Thanx and a hat tip to the alert reader who sent us this.

Interesting paper. I've contacted the first author.

My instinct is that people with very few or very many edits are much more likely to be doing gnomish work than those with a middling number of edits. Of course, it depends on whether someone cuts and pastes whole article from a word processor, or saves articles one sentence at a time etc.

A cross-reference with the name space and the size of the edits would help clarify things.

That's just the way I edited in The Before Time, when I was still under the delusion that WP was an encyclopedia. The reason was simple: I soon discovered that all of my work would be lost if I had the edit page open for more than about five minutes at a time. It was much easier for me to do most of my editing in my word processing program off-site, then import it all as one edit to the WP article I was working on. The clunky way the editing pages work essentially force you to do it that way, or to edit in small "bite-sized" chunks that take a small amount of time. Of course, the latter way is preferred by those looking to inflate their edit counts.

For a moment your answer didn't make sense to me until I realised that I missed out "sentence" in "one sentence at a time".

I try to remember to press preview to avoid losing stuff and to save one paragraph at a time, but every now and then I have a large talk page comment written, lose it and then resort to a one sentence reply instead. I wonder whether the new interface will do a better job of not losing things. Not that it matters to you now.

It is rare that an edit is truly lost because of edit conflict. It is merely a bit inconvenient. At least most browsers, using Back, will recover the edit. What really happened to lose edits, for me, would be that I don't notice the edit conflict, and left my browser and then later closed it. Sometimes, then, I couldn't recover the edit. I got into the habit of copying all of a longer edit to my clipboard before saving, if I was at all concerned about activity. That's faster, then, to re-add the material, rather than using the default edit conflict procedure.

Making lots of small edits is often rude, my opinion, it makes it harder to follow what's going on in history.

If you are SlimVirgin, it's a time-tested means of intimidating your opponent.

I'm not really buying one of their hypotheses, that the pool of "potential novel contributions" lessening is causing a quasi-biological "survival of the fittest" situation. It sounds swell, but I don't think close examination will back this up.

The norms of sourcing are tightening up as the early framework and "low hanging fruit" have been taken care of. The work to be done involves specialist knowledge. Moreover, stuff needs to be footnoted more tightly than it was in 2004 or even 2006. This means your average drop-in type editor starting a new page is apt to run afoul of tightening norms and standards and growing bureaucracy.

It's not a "struggle for food" it's a constantly rising bar in terms of the requirements for new content.

That said, it was an interesting paper. I still think that the really interesting take on Wikipedia's declining pool of editors will emerge when the metric of "characters of content added" rather than "number of edits performed" is examined.

My sense is that the number of administrative actions is steady, the total number of content creators has fallen and is still falling, but the core group of 200 or 300 or whatever "dedicated" writers -- who are, I believe, older than many people think they are -- is not declining in either number or quantity of output. I'm not sure this pool is replenishing quite as fast as it is draining, but it is close.

There is a steep learning curve to develop the technical and political-infighting skills necessary to make protracted contributions, that's the main fetter.

English WP is a maturing project and its main growth will be qualitative rather than quantitative over the next ten years.

t