to show it's in the company's best interests to have an open office environment. Doing it just because Facebook does is cargo cult thinking.
Second of all, it's also against the employees' best interests and pursuit of happiness. Employees should ideally want to choose the sort of office they have, and also be able to work from home whenever they want. Having happy employees leads to the highest productivity and employee retention.
I've worked in many different environments, and I dislike open office the most.
I have already tuned off my email client, only checking it in the morning and after lunch. Contemplating moving me workstation (computer) into my lab where I can get away from the open (short cubical wall) floor plan.
Of course, I am reading HN now, but only because it's taking quite a while to generate Gerbers for a PCB.
[+] [-] poof131|10 years ago|reply
I understand chat if you are working in operations, but for pure development, it feels like a college dorm, more about the community and less about the work, where people still confuse time at the office with getting your work done. After a decade of 70+ hour work weeks, you really appreciate what is work and what isn’t. Building camaraderie is important, but chat probably isn’t the best way. And putting everyones communications in a single channel isn’t helping communication, it’s increasing distraction.
In my ideal dev team, you have chat, use it for one-on-one communication, and for anything that is time sensitive which goes into group channels. Anything that not everyone needs to know or isn’t time sensitive should go in an email or a message or a shoulder tap. Cluttering a few group channels with a ton of individual conversations, meaningless alerts, and random banter is downright juvenile and a detriment to productivity. I recognizing that promoting free flowing communication is important, but dorm room chat isn’t the solution.
[+] [-] visakanv|10 years ago|reply
> After a decade of 70+ hour work weeks, you really appreciate what is work and what isn’t.
> I recognizing that promoting free flowing communication is important, but dorm room chat isn’t the solution.
"Dorm room chat" is a great way of thinking and talking about it, and I think you can extend the metaphor really far. A lot of young people working in Silicon Valley and startups (not all, of course) see it partially as a lifestyle choice of sorts.
Where it gets interesting and messy is– if your employees value that sort of thing, should you encourage it? Discourage it? I guess it depends on what you're trying to achieve.
I agree completely with the philosophy that communications in general should be lean and tidy. It's interesting to observe that different people have different attitudes about this, and that every organization comes to its own sort of compromise.
[+] [-] niels_olson|10 years ago|reply
Very similar observations from 10 years ago and now. And in the hospital, chat has no role. It's curious watching the Dev community obsess about how many things could theoretically do without getting off their butts, and then swear the latest standing desks.
[+] [-] crdoconnor|10 years ago|reply
>use it for one-on-one communication
^^
This often ends up hiving off critical information.
My ideal 'team chat' software would let you, by default, have one on one communications which get indexed and are searchable by everyone.
[+] [-] imgabe|10 years ago|reply
Office space is leased by the square foot. $/sf/ year. Building a new building is usually budgeted in terms of $/sf. (Aside from some base unavoidable costs, it more or less scales linearly with floor area in terms of square feet.)
The fewer square feet you need, the less your lease is, which immediately and directly decreases your overhead and increases your profit margin.
Open offices fit more people in fewer square feet. There's a specific dollar amount a manager can point to and say "I'm saving that much money".
Maybe the productivity gains of private offices would more than offset the extra cost of needing more space, maybe not. It's not as easily quantifiable. Keep in mind that office leases are often signed for 10+ years and if you're building your own building you're stuck with it for even longer. Plus once you sign a lease you have to pay to build out the office. This can easily cost several million dollars in itself.
If you're a manager making this mutli-million dollar decision, are you going to go with the guaranteed overhead savings of minimizing square footage or are you going to risk millions of dollars on a vague promise of maybe increased productivity? Unless your company has the money to spare I don't see any rational person opting for the latter, much as I would rather have a private office myself.
[+] [-] tdaltonc|10 years ago|reply
[+] [-] eeeeeeeeeeeee|10 years ago|reply
But there are ways that offices can be designed to be less likely to disrupt your work. Maybe grouping a few people in a private office instead? Or set "quiet hours"? Have designated "quiet rooms" where talking is prohibited? I feel like there are a lot of things that can be experimented with or try that are not a huge expenditure.
And I know putting a dollar amount saved on productivity with innovations in workplace design are harder to show, but I think it's a widely known fact that interruptions affect knowledge work and peoples ability to focus. So if someone who should be working on focus work is frequently being interrupted every 30 minutes by noises and people asking questions and it takes that person 15 minutes to get back into the groove each time, it's easy to see just how many wasted hours there will be.
[+] [-] ph0rque|10 years ago|reply
[+] [-] Terr_|10 years ago|reply
"Never attribute to innovation that which is adequately-explained by penny-pinching."
[+] [-] batguano|10 years ago|reply
But it doesn't explain why so many A-List silicon valley companies use open plan. Google, Facebook, LinkedIn, etc., all have intense competition for engineers, and they spend a lot on perks for their employees. And devote lots of footage to amenities such as gyms, game rooms, etc. It doesn't seem that cost would be the driving factor for these companies.
[+] [-] mirceal|10 years ago|reply
why do people fall in this trap over and over again? if you can do basic math you understand that the cost of people is >>>> cost of the space. For small startups, one person can cost you more, per month, than all your rent.
open space can work great sometimes depending on the type of work done, but I've yet to see a place where people are not simply tired of the noise and the constant interruptions.
[+] [-] analog31|10 years ago|reply
[+] [-] noir_lord|10 years ago|reply
I'd never do that again, if a company puts me in a position where I can't work during work hours I simply won't, will raise concerns and if not resolved leave.
Life is too short to do 70 hour weeks when you are only getting 30 hours work done.
[+] [-] floppydisk|10 years ago|reply
The problem is we've turned asynchronous communication and decision making into a near-real time short loop without considering the cognitive costs. Switching from my editor to email, parsing the 30 new messages, identifying the content to reply to and fabricating the correct reply costs the mental software model being used to guide the coding and it takes time to get it back into place. Switch in/out all day and your only bit of peace comes when the email writers go home.
[+] [-] epalmer|10 years ago|reply
We agree with the comments here that chat should be used for operational or time and sequence sensitive task coordination. We call quick meetings to collaborate on issues and opportunities right after standup and they tend to last about 10 to 20 minutes at most. These don't occur very often (1 or 2 a week). All of this works very well.
We have tuned the space a lot over the years. We also have a small table in the space so we can do demos (big monitor on the wall) or have meetings when we want and not be tied to formal meeting space availability. We maybe have 2 to 3 meetings a week in the space and usually they involve everyone (there are 3 of us).
My last gig was at a bank and I had no desk. I had a file cabinet and a shared admin (across maybe 80 people). Was expected to find space to work. I carried my office on my back. At times it was great. I could hide from my associates when I needed to get work done. I had 8 people working for me. But when we needed to collaborate or meet we could hardly ever find quiet space to talk through operational issues. We were never allowed to use chat so when operational problems arose people had to assume I was looking at my email.
Net net was I hated this approach. I love what I have now.
[+] [-] fishnchips|10 years ago|reply
[+] [-] brooklyndavs|10 years ago|reply
This was cruel and there is no excuse why in 2016 this arrangement couldn't work for a week or two.
As a remote member of a mostly onsite team there is some validity that you miss out on the SOCIAL aspects of work (grabbing lunch, hanging out after work from time to time, etc) but the "innovation" on our team is just fine.
[+] [-] verisimilidude|10 years ago|reply
But of course, you don't want to work for someone like that in the first place, if you can help it.
[+] [-] hire_charts|10 years ago|reply
[+] [-] lettergram|10 years ago|reply
All that has put me at odds with management to a degree. However, im given all the hard problems and churn out solutions. Based on projects completed, im doing the work of 4 employees at the same level. I should also add that I still answer 5 - 20 questions from coworkers a day, and am the go-to guy to fix many issues (so I dont completely shut myself off). This has put me in a position where my manager has hinted that im going to get promoted if I keep it up.
Of course the company I work for has "non-management management" i.e. you can become a VP without having to manage people. This lets the anti-social hard-core workers still move up in rank and pay.
[+] [-] verisimilidude|10 years ago|reply
It's not a total loss though. If they do find you valuable, play hard on your salary.
[+] [-] cookiecaper|10 years ago|reply
In aggregate, peoples' behaviors are very predictable. Before every action, ask if this kind of thing would endear a toddler to you or not. If it would, it's a good thing to do. If it wouldn't, it's going to be an uphill battle and you'll have to do many toddler-approved things to undo the damage.
This is why you should bring donuts for the office every week; toddlers love getting edible treats, and so do grownups. Make sure everyone knows you're the one who brought them, without looking like you're making sure everyone knows that you're the one who brought them. This will make people like you more.
Turn your desk around and start going to all their meetings. Say supportive and encouraging things. Make irrelevant broad, philosophical statements, literally stuff you'd find on motivation posters. You'll get popular fast.
If you care about your career, you must realize that your actual job is only ~20% of what you're paid for. You're really paid to be likeable and charming, someone that everyone wants to keep around the office. Do whatever you have to do to keep your code limping along and then put the rest of your energy into becoming popular. You'll be promoted and tasked to hire a version of yourself that hasn't learned these lessons yet, and the cycle will continue.
[+] [-] slig|10 years ago|reply
[+] [-] moron4hire|10 years ago|reply
You're not going to get promoted. That's just a bullshit game they play to pressure you into putting in more hours.
If it ain't in writing, it don't exist.
[+] [-] skewart|10 years ago|reply
But it's interesting because in the large majority of fields the best outcomes really are in some ways collaborative efforts. The lone genius is actually extremely rare.
I'm inclined to say the key to effective collaboration is that it has to be driven entirely by the collaborators, and it is only actively happening for a very small percentage of the time spent on a project.
A cargo culting middle manager might think she's helping by setting up hours of meetings with different people to get their "perspective". Those people probably can add value. But the best way to do it is to let the project lead go to them with questions or asking for feedback when they think it makes sense.
Management can help by fostering a culture where people are willing and eager to talk to their colleagues when they need help, or to get new ideas. To some extent communication tools can help, but there are other much bigger factors: Don't make people feel like they'll be penalized in performance reviews for not knowing things or needing help. Cultivate a sense of shared ownership of projects and outcomes. Give people maxuimum autonomy over their schedules.
[+] [-] abalone|10 years ago|reply
Maybe it's just because open plan is more the norm nowadays, now we hear all the downsides of that. But it's worth considering it was adopted in response to the downsides of offices and cubes (shudder), not some shortsighted irrational plot to cut costs at the expense of productivity.
One thing I know: If anyone says rarely interacting "sounds great", they shouldn't be trusted. Yes you will absolutely maximize your coding focus. Yes that is an invalid argument. Team collaboration is important too. I suspect that to make either plan work you've got to pair it with a solid culture that shores up its weaknesses: minimize distractions in open plan, fight isolation in offices.
My straw man ideal is an open plan that is relatively quiet, but let's you look at each other and leverage body language and enhanced awareness to gague when it's a good moment to interrupt. That's the only way that a whole class of "quick questions" and "crazy ideas" get fielded that would otherwise not cross the interrupt bar of a door-knock or even an IM-ping. And that stuff can be super valuable esp. to a startup team. And it actually can get those handled quicker!
But I suspect it has to be paired with a culture of minimizing extended subcritical conversations at your desk and taking them over to separate meeting areas that are less disruptive.
[+] [-] zymhan|10 years ago|reply
[+] [-] hashkb|10 years ago|reply
[+] [-] marknutter|10 years ago|reply
[+] [-] hashkb|10 years ago|reply
1) Say I'm an engineer with a manager who doesn't get it. How can I empathetically (that is, without spontaneously developing a "political performance problem" on my next review) help my manager improve?
2) Say I'm that manager, and I want to improve but meetings and spreadsheets are all I know. What can I do?
[+] [-] dhj|10 years ago|reply
http://paulgraham.com/makersschedule.html
... Most people getting work done are typically on a Maker schedule. People scheduling and benefiting from the meetings are typically on a Manager schedule. This article can help those on a Manager schedule understand why large blocks of work time should be reserved for those doing the work. Managers need to strive to make meetings short, sweet and at the beginning or end of a day and concentrated on a specific day. Those are very specific items that can improve the Maker's productivity. The Manager should be protecting Makers from meetings, not subjecting them to meetings and this article makes that clear.
[+] [-] patrickmay|10 years ago|reply
Rethink what it means to be a manager. Your job is to eliminate barriers for your team and protect them from distractions. You shouldn't be one of the distractions.
[+] [-] zymhan|10 years ago|reply
Heaven forfend
[+] [-] mojuba|10 years ago|reply
Or: "If you were Stanley Kubrick's boss, would you interrupt him as often?"
[+] [-] ryandrake|10 years ago|reply
[+] [-] floppydisk|10 years ago|reply
2) I liken a good manager to an old school point guard in basketball or a good coach. Your job isn't to score but to setup your teammates for success and create a culture of excellence they can thrive in on your team.
Be ruthless about your meeting schedule and filtering input that actually needs to make it to your team. You should be the first POC for anyone needing your team or people on your team to do things to filter it for appropriateness rather than letting your engineers handle things. I'm not talking about their buddy in IT asking them a question, but people like Joe in Accounting who thinks it'd be slick if they could add FeatureX to the software because their boss thinks it's cool and it'd be sweet if it could be in this release. "Ya Know ;)". As a manager your first job is to firewall your people and help handle task requests, analyze them, and prioritize them.
Also as a manager you help handle tasking and assignments. While you're building out your current sprint--and you're keeping up with what's happening so you can keep bosses apprised and off your team's back--you're also preparing for the next one, organizing requirements, incorporating feedback, etc. so when the team is ready for the next batch of tasking, it's already there and ready for them to process.
WRT meetings, make your meetings targeted, on point, and ruthless enforce them. You're the boss. If someone's running their mouth without reason and taking up time, politely cut them off and get back on track. Have a clear objective for the meeting and stick with it. Once you've hit it, you're done. End the meeting everyone get back t work. Schedule your meetings for either mornings or late afternoons (preferably mornings) to give people plenty of time to get "deep work" done without having to worry about only having an hour to code before a meeting. If other people are trying to draw your team into meetings, figure out whether it's something you can handle and if so, you go. Let your team do their job. Again, you're the firewall.
This also means getting to know your team personally and being able to read how they're doing. Does someone look really run down and out of it after a long week? Send them home early. Someone having a hard time getting things done on time? Sit down and talk with them, not adversarially, ask them what's going on, and what you can do to help. Politely point out what you've noticed and why it concerns you. Maybe their estimating wrong or their skill level isn't up to the work. You can help fix that. You're their coach. Know your team and help them succeed.
[+] [-] trjordan|10 years ago|reply
So, everybody collaborates constantly, so when you do something like run a pen-test before you have a customer with a CISO, there's agreement that it's OK to do it early. Otherwise, you might end up with a bunch of interesting work that doesn't have any coherence, and with only 25 people, that's not enough to sustain a company.
Of course, if everybody is working on the right things, and you have manager drive-bys via Slack/email/daily standups 3 times a day, that's a separate issue. It's the fear of working on the wrong thing that's motivating that pattern.
[+] [-] pbreit|10 years ago|reply
[+] [-] grandalf|10 years ago|reply
Culturally, collaboration can be great. Getting someone from sales involved in a product meeting could be revolutionary, or getting a product person on a sales call etc. etc.
It's possible to design an open floor plan with plenty of private spaces, but most companies don't do this because the people making the decisions are not introverted enough or their jobs don't involve achieving and maintaining "flow state".
Ideas like pair programming have created the misunderstanding that programming should always be done in a social way. There are many times when a problem demands focused thinking, and the ideal office space offers places to do that.
[+] [-] mabbo|10 years ago|reply
I have a pair of these to help: http://www.amazon.com/dp/B0000X6L78/
[+] [-] kentt|10 years ago|reply
[+] [-] amyjess|10 years ago|reply
It could be worse because I get along with my officemates, but it's easily the worst office arrangement I've been part of. The vulnerability of not having my back to the wall seriously gets to me, and it can get really, really annoying when multiple conversations are taking place at the same time. One of my officemates wears headphones 90% of the time specifically because of this.
Before this job, I've worked in cubes, I've had my own office, and I've shared an office with one other.
Of all those, I preferred sharing an office with one other, with desks arranged so our backs are to the wall. I had some degree of privacy, but I'm not totally isolated from other people.
[+] [-] dasmoth|10 years ago|reply
[+] [-] Futurebot|10 years ago|reply
[+] [-] jm_l|10 years ago|reply
That's a pretty high percentage, if you ask me. I wonder what other traits you could identify in the set of top performers that would be equally or more prevalent. The article seems to equivocate more collaboration with more distraction. These two are perhaps related when policy aims to encourage collaboration through shared office space, Slack, and meetings which interrupt deep work. But let's be clear; collaboration is a good thing and we shouldn't act like it's wrong to encourage it, just that perhaps there are better, less distracting ways to do it.
[+] [-] zachrose|10 years ago|reply
It's another thing to give people their own tasks but put them in an environment of ambient chatter and real time notifications. This would include things like open floor plans and Slack. (I'm not arguing that these things are bad, just that they're not genuine collaboration.)
It's pretty easy to make an office look and feel like a collaborative space. It's quite a bit harder to find a few tasks that would benefit from real collaboration and invest double or triple the "hourly rate" for that one thing. Then again, maybe it's totally worth the cost?
[+] [-] xyzzy4|10 years ago|reply
Second of all, it's also against the employees' best interests and pursuit of happiness. Employees should ideally want to choose the sort of office they have, and also be able to work from home whenever they want. Having happy employees leads to the highest productivity and employee retention.
I've worked in many different environments, and I dislike open office the most.
[+] [-] madengr|10 years ago|reply
http://calnewport.com/books/deep-work/
I have already tuned off my email client, only checking it in the morning and after lunch. Contemplating moving me workstation (computer) into my lab where I can get away from the open (short cubical wall) floor plan.
Of course, I am reading HN now, but only because it's taking quite a while to generate Gerbers for a PCB.
[+] [-] kentt|10 years ago|reply