Many opensource projects leaders aren't directly paid for their work on it, and consequently there is no obligation for them to accept bad code. This is less true in most closed-source projects, especially when the release date is set: various testimonies and sourcecode releases reveal how awful some well-known software products' sourcetrees are.

Even if paid to develop opensource code, having to publish may "fend off" and screen out some of those who know they will often be catched with their pants down (submitting or committing bad code).

Such exposition also sparks emulation between developers, pushing them forward and up. In some (most?) closed-source contexts the best ones (becoming fed up) leave and are often replaced with less good ones, and the cycle restarts.