The Open-office Trap
m.newyorker.com
m.newyorker.com
Truly modern offices have hot desks AND are wired for teleconferencing - right in the open areas! That's right, if "open" is good for communication then a 60" screen on the wall with video cameras pointing at you is even better! You can share in the dull roar of collaboration with people on other continents ALL DAY! In the average open space you can fit at least two, maybe three such screens. When you have them installed, get the tradesmen to come during office hours. It costs way more if they come after hours and your employees should get the chance to see drills, saws and nailguns operated by professionals, they may learn something.
We all know there's nothing that stimulates productivity like the sound of productivity, right? So when you plan your next office, leave the kitchen open too - the scrape of chairs on the tiles, the sound of the dishwasher being unloaded, the happy PING of the microwave - not to mention the aromas of hot food - all these things will make employees feel right at home. Using only the most modern materials in the office - lots of glass and steel - will ensure the entire workforce shares the joy. You don't want to be damping the precious sound waves with carpet or plasterboard.
I wish I was joking.
I visited a place that was a giant room of ~100 people working. No one had their own computer, instead you just sat down and whatever computer was there and started working. The guy running this insanity said they re-image all the machines all time taking settings from one of the machines that has been in use. His theory was that over time everyones preferences would merge.
It was interesting to see and a little frightening. Right off the bat the noise level would have turned me off. I asked how hard it was to hire and he said interviews are held in the room with him or another senior person pair coding with the candidate.
The machines were pretty much stateless. When you booted a machine you could choose between Windows, Linux, or reset the whole machine. The third one was what you did when the machine acted funny, it would wipe the hard disk and reinstall both OS' from a network image. The whole process was completely automatic and was done in 20 minutes.
So it was real hot desking, and it worked great. You sat at any computer, it became your computer. Way before Google started to talk about stateless computers and their chromebook.
I don't know what's wrong with corporate enviroments, that Windows profiles take so long to roam. That seems to be true to all of them. Also, corporate IT has an extremely irrational aversion to simply installing the software on all the computers for once. Even when it's free software, or things that everybody uses. That also repeats everywhere.
Certain people refused to put their big files on the network share because they said it took to long to load and then complained that their roaming profiles took to long to load because they had those big files on their desktop.
This is where the ability to have automatically configured (puppet / chef / ansible / cfengine) clusters of servers for 'Nix hosts, or NFS-mounted /usr, so much more powerful.
For Windows, what's actually happening is that your user profile is getting copied to the system. Which is why logging off takes so long -- the profile is getting copied back to the server.
Do this on an underprovisioned and busy network, or worse, one on which work cycles are highly synchronized (e.g., students, in standard class blocks, over the course of a day), and where account profiles can grow without limit (at one point I had tools to ID and prune large profiles), and things go all to hell.
The Linux / Unix model actually can be quite useful, and it isn't too dissimilar from my own initial experience: console logins to the campus Unix network from dumb serial terminals (precisely zero local state).
Sun Microsystems did some work with this (in conjunction with their own hotdesking workplace experiments) as well.
The downside is when you're doing highly compute- or data-intensive work, in which case the amount of information transferred across even NFS links becomes problematic, and/or you need to provision some really beefy servers. At that point you likely want some sort of shared batch compute resource. Again, more easily accomplished under Linux/Unix than other platforms.
So my information may be somewhat dated. Still generally harder to do this under Windows than Linux.
Hell, you might as well leave some of the viruses on there. Haven't some viruses integrated with the human genome? We want the same concepts to apply to our IT and office layout. Perhaps one of those viruses is closing security vulnerabilities and keeping other viruses out.
First impression: no lobby, no receptionist. I had to interrupt someone at a "desk" to announce myself.
Second impression: the "desks" were in fact folding Costco tables, arranged in ranks of around 10 and about 6-8 files through a cavernous space, edge-to-edge.
Third impression: site had some exceptionally poor security practices (and is among the more notable password disclosure case histories), which constituted a considerable part of my own interview content. The expressed interest in changing procedures was near nil.
I phoned the recruiter as I walked out (early) telling her that this was my new reference point for the worst interview experience ever.
No, I wasn't in the least interested.
The sad part is that some people see things like that in TV shows and try to duplicate them in the real world, without realizing that devices that work to communicate something on television don't necessarily actually work for doing the thing that the TV people are supposedly doing.
That particular plan is far from perfect – they have often used the lack of isolation (e.g. people peeping on neighbors through ceiling-level windows) as a plot device – but I still like the actual looks of it, based on actual experience with the open-plan-induced drudgery discussed in the article.
I worked at a place that did big war rooms per project and had tons of empty "home" cubes where you went for an hour or two per day to do non-primary work. I think a fixed-time system in the groups would be best.
Except SCDP's creative lounge doesn't have doors, see the floor plan: http://www.jordanorlando.com/other/scdp/
Well, I'm sure the savings free up money for other, more valuable purposes.
Because I have experienced both extremes (including my own startup) and they're not ideal. Having your own office is nice, but sometimes you want to ad-hoc with a small group of other people, or maybe you just want a quick change of scenery, maybe go to a place where there is some background activity. You shouldn't have to go find a Starbucks for that.
For me, these laptop-only environments don't work very well either, because at some point I want to have my desktop with a decent multi-screen setup.
Lastly, there should always be the option of working from home when appropriate.
So, if I ever start a company again, I'd try to just keep all these options available and let people find the right balance that works for them individually.
> So when you plan your next office, leave the kitchen
> open too - the scrape of chairs on the tiles, the
> sound of the dishwasher being unloaded, the happy
> PING of the microwave - not to mention the aromas of
> hot food - all these things will make employees feel
> right at home.
I understand that some people in certain roles would do much better with no interruptions in a private office but it's not a one size fits all solution. If your work consists of something you can mostly do on your own that's certainly one case where private offices might be a win. If your work consists of lots of collaboration it seems less clear. At least to me.
I'll also note that at least for me I'm much happier around other people than isolated in a private office. I'm not saying I want to be an a 1960 bullpen but a room/area with 2 or 3 other coworkers I'm working on stuff with is my personal ideal.
A private office doesn't mean you have to work on your own, it just means you have the option to.
So, IFF your company has the money to devote a fair amount of real estate to each worker, yes, private offices are great.
Even with just 4 of us, it means there is a good chance someone will be asked something, and regularly the head comes down to discuss stuff with one of us (often me). I can deal with it for the standard bug fix type work, but trying to learn something new is really difficult with constant interruption.
Folks younger than a certain age miss out knowing the social etiquette relating to visitors chairs, such as stacking "stuff" or coats on your visitor chair is anti-social and very strongly frowned upon. Also it was socially acceptable, standard even, to simply walk into someone's cube and sit down in their (visitor) chair and wait till they could talk, especially if you were (currently) working together on something.
Also "back in the old days" we had intentionally unreservable conference rooms scattered about. So if you needed 6 people on the spur of the moment you simply walked in to the nearest empty room, confident no one reserved it because it was un reservable.
This is pretty much how things were in the early 90s or so. Honestly I don't think I could handle the constant interruption and infinite distraction of an open office, which makes me think about a startup or something; I just can't imagine working in a "modern" office environment. It just sounds horrifying to productivity, like a prison yard or a doctor waiting room or something. Creeeepy.
As a parent when I need to get stuff done at home, like if I'm working at home, I kick everyone out of the room and work in peace and quiet. I never, ever, invite seven neighbor kids over and hand out toy whistles as a gift and nerf toys and then sit in the center of the maelstrom and try to concentrate. Nobody who "works at home" sneaks into a day care center with their laptop to concentrate on their finest work. There's a time and place for that kind of stuff, and its not the workplace.
That said in addition to a mandatory visitors chair, white boards everywhere. Just fucking everywhere. Nothing seemed to help collaboration like being able to start water cooler level talk about a problem and with out even noticing transition to hashing it out on a whiteboard.
So the "old timers" get very strange questions like "what if you're in an office and need to talk to someone" where the younger people don't understand the architecture and the older people don't understand the younger people's lack of experience.
In 1995 the office I worked in had 4 desks (usually 3 empty because of flex time/remote work, meetings, collaboration, etc) and 4 chairs (again, usually 3 empty) and two big lunchroom sized folding tables and I believe six chairs crowded around the lunchroom tables. Admittedly mostly covered with these new fangled (at that time) "cisco multiprotocol router" things. So if someone walked into my office and wanted to talk, or even a moderately large team appeared at my door with pitchforks or whatever, I could and did occasionally host parties of ten plus people, like if there was an emergency and we had to config and ship out like an entire data center of routers at the same time. And yet most of the time my office was ALSO blissfully empty and silent so I could concentrate on studying release notes and design new configurations and test network designs using routers set up all over the conference tables. It was extremely flexible, and therefore extremely productive, unlike an open office.
People who don't have experience with offices have these bizarre notions about office etiquette and collaboration. They also easily fall into post hoc / confirmation bias fallacies, where they see some positive result of collaborating in an open plan office and imagine that they would be forced to live in a world without the benefit of that spontaneous collaboration in an environment with offices. But they ignore the fact that such things very much do occur with offices, and they also ignore the opportunity cost of open plan layouts. Of conversations which don't happen because the work space is deathly quiet due to everyone trying to work, and you can't just hop into a friend's office space and close the door so you don't bother other people, as just one of many examples.
90s office work reminded me a lot of dorm room life and culture. Dorm culture probably hasn't changed in the past couple decades, and having dorm rooms with doors hardly eliminates all social activity; if anything it encourages it and military basic training style barracks are not considered a requirement of socialization.
PS: Loved the dorm room/barracks analogy
"Oh, some guys from floor support were doing that, something about the new cubicles."
"WHAT!! Where the hell am I going to put my books! This isn't enough space, blah blah blah". He was a hot head and immediately flew off the handle, much to our pleasure. Well, a year or two later, we did actually go to smaller cubes, and then a few more years later to the "open" floor plan (un-ergonomic tables with no kind of adjustments). People way to close to each other. While it does come in handy when we are discussing problems and bringing new people up to speed, I definitely think the negatives outweigh the positives. Sometimes I just don't want to be around people!
A few people working together on a single task is not an open office. In a "real" open office you couldn't actually talk to the artist and designer beyond a few sentences because if you did you would disturb the other 10-20 people within earshot.
10-20? There are about 80 here.
As others have mentioned this is where the definition of open breaks down. Team environments are great. Many open office advocates are small companies who, because of their size, are really advocating team environments.
Now grow that same company to 100s of people and see if the open office still works. What I have seen is that even the biggest open office advocates start putting up walls (even if movable) and segmenting once again into a teams.
TLDR; Team != Open
However, on-the-desk quants and IT are genuinely respected for their sacrifice. Sure, they have to be highly available from 9:20 to 4:10, but the rest of the time is completely their own (usually for research; they tend to work 11-hour days not because anyone is making them, but because they love the work). They're paid very well for that intense 7 hours, and no one would subject them to the humiliation of "Agile" micromanagement. Quants have traders come by to ask about trading strategies at high levels, but they don't deal with PMs and "story points" and arguments about whether what they worked on over the past week was "in the backlog" or not. Quants have something much closer to an open allocation environment than most software engineers, and if they can't hack the stress (and not everyone can, especially for years on end) of being on the desk, they're usually moved to off-the-desk roles and the change isn't stigmatized.
(Some will come in and say that quants are second-class citizens in hedge funds and, if one looks at 95th-percentile compensation, that appears to be true. Quants hit a ceiling around $500k-1.5M, which is 1% of what the best traders make, but they're respected. On the other hand, non-managing programmers are third-class citizens in most Valley startups, and they aren't making any money either.)
Open-plan offices make sense on trading desks, just as "Agile" makes sense for a small consulting firm on the make-or-break project that'll take it into the big leagues (under which it's reasonable to ask the tech staff to truly "sprint", because delivering the project will result in a quantum leap in client relations, quality of work, and compensation). The problem is that management tends to take these processes that conditionally result in high performance (for a short while) and attempt to implement the costly, negative and painful aspects (e.g. open-plan offices and Scrum) while failing to deliver the upside (e.g. quant-level bonuses, genuine equity rather than 0.05% bullshit) that makes talented people willing to tolerate the pain.
I want to focus on one thing and do it as perfectly as possible before moving onto the next thing; I hate multitasking unless the situation calls for it; I have highly (and perhaps unnecessarily) attuned peripheral vision and hearing. The chatter was impossible to tune out -- especially design discussions taking place two workstations over. People who did 'soft' jobs kept suggesting that I listen to music to drown out the chatter without realizing that mine is not the sort of work that can be done while music is playing.
People walking about in the background behind my screen would distract me from the complicated problem displayed on it. It was difficult to have private chats with people -- the moment I asked to see someone in one of the glass cubicles, everyone (including the person called) would assume that person was in trouble.
My current office layout is very 'residential' -- smaller rooms, wooden furniture, doors and windows, no sterile white partitions and ceilings etc. Needless to say, I haven't had any problems concentrating.
Big thing I notice -- in the old office, it was pretty noisy, as it was all too easy for people to think they had privacy when they were in a high-walled cubical. The new office is a bit quieter, as people tend to talk in hushed voices more, since they can see everyone around them that would otherwise hear them.
That is 100 sq ft per 4 employees whereas giving each employee a tiny 6x6 cube (although I live in a 8x8) would be 144 sq ft per 4 employees, so if you go with offices you actually use less space than cubicles and if you are flex timing it the employees feel it is much more spacious.
I currently work in a team of 4 flex timers each with 8x8 cubes, in a sea of cubes, and I'd much rather have a nice 10x10 with a door for the 4 of us. We'd save a huge amount of real estate this way and get more work done.
You can of course fit far more than 4 employees in a 10x10 office if you ultra-flex time it. More than two decades ago I was baby sitting a large IBM financial mainframe network (SDLC, source route bridging, that kind of thing) and we had like ten people assigned to a 10x10 but due to 24x5 coverage, not to mention remote work, we never had more than two people in the office at a time.
I've never in my life so far, worked a job where everyone works precisely 9-5 M-F so I can see where non-flex time lifestyles might be more difficult. I would have a serious problem tolerating rush hour driving twice a day, five times a week, but thats another matter.
I worked in office space in NYC a few years ago that was $80/sqft. 6x6 cube is $3k/year. Using your assumption that an office is 5x bigger (not sure if that's accurate or not), you're looking at around $15k/year for an office.
So the delta is $12k/year, equivalent to 160-240 hours/year or 8-16% of a salary. IMO, with many developers you'll regain productivity far in excess of that by shelling out for an office.
The reasons companies eschew offices has more to do with taxes and lack of giving a hoot than cost.
I had the impression that in some of the bigger (and higher cost-of-living) tech hubs, just their salary could easily be that much.
Buildouts and other capital expenses depreciate over longer periods.
But if the cost of office space is an issue, much more efficient than an open plan office would be just letting people work from home.
Despite what I want to believe and what I have been told (all companies only want to hire multitaskers!), multitasking is neither productive nor really possible. You can really only do one project at a time.
Edit: forgot I can't edit it -.-
Remembered why it was relevant: Since tuberculoses is still a problem in India, they already assumed that he may have tuberculoses before the diagnosis (he was coughing heavily and had signs of an ulcer). But not only did they keep him working, he was still in a tightly packed office with all the others.
The whole story was very WTF-worthy and obviously, I don't work with them any longer.
And you need not listen to distracting music; prerecorded white noise works well for a lot of people.
Companies will get rid of open-plan offices when they stop being such gift-wrapped boons to middle management, and not a second before. If you don't like it, adjust or exercise your right to be fired. This is America.
Second, not all readers of HN are American. Nor do they subscribe to the same political/ethical/social beliefs. The article was about whether open offices are productive or not, not whether we should or should not have them.
If you need to talk to Dave, first of all Dave being a SPOF is a bug or management failure to be worked around. Secondly geolocating Dave is almost certainly non-productive. If you need to converse with Dave you call/text his phone to maximize productivity.
This is an insightful observation about the core values of open office vs alternatives... an open office simply isn't interested in productive activity. Its interested in non-productive activity, which has been proven over and over in scientific studies.
Its unpleasant, perhaps even inhumane for a large segment of the population, and unproductive for everyone but in a cause and effect relationship the bug is in the core values of the organization, that productivity doesn't matter. Its not that they don't know its non-productive, its that they don't care that its non-productive.
Also note you can't fix that core value epic fail by merely generating a new, technically superior architectural fad. You merely end up with an organization that's non-productive that has really nice private offices.
In that way its kind of like a HOA... I don't want the weirdos and dirtbags to be hidden, I want to be able to glance at a companies office and tell instantly if they have screwed up values, so in a weird way I want failing companies to implement open offices, so I can identify them and avoid them.
No the problem is that most management activity, including program management, project management and the like is fundamentally unproductive. People who don't know what they're doing (ie. middle managers, project managers and program managers) can't productively contribute to real efforts.
The problem is the people judging whether this is true or not are exactly those middle managers/project managers and program managers.
Once management infects a company, it always just keeps increasing in my experience. A European bank I worked at had 3 project/program managers or architects per developer. You were lucky to get 2 hours of actual work in on a day. They were actually working to expand, not the developer team, but the management team.
You can't simply take a closed-office environment and just knock the walls down, pass out headphones and cram everyone in one room.
You need to change the work culture so that open offices work. You have to adapt things like pair programming and remove things like phones and email. Put everyone on a certain project at the same table. Switch partners out.
You can't have people who sit by themselves, don't communicate and don't work together in an open office environment. You're nuts if you think that's going to work.
I've found when working from home I'm significantly more productive on my own than two of us peer programming in the office.
I'm surprised that no hybrid solution to this problem exists, or at least one that's taken hold. The best solution we currently have seems to be giving everyone a private office and augmenting it with really good team collaboration software. In most cases however, this does not eliminate the need for direct human collaboration, which is partly why open office plans exist in the first place.
It seems you could accomplish a hybrid approach by somehow giving each employee the ability to toggle the privacy of their workspace on or off. How you would accomplish that remains an interesting question. Some ideas that come to mind:
a) Have two physical workspaces for the employee: one private, one open.
b) Some sort of The Jetsons-esque enclosure around the workspace that can be toggled. Might use technologies such as electronic smart glass (for privacy) and/or active noise cancellation.
c) Have two floors and lifts under each desk. Want privacy? Just ascend or descend into it. It could even support hot desking on both floors, provided the proper visual cues and safety features existed. The cool part about this approach is it could in theory make moving your desk much easier.
d) In the future it could just be a matter of telling your standard ocular and cochlear implants to filter your peripheral vision and hearing accordingly.
Disclaimer: Some of these are admittedly crazy.
http://javatroopers.com/Peopleware.html#Chapter_13
Edit: see also the discussion at
http://c2.com/cgi/wiki?PeoplewareOnSharedSpaces
His first rule is that there is no "one-size-fits-all" solution :) but there are patterns :)
He does identify the need for all three of
- a communal space
- a semiprivate/team space ("war rooms", etc)
- a private space
Also:
"Each team needs identifiable public and semiprivate space and each individual needs protected private space. "
But that's not true...I had about a year in the "dungeon" when I was moved to my newspaper's online multimedia team, in a windowless basement...by then I already had the habit of surfing the Internet, but I did manage to learn enough PHP and MySQL to build a crime-mapping site and do other data projects, and even do things like get Drupal working (which ended up not being very useful, but still, my first web framework).
I threatened to quit if I didn't get moved out the main room with sales (fully expecting to actually have to quit, I was already half out the door).
Boss begrudgingly gave me an office, 3 months later he said it was one of the best idea's he ever had (he was a boss after all).
These days I work for myself (small consulting supported startup) so I have an office at work and an office at home.
http://midcenturymodernist.com/wp-content/uploads/2009/08/Ma...
But it just seems that some ideas, because they are economical interesting, stick so much, that management people come up every 6 month with an other idea or excuse, why the dead horse must be ridden again (and not buried).
If work is boring, routine or extremely exhausting open spaces are great cause' you can talk to each other and lift each others spirits.
If work is any type of creative, where you need to concentrate, it's better to have your own space.
And of course this immediately leads to the situation that the workers are out to exploit the company merely to rebel against this situation. Or at least, that's the attitude it brings out in me.
I strongly advocate for private or team-member-shared offices at any chance I get. It's just that much better.
Separate office with a closed door = productivity boon.
http://fixyt.com/watch?v=QfMvqkrQkYQ
https://news.ycombinator.com/item?id=7494705
One of the best lectures I've seen on the topic.
[There's] a very long and well-established literature in psychology that getting groups of people together is no way to come up with ideas. Creativity is not a team sport. What you're looking for is somebody's individual, intellectual trunk to make new connections and come up with something new.
Let's imagine billions of neurons in my head communicating with stuff they've been talking about all their lives together: there's a high probability that occasionally they'll come up with something new. Let's now think of the line of communication that you and I have got between each other, which is impoverished, because we have to try and translate complex ideas into language, and how many times do you find you've got a good idea, it's almost in symbolic thought inside your head, and you really can't articulate it to someone. And when you do, they get the wrong idea, because really, language can't encapsulate it until it's fully formed. There's no good evidence that I know of that these brainstorming sessions will come up with a solution or a new idea.
What they might do is improve a little bit of team spirit, or show some of the people in the group, "well, if that's the best they can come up with then I'm doing OK". The idea that you can marshal creativity is an error. I'd go a little bit further than this: if there is somebody who's spending 80 hours a week running a creative team, I'd stop them right there and tell them "you don't run a creative team, you allow a creative team to run". That would be the first thing I would jump on. People ask themselves "how do I make a team be creative?" You don't. You allow a team to be creative.
(Emphasis in original.)
More: http://redd.it/21qgiv
I rarely get sick, but then again we're seven people in one office and only one have kids ( kids makes a difference ). Mentally it's different story. I can't begin to count the number of days where I get home stressed out due to the noise of my co-workers and people going in an out of the room to talk to one of my colleagues.
Maybe I'm just special, but for me to completely focus and do my best work, I need an almost eerie silence. That's doesn't mix well with co-workers talking on the phone or a radio playing.
Open offices have been known to be a bad idea since Peopleware, but they are just so cheap to build.
Open-space was major reason (together with new born children) why I quit corporate world and started my own business two years ago.
In other words, I'm completely convinced that open offices (and cubicle farms, where I work) are harmful. To me, the more pressing question is: why are they still so common?
To quote a very cynical and jaded friend of mine, "It increases shareholder value".
Yes, studies show that open plan offices, especially badly-designed open plan offices, hurt knowledge worker productivity. The problem is, knowledge worker productivity is not expressed as a nice, neat number on a balance sheet. Facilities cost is. Thus, when a business gets into the mode of increasing profits by decreasing costs, all they see with respect to this topic is that office space expense. Cue the management bullshit stream about "collaboration", switch to open plans, and watch that number drop. The productivity drop tends to lag, however. Since many executive suites can't see forward past the next earnings report, or backwards past the previous, they see a nice increase in margins for a quarter or two, followed by a drop in revenue, schedule slips, or increasing costs in other areas caused by the open plan switch. They then move on to fixing this "new" problem, not realizing they caused it with their previous "fix".
Once a story has had significant attention, reposts are treated as duplicates for about a year.
I didn't find it as distracting as I thought - but I am someone who can tune out everything when watching TV.
I miss being able to break wind somewhat discreetly though.
"What’s more, Evans and Johnson discovered that people in noisy environments made fewer ergonomic adjustments than they would in private, causing increased physical strain."
I've spent my entire career in Japan, and have never not worked in an open office. To my knowledge, non-open offices basically don't exist in Japan.
So is every Japanese company just leaving productivity lying on the floor by insisting on open offices? I wonder if there have been studies on switching to closed floor plans in Japan.
(Of course traditionally the cultural emphasis has been on consensus, not productivity, so there hasn't been a lot of impetus to experiment. Things are changing, however.)
The actions people take depend on these things. The results of those actions typically do not.
> So is every Japanese company just leaving productivity lying on the floor by insisting on open offices?
Yes.
I worked for about 8 years in Japan. Hell yes they leave productivity lying on the floor, and not just with the open offices.
This small table with a wool shell looks nice, and the idea could (with heavy modification) be used to create long benches of private working space where noise and visual distraction is reduced and people are not boxed into cubicles.
http://www.designboom.com/weblog/cat/8/view/13088/gamfratesi...
A private office for me would totally kill any social interaction I have with anyone. A semi-open office keeps it private on the team level.
The best office I ever worked in was a series of long corridors with many closed offices and two developers to each. It had large meeting rooms at regular intervals, a lunchroom with table football and that was about it.
Perhaps a better working environment would allow your day to be spent more efficiently and leave you with spare time and energy to get an actual social life, instead of requiring your job to double as your social life and end up performing both functions badly?
But in the Bay Area circa 2014 what arrangements are being tried out and seeing advantages? Low cubes, high cubes, fully enclosed offices, bull pens, mini bull pens, periodic re-arrangement, unassigned?
As a designer working under an architect and having just completed an open office building for a 150+ employee financial firm[1] - i feel i have a pretty good handle on this subject. There a a couple key anecdotal design criteria that i'd like to address in relation to open offices that the report does not address.
Natural Light - Access to natural light
Artificial light - using the correct lighting for the task with the right output measured in lumens for the particular task.
Ventilation - Natural and sufficient HVAC Acoustics - Are proper acoustic absorbent materials being used.
Planning - has the space been thought out in a thorough way - is there a meaningful program to which the open office functions in both arrangement, flow and activities.
Psychology - Has there been effort to educate the staff about the new space and general systems in place to govern how it functions.
All of these points above can be easily planned for by hiring and adhering to the advice of design professionals like architects, electrical engineers, hydraulic engineers, acoustic consultants etc. Does this happen? In my experience - the answer is generally - no. Building offices and fitting thee out is an expensive exercise and time and time again i see clients willing to cut corners and forgo professional advice at the sake of saving a few thousand dollars. It may be that the ROI in terms of employee productivity could be significantly diminished due to a insignificant saving during the build phase.
I would agree in some respects that there are limitation to open office layouts - but that its due more to the ill-conceived notion that achieving an open office work environment is as easy throwing some workstation and humans into a cavernous space and expecting it to just work.
Moreover that is why indeed planners are moving away from open office and employing the newer philosophy of ABW (activity based working)[2].
[1]http://marklawlerarchitects.com.au/commercial/hunter-street-...
[2]http://www.jll.com.au/australia/en-au/Documents/jll-au-activ...
At some point, when an organization loses the purpose for its existence and selecting leadership comes down to office politics alone, the selection process becomes attrition. Since excellence no longer matters (there isn't real work for a person to excel at) the leaders are selected by subjecting people to artificial stresses and seeing who cracks (either a full-scale nervous breakdown, or just a passive loss of interest) first. Those were the weak, the less dedicated, etc.
Open-plan offices make that attrition faster, and provide more insight into who is next to crack up and entertain the crowd with an open-plan-induced panic attack.
That's also why this toxic micromanagement, in the name of "Agile" or "Scrum", that programmers are subjected to will probably never go away. When there's no good way to pick leaders (because the work isn't challenging or interesting) the stress of being watched, hour by hour, is a powerful attritive tool.
That's not to say there aren't good people in their 20s. There are quite a few of them. But they want to learn from people who are better than they are. That's the trait of A players: they want to learn from the A+ players. If you don't have anyone over 30, then the good under-30 people you get will quickly leave.
At least the ones who are actually good, as opposed to the ones who think they are $DEITY's gift to engineering.