Why Joel Spolsky's Crazy Schemes Never Paid Off

I believe Joel gets some of his ideas about workspace arrangements from the book "Peopleware" by DeMarco and Lister.

In the book, they don't necessarily argue for an individual private office for every programmer, but argue that a quality work environment has enormous benefits. Interruptions cause huge problems, and random noise can be distracting as well. A door that shuts is one of the most important parts of being able to reduce interruptions and random noise.

A personal office isn't mandatory in order to make progress along those lines. The book actually suggests that you set aside a space for a small group of people (1-4), with furniture that can be easily rearranged.

edit: Just a another comment on the article...

BDUF Big Design Up Front: I think it might be a mistake to assume Joel's arguments in favor of software specifications imply BDUF. A specification is a fluid thing. It helps ease communication between all parties throughout the development process, and lets you save time figuring out some kinds of problems.

I disagree with his idea that you should not have a private office.

Granted, I don't currently have an office job, but given that a phone call or knock on my door will rip me straight out of the flow, I can't imagine even being able to get in it in the first place in a room with half a dozen other people all madly typing away, making noises, and moving around in my peripheral vision.

I wonder how much joelonsoftware contributes to his companies success. The blog is something one couldn't easily mimic, as few people can write as well.

I think his backup plan paid off. He has a privately held profitable company.