Page 7 of "On the Evolution of Quality Flaws and the Effectiveness of Cleanup Tags in the English Wikipedia" makes for interesting reading for editors interested in knocking over some smaller clean-ups. Fifelfoo (talk) 03:23, 26 June 2012 (UTC)[reply]
Per: "In summary, low-activity and new editors, along with anonymous users, tend to delete more than they contribute; this reinforces the notion that Wikipedia is largely the product of a small number of core editors." - So when is the foundation going to stop obsessing about making Wikipedia ultra-friendly to IP editing and to start getting serious about studying who its content-creators actually are and what tools they need to do their jobs better??? Here's one hint: we're older than you think we are. Here's another: we need access to JSTOR. Carrite (talk) 23:52, 26 June 2012 (UTC)[reply]
Well duh, Killiondude. If instead you counted on anon IPs and part-time contributors, I think the study shows pretty clearly that we soon wouldn't have much left to look at! MeegsC | Talk11:27, 27 June 2012 (UTC)[reply]
There's a good reason for the Foundation's advocacy of the myth that anyone can contribute & improve Wikipedia: funding from major foundations. It's easier to get a charitable grant for an encyclopedia "anyone can edit" than one only a select few can. Money does influence content & presentation in ways undreamed of by your philosophy, Horatio. -- llywrch (talk) 16:58, 27 June 2012 (UTC)[reply]
Anyone else notice that the contribution percentages might be skewed by reversions of page- and section-blanking? Some of the most "prolific" contributors might just be wikignomes who revert blanking vandalism. PowersT12:32, 28 June 2012 (UTC)[reply]
Yes, blanking a page or reverting it are going to look much more dramatic in this methodology than actual content edits (or justfied removals), if every character gained or lost is equal. So it doesn't really show you who the content contributors are. The study shows that accounts with only a few edits are net contributors--but that probably just shows that blanking vandals are less likely to register. And even if the content contributors are a select group, the "1%", you still need to find ways of steadily injecting that group with new blood or else it will be lost through attrition (regardless of whether the WMF caters to the preferences of current contributors, there's going to be a natural dropoff rate). The track record of Sanger's Citizendium model is clear, cliquishness does not build an encyclopedia. 169.231.98.141 (talk) 19:12, 28 June 2012 (UTC)[reply]
Since much of the text in wikipedia is interwiki links, I can guess that persons running interwiki scripts can run up a huge % of the total text.
Also, 1% writes 100%?? Sounds like a problem with counting methods. If someone adds a new fact, and I completely rewrite his text, is it counted as text written only by me? --Enric Naval (talk) 09:44, 30 June 2012 (UTC)[reply]
I've often wondered who the 1% are 20% of. I mean, 20% of the people doing 80% of the work is an old story. There is no bright line saying where the community ends or begins. So it's as if, maybe, the 5% are those we should concentrate on? Where does this lead? Charles Matthews (talk) 20:47, 28 June 2012 (UTC)[reply]
the law is generally recursive: of that 20%, 20% of them will edit 80% of the 80% of total articles--in other words, it is expected that 4% of the editors will be responsible for 64% of the content. WP is similar to other human activities. It would take a highly artificial structure to do otherwise. DGG ( talk ) 18:05, 29 June 2012 (UTC)[reply]
The thing about counting up how many characters were added by which editors really should have also mentioned an earlier similar study by Aaron Swartz.[1] In both cases I have some doubts about the methodology and in particular in the more recent study, I'd like to know if there was some attempt to separate out additions made by automated scripts rather than human editors. 69.228.171.149 (talk) 07:40, 2 July 2012 (UTC)[reply]
The story about the citation needed template isn't all that surprising. Quality content contributors (like those working on FAs) often use or encourage the use of profligate cn tagging on a specific article, so they can identity claims that need referencing in what is often an already densely referenced article. They then go about replacing all the tags with references. By contrast, unreferenced is usually a well-intentioned drive-by tag, left for some unidentified individual who has knowledge of the subject and may never respond. --Dweller (talk) 11:51, 28 June 2012 (UTC)[reply]
← Back to Recent research