Open, Closed, Remote?!?

Talk about the passion

There has been some very dynamic discussion about workspaces and work in the past few weeks. Elaine’s tweet captures much of the spice of the debate:

Six must read items

People who are way more famous than I am have written some great stuff here. Read it, you must!


My own experience (in open, closed, remote;  as a developer and manager)

In a cube


1999: ATG (dawn of the open floor plan)

A place I never worked, but I thought about it and had friends there. They were open floor plan, with many non-bookable conference rooms around the perimeter. The place was monastic in its silence. It seemed like a wonderful place to work, with good collaboration opportunities and a beautiful space.

1999: In a regular office

At Basis Technology, I was a member of a 5 person engineering team. We each had an office with a door and a window. I had a huge door desk and an enormous monitor. It was very quiet. From what I am reading above this was an earthly paradise for some people.

But not for me. None of my team EVER left their office. Virtually all communication was by email or AOL Instant Messenger. (I’m still friends with all of these people by the way). I hated it. I’m extroverted, I want to collaborate, exchange views in real time.

I was the odd duck here, I now realize the office layout matched the culture perfectly.

2001: open floor plan

I actually refrained from applying to Ab Initio because of their open floor plan. I figured my ADD was bad enough without built-in distractions. It’s a very secretive company, so there are no pictures anywhere, but trust me, it’s BEAUTIFUL. Eventually the interest of the job did me in, and I started working there. (I’m still friends with all of these people by the way).

The culture there was quite innovative at the time, and it is still pretty amazing. There was/is a very high emphasis on collaboration and communication. They dislike regularly scheduled meetings, so just about every discussion is ad hoc. That means YOU HAVE TO BE IN THE OFFICE, no exceptions. They have a lot of nice conference rooms, that (if I recall) are not bookable. The physical space is just beautiful. Teams are organized into pods. The pods are quite spread out, and if you change projects, you just move your desk to where your new team/project is. It’s very dynamic that way.

It can get quite noisy (including heated discussions), it’s a little like a news room.

The one time that the open floor plan simply did not work for me was when I was working on the user interface for an interactive distributed parallel computing debugger. That’s kind of complicated! I simply could not concentrate enough on this project. I ended up taking my computer & monitor to an unused corner of the office and working in blessed peace for about three weeks until the project got under control. I was definitely going against the culture, and made the judgment that shipping the damn thing was more important than being a team player.

The open floor plan basically lowered the cost of communication about as far as it could go. Conversely, the interruptions were as high as they possibly could be. I don’t think you can have both, they may in fact be conserved.

Now: Adobe

We’re distributed. My org cannot achieve colocation except at the team level, and even then that is rare. This has some great advantages: people get to live exactly where they want, we can recruit everywhere, not just in our office locations. And some disadvantages: just about all meetings have to be done over phone or video. We LOVE video and it works ok for us. It’s harder to have spontaneous discussions, but IM helps, a LOT.

San Jose. Every office is different. Most of my org is in San Jose on a floor in the tower with long corridors with equally sized offices with doors. The ones on the corners are a bit bigger for more senior people. The first problem for me is that the office is very drab with poor light. Many of the people are working on desktop products with long cadences and perhaps the office plan works for them. Not for me!

Hamburg. These offices are basically perfect. It’s in an old factory facing the harbor. Each team/group is in a separate room with about 6 desks in it plus conversation areas and white boards. It’s very quiet. All the walls are glass or frosted glass, so the light is wonderful and the collaboration is pretty good.

San Francisco. This is a very famous building, with all open floor plan (except for execs) and low cube walls. The building is beautiful, especially the common areas.

Boston. we’re a small team sharing space with a larger team (Adobe Campaign). The space is being reworked now but will be mostly open, Facebook style. We’ll see.

Remote we have a lot of remote people. See above.

It seems that…

Like much of life this is a matter of preference. When people complain about their office space they MAY be complaining about the culture of the office and of course these are inextricably entwined. So, if you’re designing your space for your org, here are some of the constraints/tradeoffs

  • concentration v. collaboration. To me, coding is a fraction of the job, and communicating and collaborating is a bigger part. But I don’t code at work, so YMMV. (I still thought this when I did code at work).
  • beautiful space with good light. This is a must for me. A crappy work environment would be a huge minus.
  • colocation. For me, this is ideal, but read Matt Mullenweg.
  • whiteboards. put them everywhere!
  • video. if you are distributed, make this plentiful.

The theory of multiple intelligence can be glibly adopted here to state that everyone is right. There is no size that fits all, but there may be a size that best fits the kind of work your org is trying to do. As job applicants, we can also seek out the environment that works for us.

If I was starting a new team/org/company, I would strive for space that had:

  • great light
  • non-enormous rooms, of varying sizes (to fit teams and projects)
  • plenty of non bookable rooms of varying size
  • white boards everywhere
  • lots of communal gathering spots

And a culture that:

  • respected people being ‘in the zone’
  • respected people who have their headphones on all the time
  • valued peace and quiet

How hard could that be?

bell curve

Stack Ranking: Don’t Do it!

It worked for GE!

Jack Welch is generally blamed/credited for the forced rank, ‘rack and stack’ approach to performance assessment. In its purest form a bell curve is imposed at all levels of the org with a forced percentage of people at each performance level. The gruesome details are all over the internet, including on Wikipedia.

Maybe this works great when you’re building jet engines and selling light bulbs. It is a bad way to make software! In this system, your top team in the entire company has to have one piece of toast.

Thank you, Donna!

One of the truly great things that my employer has done is to shift us entirely away from this method. We now practice ongoing feedback . Donna Morris has written and tweeted about this quite a bit. Here’s a nice summary on slashdot. The comments are gold!

The right way to do this is with continuous feedback and an engaged get well process for low performers, certainly not at the end of the year.

And yes…


did end up abandoning it.


In a fascinating linked in article about how Marissa Mayer refused to fire 5000 ‘yahoos’. She stood up at a company meeting and said ‘No’, ‘No’, ‘No’ when the topic of massive layoffs came up. She told the board:

Mayer told them that layoffs of any kind, let alone 35 to 50 percent cuts, would be too damaging for employee morale. She said that Yahoo’s basic infrastructure was so byzantine and jerry-built that it would be unwise to blindly rip whole teams of people out. She said Yahoo was going to need all the talent it could find to turn around, and she didn’t want to risk putting good people on the street.

Byzantine! When was it ever good to be Byzantine?

However, it turns out that what she did instead was to enforce stack ranking, with predictable results. I think I have to give her a pass here, since she did something that was right for the technology (not cause chaotic departures) and was better for the people (low performers out!).

That Said…

if you are not trying to do a massive turnaround you want to stay away from the stacks.

The Year End Review

2015As the end of the year, some unfortunate managers try to roll up their compensation, feedback, goal-setting, project planning and gawd knows what else into a single painful & stressful exercise. Painful for the manager, stressful for the employee. At one of my old jobs, reviews were so important we did them every 13 months!

However, like many seemingly intractable problems, this one can be solved by decomposition.

How much $$?

First, let’s look at from the individual contributor’s POV: “How much $$?” That’s it! That’s what they really want to know. Any compensation changes will probably take the form of items salary, stock, bonus, covered in an earlier post.


In economic terms, the compensation change is the only ‘signal’ that matters here.

You’ve been giving feedback all year, right? Skip it this one time. If you waited a year to give feedback, you really screwed up, and you’ll make matters worse by giving someone a bonus and reminding them of things they could have done better at the same time.


It’s actually fantastic to pivot the conversation around the contributor’s goals. It’s a new year, she is thinking about that sort of thing.

Project Planning

No! This is a one on one, project planning needs to happen somewhere else.

Gawd knows what else

Managers, don’t confuse yourself. Give them their monetary feedback, pivot to their goals and start prepping for the next one on one.

Store config information in the environment, dammit!

love, and so should you! Factor number three says:

III. Config
Store config in the environment

I’ve always thought this just seems like common sense, especially if you’ve ever suffered thru a promotion scheme where you have to modify checked in files in order to push to stage & prod. However, in ‘My $2375 Amazon EC2 Mistake‘, devfactor shows how failing to heed this advice can lead to high stress, and big losses of real money:


When I woke up the next morning, I had four emails from Amazon AWS and a missed phone call from Amazon AWS. Something about 140 servers running on my AWS account. What? How? I only had S3 keys on my GitHub and they where gone within 5 minutes!

Turns out through the S3 API you can actually spin up EC2 instances, and my key had been spotted by a bot that continually searches GitHub for API keys. Amazon AWS customer support informed me this happens a lot recently, hackers have created an algorithm that searches GitHub 24 hours per day for API keys. Once it finds one it spins up max instances of EC2 servers to farm itself bitcoins.

Boom! A $2375 bill in the morning. Just for trying to learn rails.

AWS + Git + Bitcoin makes for a dangerous playground. Stay safe!!

UPDATE: according to this article, the moment you commit, it’s too late. Ouch

Questions that come in threes

Jessica Mah is the CEO/founder of inDinero, a startup to help small businesses with their accounting needs. In an otherwise dull NY Times interview lurks this gem:

Q: Do you have any favorite interview questions?

A: Yes, I ask the same ones to every single candidate. Some are, “What’s the hardest you’ve ever worked in your life, what’s your most lofty ambition and what are you doing for self-development?”

First off, these are great questions! I especially love the question about self-development. The questions have an interesting parallel in the questions we ask in the daily standup:

What did you do (since we last met)?

What will you do next?

Are you blocked?

Past, future and present!


How do we get paid? For some of us very lucky people it’s like this:

  1. Salary
  2. Stock
  3. Bonus
  4. Work Environment
  5. Time Commitment
  6. Knowledge

making ‘enough’ …

Finding a job, and deciding to stay at a job becomes an exercise in seeking the best score across all of these features, although most of us don’t write out the formula much beyond ‘pros and cons’. I don’t have too much to say about the monetary aspects of work (one, two & three), though I do agree that ‘paying people more doesn’t motivate them more‘.

Living in Boston, New York, California is damned expensive, & so is sending your kids to college. So ‘making enough’ could be all that there is, but if not …

we should reflect on four, five & six!

Cesar Kuriyama (one of my big heroes) writes poignantly about making good money in a soul-deadening job, working over 80 hours a week. I’m guessing that was great for one, but a disaster for five & six.

four – work environment

There are a lot of ways to succeed here and probably a lot of ways to fail. But there are some goto invariants for success:

  • Good Tools — in my trade, this means fast computers, big monitors, nice desks. At one of my best jobs ever, they simply never questioned a developer request for a software or hardware tool, everything was pre-approved. Thanks Carl!
  • Smart, Nice people — when I interviewed at ATG, Joe Chung told me that they were looking for two things: ‘smart, nice’. This was the first time I ever heard this formulation, although Joel Spolsky wrote about ‘smart & gets things done’ eloquently later. Oddly, he didn’t mention ‘nice!’, or invertedly:
    • No Jerks — according to google, many companies say they do this. I have my doubts, perhaps because the number of jerks who have been let go is small (in my own experience anyway). But if you are lucky enough
  • Great boss — the converse is even more true: ‘people quit their boss‘.
  • Open floor plan — good for some (I love it), bad for others (Dan hates it!)
  • Plenty of light, not too much gray/beige furniture
  • A mission you believe in

five – time commitment

I have some friends and co-workers with epic commutes. They are trading in a lot of five, hopefully gaining elsewhere, e.g. the opportunity to live in a picturesque Scottish village, or in bucolic New Hampshire, or in a place with great schools, family nearby, what have you. A truly extreme form of this is the fascinating world of remittance flows back to the home country.

Or, your job may not tax you too much, freeing you up to work on your true passion. Before he was a famous novelist (and way before he became another of my many heroes), Anthony Trollope worked as a Postal inspector. Business travel back in the 19th century meant long train rides, so he wrote his first three novels while ‘on the job’.

I, on the other hand, can walk to work. This is an incredible benefit, almost invaluable.

A job that pays 20% more and takes up 100% more of your time, may not be a great deal.

six – knowledge

To my mind this is the most underrated form of compensation. Talk about burying the lead! When you can be paid to learn things, life is pretty sweet. Even sweeter if the things you learn will ultimately help you move ahead in the other five categories.

HBR reports something that sounds intuitively true: ‘women don’t apply for jobs unless they are 100% qualified’. This is very sad to me (especially as the father of three girls), because it turns your career from an inverted funnel into a tube: can’t do it unless you already know how to do it.

My guess is that one of the best ways to be paid to learn new things is by getting hired to do one thing and then convince them you could also do this other thing. Try it and let me know!

on the other hand

HBR reports that this is what people say they value. YMMV!


One Second Every Day

year in the lifeMy band mate Kirk used to come to practices and gigs and discreetly whip out his phone for a bit of filming. Then on FB I saw that he was sharing short videos. “What Ho?” I asked myself. Kirk clued me in about One Second Every Day. Kirk’s been at it for a couple of years now. Cesar Kuriyama’s story is totally inspiring, so go read it (here), make sure to watch the TED talk and read about how it came to be. Nerdvana!

The plan is pretty simple: take one second of footage every day and then assemble it all into a super-cut after a month, quarter, or in my case a year. After a year, your film is six minutes long.

I found that having to film every day made me more aware of what I was doing: “was it interesting, fun, worthwhile?” It also made me much more alive to the inherently interesting things that flit in front of our eyelids every day. I found I did not care very much what ‘other people’ would think about the film, it was simply something that was interesting to me. It also fascinates me how some circadian trends emerged: dogs, cats, my awesome wife, beer, birthdays! The passing of the seasons (a New England special dish) becomes quite beautiful.

A few technical notes. I was running the app on on an iPhone 4s, and found it to be pretty buggy, but I heard from Kirk that he had no such trouble on a later model phone. The editing interface on the phone is primitive, but effective. I actually delayed upgrading my phone because I didn’t want to worry about discontinuity in the middle of the year, although my actual upgrade later was flawless, so I needn’t have worried. Apple is great at upgrades, in my experience.

Two months into the project, my father temporarily died. He was then resuscitated and stayed in a coma for almost a month, and even when he came out of it (which was a miracle!) he was practically a vegetable. Filming my life at that point, felt more than a little weird, so I shot some circumspect footage of the hospital, me, my brother, my uncle at the hospital, footage of the many walks I took around Beacon Hill at that time. I also captured the non-tragic parts of life, since we must go on! But then when he miraculously recovered, I had some documentation of him slowly rejoining the compos mentis. Please check out his fantastic TEDxBeaconStreet talk about the experience.

A few days in the year, I just didn’t remember to film. Whoops!

I filmed for a year from August 2013. Enjoy!