If we list all people, do we list editors on the frontpage (exactly like PEARdoc does)? Who are editors? They need to be picked by hand, no? If we list all people, who we list on the phpcredits() page? Noone? All people do not fit there.
Then credit them as a "cast of hundreds" and make it link to the list in the back of the manual. My point is that it is a waste of time trying to track down just "how much" to credit someone; everyone that contributed, contributed. If you really want to rank them, do it in reverse order of commits, but you should still list them all.
Authors of user notes sometimes contribute a lot more than those we have tracks of in CVS. We just integrate user notes without giving credit to their authors. It is just not possible to list everybody. Big chunks of the manual are based on user notes. Even if we say now we add the author tags to sections based on first commits, it will turn out that quite a few of those commits are actually user note integrations.
They make up the small list which we can put into phpcredits().
Ah, now I see the motivation. It's a tricky proposition. There is no fair way really to credit without crediting all involved parties, IMO, and that would be too long a list.
Right, this is our problem. If you look at the phpcredits() output, you see some names picked out of those who contributed to the PHP project. You know more people contributed to those extensions, APIs, etc, but you still have some names picked out on the phpcredits() page as *most important people*. Now I don't see if all the PHP source code crediting system is based on this picking, then why this is a problem for the PHP documentation to be based on a similar system?
Anyway, I don't have particularly strong feelings about what does or does not happen, I just wanted to point out that I don't think ranking based on an arbitrary cut-off point in CVS commits is really all that fair (for instance: one person documents a new extension in great detail and commits it but doesn't get higher than the bottom of the list; is it fair on them to not be credited?), and can't see a practical fair solution short of listing everyone that contributed.
I have provided the raw commit list as a base and already made three exceptions (listed Hartmut as a previous author, but said I would not grant two of those in the top list "top author" credit). Plus we are going to make an exception with Youni too, since he has not reached the commit limit I have proposed. I have also said I have no problems with more exceptions, actually I have started this discussion to make exceptions.
Goba
