1. 42

  2. 18

    Nice to see all these points in one place. I like that he discussed productivity as something that is often said to be important but not actually prioritized. Business culture selectively invokes productivity when it justifies existing preconceptions, and ignores it otherwise. Prime example: studies showing open offices hindering productivity. Knee-jerk retort: “But, collaboration!” The implication being the incidental ‘collaboration’ that occurs manages to offset the negative productivity. Quite a remarkable claim!

    Programmers are not immune to this sort of thing, either, but that’s another discussion.

    1. 8

      The implication being the incidental ‘collaboration’ that occurs manages to offset the negative productivity. Quite a remarkable claim!

      What I don’t understand is why it has to be 100% one or the other. Have quiet, individual working places and open office space for when you need collaboration.

      1. 9

        Also, I’ll unpack this a bit more. There is some resentment among techies about open offices, particularly since the culture skewed to be quite introverted before tech sold its soul. It can feel like extroverts imposing their will on everyone to have only open offices. Obviously both types of people can function within these environments, but it should be a question of whether they thrive.

        1. 6

          The offices where I work right now are really good that way; I brought my daughter (4 yo) to work today, and we’re just chilling in a room filled with bean bags. Earlier she was drawing a bit in the open office space; when she got noisier we isolated ourselves a bit, because that is an option. I’m visibly not doing time-crunchy “deliver this 5 seconds ago” kind of work right now, I’ll work a bit more from my home office this evening to make sure that I give them a fair share of my time against the money they pay me, but that’s fully on me.

          Anyway, yeah, we have enough space that it allows us to host yoga sessions within the walls of the office once a week, to which quite a few devs participate. We mostly suck (where “we” is mostly “I”), but that’s fine, it’s still healthy and fun (to me, that last one is a surprise). They issue headphones to everyone to mitigate ambient sound when there’s noise and you can’t/won’t isolate yourself. Hands down, best offices I’ve worked in.

        2. 3

          It doesn’t. But best practices rarely have an element of subtlety to them ;)

        3. 7

          Some of my days I work in a building from 1935 that is completely divided into offices that fit 2 people luxuriously to 6 people cosily. (It was probably originally a dormitory – the rooms are all the same size.) Those rooms are so much nicer to work in than an open-plan office.

          Working as a team of 3-4 people in one private office room is a joy, I didn’t know what I was missing until I got to experience it.

          1. 3

            Prime example: studies showing open offices hindering productivity. Knee-jerk retort: “But, collaboration!” The implication being the incidental ‘collaboration’ that occurs manages to offset the negative productivity.

            He covers this in the tendency of management to favor control over profits.

            Open-plan offices suck and management knows this, but they appear productive and they offer a sense of control.

            Middle management likes the intimidation factor. Executives of startups are managing up into VCs and recognize the optical advantage of the open-plan. (Look at all my “resources”; they practically live here.) The non-startups who’ve adopted open-plan offices are huffing jenkem because it’s what the cool kids do.

            1. 1

              We have an open office plan that I and many other dislike but the reason I was given was not “collaboration” but flexibility: they have to move people around frequently in response to product changes and having real walls makes it hard to do that. Doesn’t make it any better of course.

              1. 3

                One bit of circumstantial evidence that this is the “real” reason is that it’s the one almost invariably cited in the commercial real-estate business for the trend. They don’t usually bother trying to pitch their open-plan inventory with “collaboration” or “creativity” as the selling points. The two things they emphasize are flexibility due to it being easier/cheaper/faster to reconfigure the space, and lower space costs due to increased density: typically average square foot per worker is lower in open-plan offices than in traditional separate offices with doors.

            2. 15

              Hero syndrome is the biggest one I see, and it’s a very widespread issue.

              Literally in our society the reason we value Firefighters more than a person who walks around telling people not to burn candles in their house.

              We value doctors more than the sanitation workers who extend our lives by years by making sure we aren’t living in our own feces.

              We just care alot more about our pound of cure than we do about the ounce of prevention. Especially if the cure is exciting.

              Nobody gets to give conference talks about how they picked the right tool for the job. It’s much more beneficial for your career to launch a messaging platform on rails and spend the next several years fighting the inevitable fires.

              I’m literally watching history repeat itself with several “exciting startups” right now. They are getting tons of mileage and interest that they never would have received had they tested their backups or learned a little bit about data storage. Nobody ever wanted to watch a livestream of me setting up a cron job to test backups.

              The really sad/unsettling thing is that as you become a better engineer you have to seriously ask yourself: “How fireproof should I make this thing? If I spend another day on it, I get nothing. If I spend 0 days on it now and a week firefighting when I deploy to production I get to write a blog post and get thanked by the CEO for saving the company”. I still believe it’s important to do a good job, but it’s also important to go out of your way to make sure people understand the value your robust solutions bring to the table. When shit hits the fan and the foresight you had 1 year ago to degrade gracefully pays off in 0 downtime, you need to gently remind people.

              1. 3

                I came here to say exactly this. Out of all of those, Hero Syndrome seems like the hardest habit to kick. It is very tempting as a manager to praise the work of a team that battled through a crisis. It is almost a gut reaction. As a manager, I try to counteract this by recognizing and elevating the everyday, but I still don’t know how to do that and still acknowledge the important work that happens when people rise to a crisis. Any thoughts?

              2. 5

                The startup I work has reasonable hours (10am to 6pm), and no one cares if you’re 10-30 minutes late. We value actual output vs being seen working. We sometimes have a problem with “hero syndrome” but it’s rare. We’ve had some people that willingly put in long hours but we quickly realized they were not any more productive.

                1. 4

                  Measuring hours worked instead of productivity and impact is probably a decent red flag for leadership that is losing its bearings and trying to grasp at something in response. Luckily many smaller orgs simply don’t have that sort of organizational scar tissue built up yet.

                  The only downside to results based environments is they can push you to work harder to justify yourself. However, this is a temperament thing as well, and better than the alternative.

                  1. 5

                    The only downside to results based environments is they can push you to work harder to justify yourself.

                    It’s also really sensitive to how “results” are measured. Can discourage people from doing anything that they don’t perceive to “count”, and to over-emphasize things that are highly visible (closing bug reports, increasing number of commits, being unnecessarily public about all contributions so they get noticed, etc.). Especially true of course in cases where management is openly using quantitative metrics, but even when that isn’t the case, people who think their productivity is being judged tend to worry about what counts and doesn’t-count.

                  2. 1

                    Nice! I work at a startup too, and yeah, many of them are fine places to work. But probably not most.

                  3. 9

                    Great article.

                    And since the founders love what they’re building it never occurs to them that long hours might not be for everyone

                    Most founders don’t “love” what they’re building. They love the idea of being able to retire in 4 years and buy bigger yachts. They pretend to love the product because it draws attention from the inequality inherent to a system where founders get 20+ percent and employees get 0.05%.

                    If the founders fail but their investors like them, they get another startup or they get hired by the VC firm. If the employees fail (either individually or because the company goes belly-up) they don’t even get a severance.

                    Also, work is fun when you’re in charge. It only sucks for most people because most people end up as subordinate losers on 0.05%. So, it makes sense that founders would work hard.

                    Do they love their products, though? Not most of them. I know plenty of founders. They build the product that they can sell to investors and get acquired, not a product that they care about. They enjoy their work enough to do it, but “visionary” founders don’t last in Silicon Valley.

                    Jason Fried describes this as a Managerial Entitlement Complex: the idea that if someone is paying you a salary they are entitled to every minute of your time.

                    Also, it’s a shitty job market for most people. Yes, if you’re an under-30 programmer in the Bay Area with 5 years of Spark experience, you can find a job quite easily. Most people cannot. The new economy sucks, and employers take advantage. Middle managers are in the position of being exploited and exploiters (the middle piece of the human centipede). On one hand, they have control over other people. On the other, they know that if they lose their jobs, they’re going to be eating shit like everyone else. (They don’t have the social contacts and private welfare system that executives get.) They have more to lose.

                    Control beats profits

                    Ding-ding-ding-ding! Yes, absolutely. “Companies” don’t want things. “Companies” are pieces of paper. Executives are the people who drive companies. They’d like to be richer, but what they care about on a day-to-day basis is maintaining control. They only care about increasing profits when it helps them, and otherwise don’t care about company P&L, but they will shank you if you are a threat to their own positions or reputations.

                    Executives aren’t really capitalists and they certainly aren’t entrepreneurs. The Capitalist Party is about as capitalist as Soviet Russia’s or China’s Communist Party was committed to egalitarian communism– that is, not at all. They don’t care about shareholder value. They care about their own careers (private-sector social climbing) and nothing more.

                    When you’ve been set up to fail, your primary goal is to demonstrate that the inevitable failure was not your fault. The obvious and perhaps only way for you to do this is to have your team work long hours, a visible demonstration of commitment and effort.

                    Yep. And this is (sadly) rational. There’s a social glue that forms under shared suffering (even pointless suffering) and the person who gets blamed and fired when things go to hell is the 9-to-5er. (Or, the person who’s on vacation, which is why you should never go on vacation after a stressful launch.)

                    Middle managers get the advantage of being able to play white knight. “I’m sorry, but I can’t ask my team to do that.” It makes them sound reasonable, where as “I’m sorry, but I’m not willing to work 12 hours a day for this shitty, doomed project” makes someone sound like a so-called whiny Millennial. And then, if said manager does push his team to work 12-hour days and does solve the problem, he comes out as a hero.

                    The result is hero syndrome: the organization rewards those who save the day at the last minute, rather than work that prevents problems in the first place.

                    Managers are most grateful to people who save their asses. If corporate longevity is what you want, find (or make, if you’re playing black-hat) a crisis and be there to fix it.

                    Managers are most inclined to fire not low performers but people who cost them time. This is why high-performers are often fired. It doesn’t matter if your brilliance is going to improve profits by 10%. If your manager has to spend 5 hours per week dealing with the consequences of your suggestion, you’re going to get fired just as fast as the one who has to spend 5 hours per week dealing with complaints about an underperforming employee.

                    Managers are also incentivized toward reliable mediocrity and team-playerism. A star underling gets the limelight for herself. Managers get the downsides of employee performance but not the upsides. If you mentor an 8 into the 10, then she’s the brilliant one and you were just lucky enough to be there. Meanwhile, taking a team of 3s and making them perform as well as a team of 5s is an impressive managerial accomplishment.

                    1. [Comment from banned user removed]

                      1. 5

                        Truly, most human interactions are theater. Even if you go out and start you own thing you’ll be doing theater: sales, hiring, management, what-have-you.

                        More than that, you can’t really opt out of the theater. People judge the result of your performance as the primary means of learning about and valuing others. Most of the time.

                        The only way out of the theater is to hermit yourself or to tackle things with such obvious universal value that you can be said to be justifiable without anyone around. Some things transcend these stages. But we rely on the theater for things like payment, collaboration, survival. Even if you’re wealthy it’s hard to cut those dependencies so as to have the situation I just described.

                      2. 3

                        These are good, but in my experience (both as a developer and as a manager), the dominating reason has been bad estimation/planning.

                        There’s, essentially always, a disconnect between reality, what the engineers think they are supposed to do, what management thinks the engineers are supposed to do, what product thinks the engineers are supposed to do, and what the company thinks the engineers are supposed to do.

                        Agile-as-originally-described-and-constituted was supposed to help fix this by adding authority and responsibility to the developers, who are the closest to executed-reality. In many cases it was misunderstood and coopted to try to achieve other aims (frequently, enhancing the visibility of management into development activity and keeping an even tighter rein).

                        Fundamentally, though, the work that’s required to do proper planning and estimation is greater than the available skilled resources to do that (whether that’s devs, mgmt, PMI-certified project managers, whatever), and this is more often than not judged by management/product/company as an acceptable tradeoff given market pressures, product pressures, pay incentives, and so on.

                        So what ends up happening is engineers work late hours not out of any immediate or direct personal failing on the part of managers (or their managers), but because holistically the incentives for the entire company are lined up to have that happen. Of course I would say that because I’m a manager. But I promise you that there’s a little nuance. :)