Do the dishes in the office sink get washed? If so, by whom? And how clean is the kitchen area?
This may seem like a joke, but the answer says a lot about an organization. For example, last time I was at the Wikimedia Foundation offices, this sign was above the sink:
After the first night of the 2008 SciFoo conference at Google, @gnat Torkington's twitter stream contained a gem: "At SciFoo opening session. 300 people standing around networking. Meanwhile, Larry Page is quietly unpacking chairs at the back of the room." (from memory, not exact)
Long ago, my previous company ditched paper cups in favor of ceramic cups for coffee. You should have seen the disruption in everyone's workflow.
Since you have to actually wash the cup after you drink from it, there were long queues before the kitchen sinks.
People started reducing the number of cups of coffee they drink :) and instead started dozing off (not joking)
People started congregating around the pantry or kitchen because if you wander off with your cup around the office, you still have to come back to kitchen to clean it up.
I used to wonder the whole 'go green' thing at the time because anyway the paper cups are going to be recycled :)
Why not just use a cup more than once during the day without washing? I go through at least 7 cups of coffee a day in a single mug; only washing that's done is just before the first of the day.
Does this assume that developers should wash the dishes? If so, why is that a good idea? I would assume that developer time is too expensive for that...
Please shoot me if I ever start thinking my brain is too valuable to waste on basic hygiene.
There is a dishwasher. It's five seconds to rinse your dish and put it away.
At the WMF, there is someone whose job it is to make sure the kitchens are clean and the dishwasher gets run, but that's only effective because most people exercise common sense.
I was genuinely interested in the answers, so thanks for the interesting replies. I don't have a dishwasher so I didn't know it could be so quick. I agree spending 10 or 20 seconds or a minute on something that's useful for everybody is not a bad idea.
I'm just not sure how that scales for tasks that require more time. If there is some task that requires 20 minutes, and the company is big enough, do the answers change or do they not? Maybe it is better to hire a dedicated person for the office to do that than to make everyone use 20 minutes of their time every day on such task?
More seriously, I think that allegedly financial calculation often ends up covering up for laziness, arrogance, and hierarchies of social class. Make sure the code works? That's QA's job. Make sure it runs well? Oh, that's ops; file a ticket. Wash my own dishes? Hey, office manager, get on that. Oh, and and honey, make me some coffee while you're at it.
I know some companies work that way, but I'd never want to work at one.
If you have a unique skill, it makes sense to pay you to mostly utilize that skill. If you're great at programming and terrible at system administration, why should you have to administer your servers, especially when there are people that are great at it?
Testing is a little different: most professional software engineers consider testing an intrinsic part of software engineering, and so would never say, "that's QA's job". (But even then, there are some especially hairy testing problems that would benefit from a specialist's touch.)
Anyway, I think you're conflating dysfunctional organizations with specialization. In your example, developers say "that's operations' job" because they are lazy and don't care about maintaining a sane production environment. In a functional organization, they'll say the same thing, but they'll mean, "that's operations' job because they will do a better job than me". It sounds the same, but it means something entirely different.
(The same goes for washing dishes. I could wash my own dishes, but I would use a lot of water. If we batch up all the dishes in a 3000-person company each day and wash them in an industrial dishwasher, it will take less aggregate time and use much less water and energy. So while washing your own dishes may be symbolic of teamwork, it's actually a dumb thing to do.)
I would love to see some evidence that complete and total specialization is the maximum possible way to achieve business goals.
I know in my work, I've found that doing a variety of things, including "grunt work" leads many times to great works of creativity and discoveries of new ways to be more efficient. If you have a Ph.D. sweeping floors, you can bet they're going to take a stab at making the entire "floor sweeping" problem disappear. Not so much for a professional janitor.
Putting your head down and focusing is great when you have a mountain of obvious work to do. But that's not always what the business needs.
The high-end design firm IDEO specifically looks for "T-shaped people", by which they mean people who are deep in one area but have a broad set of skills outside their specialty. They believe that creative work is essentially collaborative, and that you can't be a great collaborator without a good understanding of what people are up to and the ability to step in and take a swing at anything that comes up, expert or not.
Another reason to avoid specialization comes, I'm told, from queuing theory. Unless your workload is perfectly regular, specialization leads to bottlenecking and global underutilization.
I'd also consider how Toyota, the world's #1 car company, looks at this. They do an immense amount of crosstraining, and line workers are specifically discouraged from specializing too much. Much of their efficiency comes from bottom-up innovation, which you don't get if people are focused only on their one little piece of the problem. (For more on this in particular, Toyota Kata is a great book.)
I spend my day building a castle in my head. Sometimes I may go outside and pace around the parking lot, sorting through the rooms in that castle.
I drink 10-15 cups of whatever a day. The cups get tossed in the trash when empty, or left ..somewhere.. when I get an idea.
But stop me every time my cup is empty, make me break my train of thought and go fool with soap and water and a breakable cup? Likely the castle comes crashing down, and 40 minutes to rebuild it.
Might work for some, but definitely not for me. Wash your own cup; but leave me alone.
Specialization introduces opportunities for dysfunction, because it breaks a feedback loop. Dev -> QA is one of those. Google "throw it over the wall" to see how common the problem is. E.g.: http://kdc-blog.blogspot.com/2009/01/dont-throw-it-over-wall...
People are skilled economic resources, but they are also status-seeking monkeys that were mainly evolved for foraging, fucking, and fighting. If you don't keep that in mind when setting up an organization, there are risks. I'm asking how Valve handles them.
Regarding dishes, I'm not saying that one has to pick suboptimal solutions just for the sake of symbolism. Google, for example, has industrial dishwashers. But still, everybody picks up their trays, sorts the compost, recyclables, and landfill into proper bins, puts the utensils in the bins for used silverware, and then slots the tray and dishes into a conveyor that feeds into the kitchen. Could Google hire busboys? Sure. But they didn't, and I'd bet it's not an accident.
Not that I have the most experience here, but I've never felt my time was too expensive to wash dishes. Your ego may vary.
Maybe there's some marginal value in saving highly paid employees from doing menial work, but there's also a cost in terms of culture when you treat people like they're better than anyone else.
Whenever I did anything not programming related at the place where I worked I had a feeling that I'm getting money for free.
I was selling work of and inexperienced tea maker or dishwasher or cleaner or delivery man to my employer at hourly rate of a skilled programmer. One of the best deals I've ever done albeit short one.
By extension, developer time is too expensive to be wasted on long lunch breaks, socialization with co-workers, or toasting your bagel in the morning rather than eating it untoasted.
Basic hygiene/workspace maintenance is everyone's job. There are people who specialize, of course (the cleaning staff; if you have them, developers aren't scrubbing toilets and steam cleaning carpets, of course), but it takes literally 5 seconds to rinse a dish and put it in the dishwasher, the act of which improves the entire workspace and contributes to overall order in the environment. If you don't have a dishwasher, it might take upwards of 30 seconds to wash your plate, dry it, and put it away.
Developers are actually in a unique position where they can do their job while doing other things; much of a developer's job involves thinking through problems, and menial physical tasks are actually a pretty great way to get your brain engaged on the problems.
The two aren't totally separable. Status calculations are a fundamental human behavior, and social systems tend to move in directions of class and hierarchy. (For more on this, the books Chimpanzee Politics by DeWaal and Impro by Johnstone are readable and very compelling. Consider also that absolute salary differences matter less for happiness than relative salary differences, which is economically insane but very human.)
I strongly believe that a lot of corporate idiocy is driven by hidden status games. E.g., the way that managers often have status in proportion to the size of budget or staff, which gives an incentive away from maximum efficiency.
Unconventional approaches to management push against this. E.g., the notion of servant leadership, or Toyota's explicit focus on management's responsibility to support the people doing the real work. Valve presumably has something similar.
It takes literally 10 seconds to wash (not just rinse, but entirely wash) a bowl and a spoon. Nothing has to stay in the sink - just wash things right when you're done using them.
The mindless job of dishwashing is just another way to think about a problem away from a computer, plus it's useful.
This works if your coworkers are happy to leave you alone to think. I often pace the halls at work while I think, but the moment I step into the breakroom I get greeted with "Hi Michael how are you so nice to see you how was your weekend" which makes the exercise kind of pointless.
Not that I have anything necessarily against small talk, but the time spent in the breakroom is unproductive time.
What if one of your fellow developers is looking for a clean dish and glass and there are none to be had? All he sees is a sink full of dirty shit that nobody has touched.
So you may have saved some time dropping your stuff in the sink, but now you have another developer who's spending the identical amount of time cleaning up...but in addition he's pissed off as well. Sounds like a positive outcome to me.
Yes, because only male developers might not want to spend their valuable work hours scrubbing dishes because the company is too cheap to have a working dishwasher.
Who said nobody could have an automatic dishwasher? But even if you have one, at some places dishes pile up in the sink, and at some places they don't.
If it isn't a K-cup, it probably isn't fresh enough. Also, the place I just left had awful coffee. It was so bad that once when a coworker cleaned the grime that built up in the coffee pot for the past few years of use, it made the coffee taste worse. When grime is the only thing making the coffee better, you need to get better coffee.
No, people will still complain. At Google, we have Stumptown, Intelligentsia, and Gorilla coffee. People complain that the coffee is not fresh enough (unused coffee is only thrown away every 14 days), that the blends purchased are not acceptable (blended coffee is evil, donchaknow), and that the full-auto single-serving coffee machines tend to retain dust from the past user and hence mix a tiny bit of Blend A with Blend B.
(I've personally complained about how some of the espresso tampers are not heavy enough, and how the Barista station closes at 3:55 instead of 4:00. If you give a moose a muffin...)
i don't work at valve, but i work at an organisation with a very similar management structure (there is none) and payment scheme (autonomous self organizing teams are paid by clients and distribute payments as negotiated by the group). we have virtually zero overhead and minimal residual or recurring income so we are not identical to valve, but their employee handbook could very easily be ours
we have a number of 'bad apples', but they generally find themselves ostracized very quickly. as we have no downside income guarantees (we are organized as a group of independent contractors), people very quickly fall out of the organisation
i assume valve accomplish similar by minimizing payments to individuals who act in bad faith. as they are probably primarily financially motivated and valve is likely a great entry on a cv, i'm sure they don't stick around long
That is incredibly interesting. If you're uncomfortable saying who you work for in public, please contact me. I would love to understand more about your organization and put it on my list of potential companies to hire.
In the dark days when I worked for financial traders, once a year the company would set aside a pool for bonuses and then try to figure out how to share it out based on contribution. If you went around and asked people what percentage they deserved and then added them up, the total would never be less than 300%.
I don't think most people are quite as bad about this as financial traders, but I imagine no group's numbers add up to 100% exactly. What's the dynamic like at Valve, and how do you make sure people feel fairly treated?
So you are saying the requested bonuses were 300% of the pool set by management? There are two possibilities. One, the expectations could be too high. Second, the pool could be too small. Your assuming only the first is correct?
In this particular case, the bonus pool was basically the profit of the firm, so the pool was not variable. But my point was more about the perception of one's own contribution versus everybody else.
Another way to look at it: a friend's marriage counselor said that if each member of the couple thinks they're doing 50% the work, that leaves another 50% of the work undone.
Unlike the traditional management structure that Bill Gore had experienced at DuPont, he proposed a flat, lattice-like organizational structure where everyone shares the same title of “associate.” There are neither chains of command nor predetermined channels of communication. Leaders replace the idea of “bosses.” Associates choose to follow leaders rather than have bosses assigned to them. Associate contribution reviews are based on a peer-level rating system.
I always assumed this sort of org could only function with relatively small companies (e.g. Github, maybe even Valve) but Gore has 9000 employees - pretty impressive.
Gore divides the company into business units, each in their own building with their own parking lot and staff. In fact, they know when to divide a growing unit - when people start parking on the grass (the parking lot gets full).
So no not 9000 people; 90 units of 100 people ( or something like that)
1. How do you handle the logical conclusion of letting your workers do whatever they want, they don't do anything? This may not apply to all employees, but it surely applies to at least one, and I wanted to know specifically how you might handle that situation.
2. What sort of system do you have to address problems with the unconditional hierarchy? For example, if an employee disagrees with Gabe, who wins and how is that handled?
Real management isn't telling people what to do. It's helping people find the intersection between what they want to do and what customers need, and then supporting them.
That strikes me as especially easy at Valve, where the whole point is to create something that other people enjoy. If you can't find fun somewhere in the vast space of "create something fun," you're dead inside.
I understand the parent comments point as well as yours. I realize this won't be an issue for most of the employees, but bad apples sneak through everywhere, it's human nature, and I specifically wanted to understand how they might handle that.
Well, "creating something fun" is not an easy thing to do. When you get down to pure game design there can be wildly divergent viewpoints of what to do. Should we make this new game more interesting for hardcore players, or for casuals? Is increasing accessibility of the game worth the decrease in depth?
1. How do you handle jobs that suck, but must be done.
Like backups ... nobody really likes the job of 'backup guy'.
It's boring. Tedious. A backwater of a role with no meaningful career progression. Also - the software typically sucks, and is expensive. Also - the hardware typically sucks, and is super-expensive.
But someone must pay attention to backups because stuff happens and you might actually need to recover everything from tape to get the company online.
How does Valve get the tedious, boring, nit-picky, utterly necessary stuff done?
I would imagine no one at valve is an expert at managing backups for hundreds of machines and keeping an appropriate level of redundancy. Valve is definitely a company that is going to focus on where it can do better then others. They probably use many outside services to handle the things they don't want to have to worry about or specialize in. Other companies will handle it better because that's their focus, Gabe definitely gets that.
We experimented with out-sourcing a few years ago. It was a disaster: the out-sourcing company was 'fair-to-excellent' with standard products and services.
But we're a manufacturing company. The majority of our software is specific to our industry. Suddenly the vendor is dealing with dozens of apps they've never even heard of. And - whoops - some of the guys who dealt with that stuff are laid off.
The rest of us got calls from confused techs in Austin plaintively asking what in the heck was the UmptyFratz service and who can restart it?
Long way of saying that - I imagine - some of what Valve does is industry specific and hard to out-source.
When I worked at the office supply store across the street from Valve HQ I had a guy come in and order 8 laser printers or delivery and he told me he played the role of IT over at Valve. So I'm sure they actually have people who choose to focus on things like that.
I don't think this question should be downvoted. There's a lot of unsexy infrastructure stuff that just needs doing. If people can really do whatever they want, what happens to the stuff that nobody wants to do? Lightbulbs are a great example.
This is a common "issue" in design studios and the like. I've always been the kinda guy who enjoys using a lunch break to clean/fix up/replace the small things, because it's just what I'd do at home. Obviously there are places where people feel like they're dumb tasks, but I'd imagine that as the quality and mutual respect goes up (as in a place like Valve) there are more people likely to behave similarly.
I would like to imagine that. But you run up against the "neatest roommate" problem.
Suppose you share a house with a half-dozen people. You are all good people, so whenever something seems messy to you, you clean it up. Problem solved, right?
In my experience, definitely not. Because the person who's best at perceiving mess tends to clean things up before other people notice anything. That person feels like they're doing more of the work. Because they are. And they feel unappreciated, because people generally don't notice the messy-to-clean transition.
I'm a recovering sysadmin; I burnt out because I grew so very tired of caring about things that people never noticed until they were broken. What they said then wasn't, "Gosh, I really appreciate how well the printer has worked these many months." They said, "God damn it! The printer's broken just when I need it the most!" I'd love to know how stabby Valve's sysadmins and IT people feel.
If I can do whatever I want, I would change lightbulbs when broken. Why would I want to work under a burned-out light bulb. (Pedantic note: light bulbs are pretty much illegal now, so I assume you mean fluorescent tube or something.)
I think it's sad that so many commenters genuinely can't seem to fathom anything positive happening without a hierarchical / top-down / command-and-control approach. Especially when a few billion years of evolution has basically never selected that approach (something like a bee colony may seem similar, but is radically different... it's not like there is a manager bee for every 5-7 worker bees). Please don't vote me into oblivion for saying this.
How many hours a day does the average Valve person spend in meetings? How are they organized?
I've looked into some co-ops, including San Francisco's venerable (and very profitable) Rainbow Grocery. One of the biggest complaints is the amount and/or difficulty of meetings, but they see them as necessary to settle issues with sufficient buy-in from all stakeholders. How does Valve minimize that pain?
This is actually great, I had the same idea last week coming across that handbook, so hopefully someone at Valve can shed some light.
I was mostly curious about project management. It seemed like everyone could be very fluid going from one project to another, or even proposing one on the spot and going on to execute it. There has to some sort of enforcement for this though. Are there expectations, do the self-selected leaders lay out milestones or goals, what happens when those aren't met?
Valve doesn't necessarily seem to have a reputation for having too many product delays vs. always shipping on time. They definitely seem to have constant flow of different products getting out the door though.
Hopefully someone answers, otherwise I might just hitchhike across the lake into Bellevue and see if I can meet with someone there. Very curious to learn what makes their system actually work well.
In hierarchical organizations, people often look to authority figures to decide between competing alternatives. How does Valve avoid or deal with deadlock, forking, cliques, cabals, butthurt sulking, and other common group dysfunctions?
And they're all going to say, "We're just like Valve's flat organization, except here we do it a little differently because we have managers, and a CEO and a few VPs, and ..."
(A cow-orker of mine went to Valve last year; he's incredibly, insufferably happy).
And after that, you'll see the inevitable HN posts about how the "Valve" adoption at their company didn't work, which proves the whole idea was bankrupt to begin with, and probably a conspiracy to increase book sales, consulting rates, and the amount of underpants collected by the gnomes.
ROWE isn't similar. You still have people telling everyone what to do. Only now the deadlines and expectations will be ridiculous. "I don't care how many hours it takes you this week, just get it done! We only care about results here.". The valve method would imply that middle manager would not exist.
I couldn't say. But it's implemented at the company headquarters and they are experimenting with it in the stores. They haven't gone back on their commitment to it. It's also been implemented at The Gap:
A Rails consultancy (I used to work for) in Atlanta called Highgroove also implements ROWE, going so far as to incorporate it into the way they deliver to the customer. It was a great way to work and operate as a business.
Job postings on Valve site tend to be very specific about the person's duties. To me, it seems to be in contradiction with being able to choose how one contributes to the company.
The short version is that people who are good at things generally get that way because they have a strong ability to tell good work from bad. People who are bad at things can't tell the difference, so they a) have a hard time improving, and b) think their low-quality work is pretty swell.
Is your culture perhaps unusually frank? Alternatively, is it very supportive in a way that makes critique more comfortable? Might you have a formal (or informal?) mentorship program so that people get useful feedback?
Are there peer groups that meet around particular skill areas? E.g., do visual artists get together regularly to show recent work and discuss it?
I agree my summary was a bit breezy. But I think an essential component of their theory is the focus on domain knowledge and metacognition, not skill as it is normally used. Consider the title of their first paper: "Unskilled and Unaware of It: How Difficulties in Recognizing One's Own Incompetence Lead to Inflated Self-Assessments". In the conclusion, they write, "We propose that those with limited knowledge in a domain suffer a dual burden: Not only do they reach mistaken conclusions and make regrettable errors, but their incompetence robs them of the ability to realize it."
Competence isn't a single axis. Nobody's a genius at everything. E.g., great programming skill does not automatically confer high social skill, good business sense, appreciation of experience design, or ability to wrangle meetings.
If it will help everybody, please forget I mentioned Dunning-Kruger in specific. That's the inspiration for the question, not the heart of it.
This may seem like a joke, but the answer says a lot about an organization. For example, last time I was at the Wikimedia Foundation offices, this sign was above the sink:
http://commons.wikimedia.org/wiki/File:These_dishes_may_requ...