In a lot of ways, authorship doesn't really matter in wiki. In fact, it kind of gets in the way of the development of good wiki pages.
But authorship can be important factor for, say, licensing issues, or for MetaData. It's worth thinking about, if just to dismiss it.
Note that we're not talking about copyright, ownership, or other closely related, but different, issues.
In a spatial authorship model, different parts of a wiki page (sections, paragraphs, sentences, words, etc.) are authored by different contributors. For example, when Alice starts a page, it looks like this:
A A A, A A. A A? A A A.
Bob later edits the page, adding a second paragraph:
A A A, A A. A A? A A A. B B. B B B, B B. B, B B B B. B.
We can say that Alice is the author of the first paragraph, and Bob is the author of the second. This passes some common-sense tests about who is author of what. However, it's hard to answer the question, now, of "Who is the author of this page?" Alice? Bob? Are they co-authors?
When Charlie comes in and makes modifications to both paragraphs, though, we start having some more difficulties:
A A A, C A A. A A? A A C A. B C B. B B B, B B (C C C). B, B B B B. B. C C C, C C.
Is Charlie the author of a few words, and a sentence? What is Alice author of, now? Who is the author of the first sentence of the second paragraph: Bob? Charlie? Are they co-authors?
And when Darlene refactors the page, without adding any new text, things get a little hairier:
A A A, C A A. B B B, B B (C C C). A A? A A C A. B. C C C, C C. B C B. B, B B B B.
Darlene has significantly changed the layout, presentation, and probably expression of the ideas of the page. And yet, because she didn't leave any visible marks (except perhaps some whitespace), she's not really the "author" of anything in a spatial model.
In a few more rounds of editing, Bob's contributions might be entirely replaced by other words, sentences or phrases:
A A A, C A A. E F F, G G (C C C). A A? A A C A. F. C C C, E C. F C E. G F E E E, E F.
...and yet his contribution to the ideas and format expressed on the page may still exist.
This kind of mishmash is not at all uncommon in DocumentMode pages.
Some advantages of a spatial model:
In a temporal authorship model, each contributor creates a brand new work based on the previous version of the page. Each page version is "authored" by the person who makes the last edit, with "contributions" or "co-authoring" by all the people who have made edits previously.
Considering our example above, in a temporal authorship model, Alice would be the sole author of the first version, and might be listed as such in some theoretical version history of the page:
* 1997-04-02 SomePage author: Alice
When Bob makes his change, he is now the "primary" author, since he's in some abstract way "responsible" for the way the page looks now, even though he didn't write all of the page. Alice is now a contributor.
* 1997-04-08 SomePage author: Bob, contributors: Alice
When Charlie works on the page, he get credited as the primary authors:
* 1998-10-30 SomePage author: Charlie, contributors: Alice, Bob
The justification here being that Charlie took the work that Alice and Bob had made and made something new and original out of it. He made additions that change the nature of the work. Alice and Bob deserve recognition, but the page as it is now is primarily due to Charlie's effort and thought. This is more striking with Darlene's contribs:
* 1998-11-03 SomePage author: Darlene, contributors: Alice, Bob, Charlie
Remember, Darlene didn't actually write anything: she deleted, re-arranged, and refactored. Again, the idea of crediting Darlene as the author is that the new version of the page is due to her judicious review.
Similarly, we can view the entire version history with different authors and contributors:
* 1997-04-02 SomePage author: Alice * 1997-04-08 SomePage author: Bob, contributors: Alice * 1998-10-30 SomePage author: Charlie, contributors: Alice, Bob * 1998-11-03 SomePage author: Darlene, contributors: Alice, Bob, Charlie * 1998-11-03 SomePage author: Evan, contributors: Alice, Bob, Charlie, Darlene * 1999-03-17 SomePage author: Frances, contributors: Alice, Bob, Charlie, Darlene, Evan * 2001-08-12 SomePage author: Greg, contributors: Alice, Bob, Charlie, Darlene, Evan, Frances
Note that even though Bob's actual language isn't present in the last version, he still gets credit for making a contribution to the page.
One thing to note, though, is that Greg ("G") only changed a few words on the page. Common sense rebels at calling him the "primary" author. Our idea that he has reviewed and judged the rest of the page to make a new work kind of stretches credulity here. "C'mon! He changed three words."
Some advantages of temporal authorship model:
Authorship can be modeled by combining both temporal and spatial notions of authorship, as follows. First logically split the page up into different areas. Then, for each area, the primary author of that area is the last person to make changes to it, and the contributors to that area are all the people whose edits have (directly or indirectly) effected it.
Another way of looking at authorship is to consider the set of all editors of a page as "co-authors". In other words, the page is an indivisible unit, created, blended, remixed and refactored by all the folks who took time to edit it. Each contributor gets equal credit as co-author.
In other words, a group of people got together and created a text. Their mechanism for creating this text is immaterial; what matters is that each contributed. Individual page versions or parts of the page don't matter; it's the text as a whole, and as it exists right now, that matters.
Some advantages of group authorship model:
It's also possible, and perhaps necessary with some WikiEngines, to eliminate the authorship role of individuals altogether. Any particular page is authored by the community as a whole.