If we use a metric to determine who belongs on that list, and someone
complains that he/she is not on it, all we have to do is explain the
metric -- for instance, "You haven't contributed at least 5 commits in
the past 6 months." Any of us on the list could respond to that
complaint with the exact same answer. Having a well-defined metric is
important, because then we're able to present a concrete reason for not
having that person on the list.
Having only the last 6 months contributors listed would be a very bad idea, as most of the old authors, who contributed lot to the manual, won't get credited. Have you seen Hartmut doing work in the last six months? I have not (forgive me, if I am mistaken). But he definitely deserves a top place on the list.
On the other hand, if the list is hand-edited, not only do changes have
to go through a "select few", but if those "select few" deem that
someone isn't going to be on the list, it could really get personal.
That's the danger inherent in building something without a well-defined
metric. The problem then becomes not "I haven't contributed enough", but
"Goba/Philip/whoever doesn't like me" or "They don't feel like I'm a
valuable member of this team." This could lead to the manual losing
contributors because there would be no easy way to answer the question
"Why am I not on that list?" Making it hand-edited invariably makes the
decision of whether or not someone goes on the list personal, and by
bringing personal feelings into it, someone is going to feel slighted.
Are we really seem to be that scary? Well, we had Jeroen here for example, who some of you may also remember. He was not really able to work in a group, but added many stuff to the docs, especially into the types section, and clarified many parts (there are still XML comments labeled with his name in the doc sources). I don't think that there are too many guys here, who liked him, but he was part of our history, and he definitely added value to the documentation. I a voting comes on his name, I'll vote on adding it. We just need to dig up, what he had contributed, and categorize him to the right part.
As I said before, any addition requests can be open. It can be discussed on the list (as IMHO there won't be too many after we finalized the list we would like to change the current one to first). As I proposed, one would need to have a fixed number of 'PHP Documentation Group member' votes, and no negative votes from any 'PHP Documentation Group member'. If any negative votes come in, then it need to be discussed, and the negative voted member should be convinced that the propsed contributor deserves to be listed.
As far as I see, only a human edited list can be valuable. Any automated process will degrade the recognition of those who are on the list, as they are not based on added value, but added quantity.
I don't think that political problems will come up, as we are reasonable guys, and have enough members on the list to stop any conspiration.
Goba