Hey folks,

I'm glad the presentation came across so well.  I really appreciate the discussion.  

Pine, I really appreciate those plots that you linked.  It seems that you can identify the progression through barrier types by following the hexagonal graphs clockwise.  Concerns start with complex rules and (to a lesser extend) the difficulty of editing and progress to concerns negative social behavior and access to reference materials.

Regarding editathons, I'm not quite sure the right way to measure their effects.  I suspect that one of the biggest effects of editathons are the result of discussions that people have with their friends and family after the event.  "I edited Wikipedia and it was fun.  It turns out that there's a lot of different types of ways to contribute.  You don't have to be an expert." -- is a conversation I imagine is relatively common after an editathon.  The awareness (I can edit Wikipedia?!), new registrations and contributions that result from such once-removed discussions would be nearly impossible to track.  

Jane, seem more of my work exploring the rising social/motivational barriers here: https://www.youtube.com/watch?v=bozyc1z25aQ#t=24m49s  In the conclusion of that talk, I bring up Snuggle[1] as an example of a technological strategy for supporting desirable social behaviors.  My recent work on the Revision Scoring[2] was originally inspired by my work to extend Snuggle beyond English Wikipedia -- I needed vandalism prediction scores beyond English Wikipedia!  Generally, I think we (as Wikipedian community members) have a lot deeper insight into the types of behaviors (e.g. reactions to newcomer contributions) that are desirable than we had in 2006 and that, if we were to redesign counter-vandalism tools from scratch with these insights in mind, we'd be able to dramatically reduce this type of social/motivational barrier.  I think Snuggle is a good example of such a new type of tool and the idea with Revision Scoring is that I'd like to make it *really easy* for others to experiment with their own strategies.  The next thing I want to do is to try empowering WikiProjects with automated quality control/socialization tools.  I suspect that, WikiProject members will be highly motivated to socialize potential good newcomers and help them work productively within the topical context of their WikiProject -- if they had the means to do so efficiently.  

1. https://en.wikipedia.org/wiki/Wikipedia:Snuggle
2. https://meta.wikimedia.org/wiki/Research:Revision_scoring_as_a_service


On Mon, Aug 3, 2015 at 7:36 AM, Jane Darnell <jane023@gmail.com> wrote:
OK I am replying to this mail, as this one has the link to Youtube in it with the two presentations. I am only responding to the first presentation by Aaron here.

In general I like the idea of focussing attention on the "New Editor Activation Funnel". This area is of course the reason why we have a decline in new editors, and it all has to do with an increase in "barriers to entry" (which btw I am not convinced is the same thing as "technical impediments"). It is useful to split these barriers up into Permission, Literacy (here wikimarkup is lumped together with policies), and Social/Motivational (human interaction) issues, but I think the whole presentation misses the point on the need for more dissection of the reverts problem (shown a bit towards the end). 

I personally think that demotivational behavior by experienced Wikipedians is the biggest factor in the decline of new editor contributions, but unlike most people I don't think this has to do with what the experienced Wikipedians do, but rather what they don't do. They don't welcome people in person (because they don't see their edits) and they don't give timely feedback on first edits to pages on their watchlist (no way to see if those edits are first time edits). They don't show them the ropes in that if one wants to make a BLP, or an article about a company or building or place, or an article about an artwork, you should look at existing examples and start from there. Having said this, I do think we spend an inordinate amount of time on things like extending the page about WHAT WIKIPEDIA IS NOT (which btw I have yet to read). It seems that our best way of dealing with newcomers is to throw CAPS at them, though we all hate CAPS.

The point of this study was to prove these two: H1: VE will increase the amount of desirable edits by newbies and H2: VE will increase the amount of undesirable edits by newbies (aka VANDALISM). Guess what? Both H1 & H2 show no significance and if anything, less vandalism came from VE editors. I could have told you that beforehand - yawn. It angers me when people assume that others are not technical enough for Wikipedia. Sorry, but it is not rocket science. 

This type of thinking is not just on Wikipedia, I see this also in health occupations, where doctors tell their patients not to go look things up on the Internet. Just trust the doctors because they studied it! Yeah right, like I am going to trust all aspects of my future health and well-being to someone who sees my future health and well-being as a 10-minute interlude in their 9-5 workday. No, I will nod politely (one must always remain friendly) while googling my way to better health, thanks. And if I want to make an article about something that I think needs an article on Wikipedia, I am going to try to do it on my own as far as I can get, and I am probably not interested in talking about it until I am done. The whole AfC queue thing is absolutely horrible because it puts these edits on ice until the person totally forgets what the password was that they dreamed up for their user account. As far as spelling corrections go, if I correct an error and see it deleted (like from Kiev to Kyiv, which will be reverted by a bot probably), then I will probably not come back.

I am very eager to hear more about the revision scoring though! I wish there was a better way to do that than manually however.

On Wed, Jul 29, 2015 at 8:07 PM, Leila Zia <leila@wikimedia.org> wrote:
A friendly reminder that this is happening in 23 min. :-)

YouTube stream: https://www.youtube.com/watch?v=vGyrVg_qKSM
IRC: #wikimedia-research


On Mon, Jul 27, 2015 at 2:47 PM, Leila Zia <leila@wikimedia.org> wrote:
Hi everyone,

The next Research showcase will be live-streamed this Wednesday, July 29 at 11.30 PT. The streaming link will be posted on the lists a few minutes before the showcase starts (sorry, we haven't been able to solve this, yet. :-() and as usual, you can join the conversation on IRC at #wikimedia-research.

We look forward to seeing you!


This month:
VisualEditor's effect on newly registered users
By Aaron Halfaker
It's been nearly two years since we ran an initial study of VisualEditor's effect on newly registered editors. While most of the results of this study were positive (e.g. workload on Wikipedians did not increase), we still saw a significant decrease in the newcomer productivity. In the meantime, the Editing team has made substantial improvements to performance and functionality. In this presentation, I'll report on the results of a new experiment designed to test the effects of enabling this improved VisualEditor software for newly registered users by default. I'll show what we learned from the experiment and discuss some results have opened larger questions about what, exactly, is difficult about being a newcomer to English Wikipedia.

Wikipedia knowledge graph with DeepDive

By Juhana Kangaspunta and Thomas Palomares (10-week student project)
Despite the tremendous amount of information present on Wikipedia, only a very little amount is structured. Most of the information is embedded in text and extracting it is a non-trivial challenge. In this project, we try to populate Wikidata, a structured component of Wikipedia, using DeepDive tool to extract relations embedded in the text. We finally extracted more than 140,000 relations with more than 90% average precision. We will present DeepDive and the data that we use for this project, we explain the relations we focused on so far and explain the implementation and pipeline, including our model, features and extractors. Finally, we detail our results with a thorough precision and recall analysis.

Wiki-research-l mailing list

Wiki-research-l mailing list