The Cathedral and the Bazaar book club questions

1. 1.

   Raymond contrasts cathedral and bazaar models of development. Where else do you see these two modes of production competing — in fields outside software?

2. 2.

   Linus's Law assumes many eyeballs are reviewing code. In practice, most open source projects have very few active contributors. Does this undermine the argument or just change its scope?

3. 3.

   Raymond argues that distributed open development can outperform closed expert development. How does this claim hold up given the security vulnerabilities that have appeared in widely used open source libraries?

4. 4.

   The essays explain open source contribution through reputation and reciprocity. How much of your own professional or creative work is driven by those same incentives?

5. 5.

   Raymond writes about hacker culture as a gift economy. Does that description still apply to the open source ecosystem twenty-five years later, or has commercial adoption changed its character?

6. 6.

   The book was written before GitHub, Stack Overflow, or modern open source infrastructure existed. What does the current open source ecosystem confirm and what does it complicate in Raymond's argument?

7. 7.

   Raymond was also controversial for cultural views outside software. How do you read technical arguments from an author whose broader values you might not share?

8. 8.

   Release early, release often is good advice for some software. Where does that principle break down, and what does that tell you about the limits of the bazaar model?

9. 9.

   Large companies now employ the majority of contributors to major open source projects. Has open source been co-opted, captured, or successfully scaled — and how do you tell the difference?

10. 10.

    What is the equivalent of open source collaboration in your own field? Where could distributed peer production outperform centralized expert production, and where would it fail?

11. 11.

    Raymond uses his own fetchmail project as evidence for his thesis. Does a self-reported case study like this carry weight for you, or do you require independent evidence?

12. 12.

    The title essay was written in 1997 and the book published in 1999. Which of Raymond's predictions turned out most accurate, and which did not?

How to run the discussion

• Pick six to eight, not all of them. An hour of real discussion covers fewer questions than anyone expects, and racing the list is what makes a book club feel like homework.
• Open with what people noticed, not what they concluded. Asking for a verdict first hardens opinions before anyone has heard each other.
• Let the disagreement run. The questions below that split a room are the ones worth the time; the ones everyone answers the same way can be skipped.
• Ask the quiet person a specific question. A direct, narrow question is far easier to answer than an open invitation to contribute.