One co-founder of Fog Creek has this to say, as a goal for the company [1]:

"* great work environment, with the goal of every engineer having an office with a door that closes"

And, in an oft-cited "Joel Test" [2]:

"8. Do programmers have quiet working conditions?"

Furthermore, the company itself says [3]:

"The average Fog Creek developer has 694 square inches of screen real-estate, 2 desktop computers, and an Aeron chair. Most have private offices with windows and doors."

Googling for "Joel on software door that shuts" will turn up similar results.

A shutting door is clearly important to the management, remote or no.

[1] http://www.joelonsoftware.com/articles/fog0000000038.html

[2] http://www.joelonsoftware.com/articles/fog0000000043.html

[3] http://fogcreek.com/about/

From reading the comments in this thread, I don't think very many people in this thread know who Joel Spolsky is.

Or the alternate explanation, most of the people in this thread disagree with Joel Spolsky. Given occam's razor and his fairly high profile, which is more likely?

Disagreeing with Spolsky's writing is one thing, but unless you've seen or heard a lot of Fog Creek employees calling it a horrible place to work, then I would say that any speculation about how horrible a place Fog Creek must be because of the tone and content of the OP would suggest that Mistermann is right.

> Given occam's razor and his fairly high profile, which is more likely?

I'll hazard a guess that you're "older", like me. I don't think Joel Spolsky has a high profile at all with forum active programmers, I mean, when did he drop out of public writing, it was many years ago.

At least to me, most of the criticism here is generally of the nature "you're wrong"....not subtle, considering the variety of things involved, just that programmers are, without exception, incredibly productive, and all work should be able to be performed asynchronously (or you're doing it wrong).

I just don't find the comments in this thread very reasonable, it seems like a bunch of young and very smart/competent kids assuming everyone is as smart and competent as them.

Do you get a sense of humility from very many of the comments here?

Fair enough, i certainly fit in the "older" category. I find it less than intuitive that those active here don't know the co-founder(and current CEO?) of Stack Exchange, or the team behind Trello. But i also accept I am not the target audience of HN and times perhaps have changed.

I also spent three years recently working for a company with < 10 employee's and no office, we all worked online from around the world. I don't think we followed any of the listed requirements. No video chat, no headsets(just text communication, same as how we developed open source software). Team members took care of their children whenever necessary and we did not have specific start/end times to our day. Some days i would take off for a 3 hour "lunch" (really an energizing ride through the mountains) and no-one ever asked about it. I believe they never asked because I was judged by my deliverables, little else mattered. We also didn't have any specific "work hours". Having team members from both coasts of America, Brazil, Bulgaria, Zimbabwe and the Philippines meant time of day meant rather little. Not having a specific start/end of my day often meant i worked much more than a 40 hour week, but it was a labor of love anyways.

Is he wrong? Hard to say, many different things work for different people. Is it right for me? Certainly not. Do the other people in this thread saying he's wrong believe they work better in alternate conditions to the ones proposed? Probably.

But what's the point of having a door if you're constantly chatting or video conferencing on Google Hangouts?

Rules like this usually exist because there were people who did the opposite at some time in the past. It's a simple enough requirement; isn't having it better than disallowing remote work, or disciplining someone for problems created by not having a door (kid runs in in the middle of a staff or client meeting) when no such requirement was previously specified?

I would say one of the tools required to be productive as a programmer is a focused work space. You can give that tool to people who come to the office. But if someone works from home, that's the one tool they have to provide themselves.

> "Have a room with a door that shuts." Seriously?

Yes. Seriously: http://www.dorsethouse.com/books/pw.html

> How about this: hire smart people you trust, give them whatever tools and support they need to do their jobs most productively, then get out of the way.

Good idea: http://www.dorsethouse.com/books/pw.html

I don't understand -- if you could give some rational, reasoned argument or study that shows that smart people are very less productive on an average if they are working remotely instead of repeatedly chanting the link to a book whose reviews read "Bill Gates has built a company full of managers who read Peopleware" (that is supposed to be a plus?), your argument might make more sense.

Firstly I was pointing to the origin of "a door that shuts". It's not an example of micromanagement, it's an example of a finding from a study performed by IBM that programmers in private offices were more productive on any measure.

Secondly I was pointing out that the same book also tells managers to recruit bright people, give them goals, enable their work, and then go away.

I suspect Peopleware is the most widely not-actually-read book in this industry.

Do you really think this is a literal rule?

"Can I work remotely?"

"Will you have a room to work in?"

"Yes"

"But does it have a door?"

"Well, no, it's a desk in the living room, but I'll be home alone all day."

"Sorry, no door, no remote work."

Consider, Fog Creek has spent substantial funds making sure that their on-site devs have offices with doors that close. Is it unreasonable, then, to expect a remote worker to have a distraction-free work environment?

Well said!