Robin wrote:
[*]And lastly, I believe (2) causes a paradox (or at least an infinite loop) if two votes are made at the same time by the same member: ... I know the way out is that one should be considered valid and the other not (it does not matter which, since they're the same anyway), but the file does not specify that.[/list]
I propose changing it to:
New Community File wrote:
A valid vote is any post in this thread made by a member of
the LPCL development team which
contains "yes." or "no.", but not both or as a subword,
does not have any other valid vote cast by the same member of
a newer time,
(...)
If two valid votes exist by the same member of the same timestamp, one of them will be declared unvalid.
Yes, you're right about (2). The intention was that neither post should be seen as valid if two were posted at the same time. No newer posts may cause a vote to be counted twice. How about "no valid vote of a newer time. If a user has more than one valid vote, then none counts." LPCL is scheduled to be rephrased as a policy regarding the trademark (the project name more specifically) rather than anything about copyright (where BSD, WTFPL, or the like may be used instead).
To bring this back up, something relevant: one of the Glibc developers (one of the key ones, too) is causing enough strife for Debian to switch to a fork of it. Community-versus-developer(s).
