I recently read (most of) the 37signals book and I highly recommend it to anyone. These were some observations I made with logging for my Dynamic Web Development Course:

I’m going to write this log as I read it, so I can get the most out of this documentation. (I’ll try not to contradict myself). I really like the less is more philosophy they present in this reading. It is becoming quicker, cheaper and easier to publish poorly designed Web 2.0 projects on the Internet, so you see them everywhere. The only way to stand out as an educated designer of new media, is to design less well. To perfect the craft of solving a single problem with a custom solution, rather than throwing all the latest widgets at it to see what sticks.
This would seem to contradict my reasoning for voting for the Drupal path in this course rather than the Flex, and it seems that way because it DOES. I think it is incredibly valuable to learn how to perfect certain key features of a web project. However, I stand by my decision, because I still believe that is a self-taught discipline and requires more self-conditioning than peer critiquing. I want to use this course to learn how to take all the power of Drupal and whittle it down to a specific solution to a specific problem.
The concepts of the third chapter seem repetitive of the second. While I believe you can produce a lot quicker with a small qualified team, I believe it is difficult to the sort of people that can fill that role effectively. These are people who are multifaceted, end-user-driven, and can communicate well outside of their expertise. In other words, not everyone is a graduate of the Multimedia department of UArts. (I kid). (Sort of).
Chapter 5 speaks to me when I think about wanting to provide users with 1001 ways of accessing the same content. What I’m getting from this chapter is that users feel more comfortable finding their own way with fewer paths that are wider and more flexible, then many very specific, very constrained paths.
I really enjoy the definition of Web 2.0 by Seth Godin, “Web 2.0, the world where we start by assuming that everyone already uses the web.” While some might say it over-simplifies the user, I think it admits a rare level of respect for them. It simultaneously gives the developer a sense of humility when it comes to thinking that it is ever possible to anticipate every problem, and a sense of confidence to move forward and let your users debug your product.
A lot of Chapter 6 to me speaks about confidence. “This isn’t brain surgery, it’s a web app.” “I don’t want to hear people’s ideas. I’m not interested until I see their execution.” “Next time someone tries to pin you down for an exact answer to an unknowable question…say ‘I don’t know.’” If your core concept is strong (where there is strength in simplicity and specificity), just keep moving forward and don’t get hung up on having all the answers right now.
Chapters 9 and 10 have very good referential information. Balances to keep in mind throughout the entire project. Don’t expect to know every hole, just have a general net to catch errors underneath. Code quickly, but remember to revisit recurring glitches. I’m also, so glad to hear I’m not the only one who hears code speaking to them sometimes…
I have experienced the documents referred to in Chapter 11 from both the developer and customer points of view, and agree that they can be extremely counter-productive. I also like the priority they put on using “real words” instead of Lorem Ipsum. I picked up this bad habit, and I use it as an easy way out. This is why interactive designers are not supposed to hang out with graphic designers.

Overall, I think this is a brilliant book conceptually, and as a reference. I love the humility they show in short sweet points rather than long tirades repeating the ideas, and in quoting so many outside sources. There were times when I felt like this book was just an excuse to brag about how great Basecamp and their team are, but really they are just exemplifying the confidence we should all show as designers that takes a good idea to fruition.