---

Some Implications of Bazaar Size

Introduction
One form of open development model ("bazaar") has been described by Eric S. Raymond in his paper, "The Cathedral and the Bazaar" (available at http://tuxedo.org/~esr/writings/cathedral-paper.html with related links including a link to this page at: http://tuxedo.org/~esr/writings/cathedral-bazaar/)

This document assumes the reader is familiar with Raymond's paper. This document references the Raymond paper, and attempts to use the same terminology.

One of the strengths of an open development model is that it allows (and even encourages) many people to share enhancements and defect fixes to software (collectively called "modifications" hereafter.)

Frequently the existence of a modification is a result of a user satisfying an individual need. The existence of a modification is made known to others by a mailing list, newsgroup, or other discussion method. Keeping with the bazaar metaphor, this is equivalent to "advertising your goods."

When the bazaar model described by Raymond is working well, (preconditions met, etc), there will be an abundance of advertised modifications.

In this paper, I postulate that

• The size of the bazaar (number of participants) affects the "health" and success of the bazaar, even if not a pre-condition to successful bazaar startup.
  
  
• Not simply "size" but "effective working size" has important implications for the continued progress of the product and strength of the bazaar.
  
  
• Debugging by testing includes factors which give the bazaar a very much smaller "effective working size"
  
  
• There are alternative debugging methods which might be used to counter a small "effective working size."
  
  
• There are activities and natural progression which tend to erode the "effective working size" of the bazaar over time.
  

I would like to thank Eric S. Raymond for sharing his paper, (which led to this one) and for valuable comments and input which fueled new ideas incorporated into this document.


1. The Importance of Bazaar "Effective Size"
2. Debugging by testing dramatically lowers effective size
Commentary by Randy Boring on debugging by testing : Open is still much better than closed source
3. Alternate Debugging methods (useful for lower effective size.)
4. Other Factors which erode effective size

Revision History

---

Up to Mib Software main page