Grants talk:IEG/Wikipedia Massive Open Online Courses

This project has not been selected for an Individual Engagement Grant at this time.

We love that you took the chance to creatively improve the Wikimedia movement. The committee has reviewed this proposal and not recommended it for funding, but we hope you'll continue to engage in the program. Please drop by the IdeaLab to share and refine future ideas!

Comments regarding this decision:
We really enjoyed learning about this idea, and wish you great success if you move forward with it via other means this semester. If not, hope to see you back in a future round!

Next steps:

Review the feedback provided on your proposal and to ask for any clarifications you need using this talk page.
Visit the IdeaLab to continue developing this idea and share any new ideas you may have.
To reapply with this project in the future, please make updates based on the feedback provided in this round before resubmitting it for review in a new round.
Check the schedule for the next open call to submit proposals - we look forward to helping you apply for a grant in a future round.

Questions? Contact us.

Aggregated feedback from the committee for Wikipedia Massive Open Online Courses[edit]

Scoring criteria (see the rubric for background)	Score 1=weakest 5=strongest
Potential for impact
(A) The project fits with the Wikimedia movement's strategic priorities	4
(B) The project has the potential to lead to significant online impact.	4
(C) The impact of the project can be sustained after the grant ends.	3
(D) The project has potential to be scaled or adapted for other languages or projects.	3
Ability to execute
(E) The project has demonstrated interest from a community it aims to serve.	3
(F) The project can be completed as scoped within 6 months with the requested funds.	3
(G) The budget is reasonable and an efficient use of funds.	3
(H) The individual(s) proposing the project have the required skills and experience needed to complete it.	4
Fostering innovation and learning
(I) The project has innovative potential to add new strategies and knowledge for solving important issues in the movement.	4
(J) The risk involved in the project's size and approach is appropriately balanced with its potential gain in terms of impact.	4
(K) The proposed measures of success are useful for evaluating whether or not the project was successful.	4
(L) The project supports or grows the diversity of the Wikimedia movement.	3
Comments from the committee: Highly ambitious proposal. Proposer and collaborators seem to have the required skills and experience. A big project, but a seemingly managable one. It has a clear purpose that leverages an existing platform which make for a big plus. This could be the next big thing. Potential to have a big impact on getting new active editors, although perhaps not several thousand as stated. The English Wikipedia community seems at best to feel cautious about the idea of flooding the site with thousands of new editors, making the project seem risky. The cost and time accounting for this project is more vague we'd like for a large request, stronger justifications for time and costs should be provided. Course design and contents are not yet well-defined, the project may be premature. Well-attended courses on sites like Coursera inherit a lot of materials and experiences from well-attended courses in the traditional style, performed by the same lecturers. This project doesn't seem to have a lot of them - there are open-content materials about Wikipedia, but if the new course were to reuse those already available extensively, it could attract less attendees who would like to see something new. To complement, it would have to be considerable amount of research and design. Would like to see numbers in the measures of success.

Chat on IRC[edit]

Hang out in #wikipedia-mooc^connect on IRC if you want to chat about this proposal and Wikipedia MOOCs in general.

Comment, re. Coursera etc.[edit]

I like this idea in principle. Obviously there are many pragmatic issues to consider: the devil is in the details. But just one thing for now: Is there any chance of running the MOOC *outside* of the big behemoths such as Coursera et. al.? (I write as someone who is working on something of a DIY MOOC at present.) --Jbmurray (talk) 07:02, 12 February 2013 (UTC)[reply]

From my perspective, at this early point point I'd prefer to start with whatever MOOC platform gives us the most audience, which means Coursera or edX as the best options. But it's complicated by major institutional turf wars that are happening around MOOCs. Many universities (especially ones that started experimenting with online education before the current wave of MOOCs) don't allow faculty to do Coursera or the like, because they have their own systems they want to develop.

For a Wikipedia course, I think having peer evaluation as part of the platform is critical to doing it right. (It would certainly be possible to do a basic Wikipedia MOOC without peer evaluation capabilities, but that would mean serious limitations to the homework assignments we would want to try.) So far, I think only Coursera has this.

Although I don't think either of them have peer evaluation yet, edX and Class2Go probably will soon. Both are much more closely aligned with Wikimedia: they are non-profit, and allegedly open source (although I can't actually find any source code for edX). Class2Go in particular is a (very new) DIY-friendly platform; it's being developed by Stanford professors who were initially part of what became Coursera (before it spun off into a for-profit), and the philosophy is to cobble together as many usable existing open source components and/or open (although closed source) platforms like YouTube to get all the necessary components of a modern MOOC.

I'm open to a more DIY approach, and I think something like an independently-run ongoing Wikipedia MOOC is an important long-term goal. But for a first foray, being part of a well-supported technical system and having access to the huge pool of students on Coursera (in particular) is very attractive. If I can't find a good opportunity with one of the major MOOC systems, though, then I'd probably try to build a small-scale DIY course that could later be the basis for other MOOCs.--Ragesoss (talk) 11:35, 12 February 2013 (UTC)[reply]

To add a bit in response to your anything but Coursera comment at the en.wiki Village Pump... I agree with a lot of the sentiment there. I don't want for-profit, market-focused Coursera to 'win' the MOOC competition in the long run. (To that end, I put a very high value on the course content being free, so that it can be used on other platforms and built upon.) But I think the Wikipedia participation crisis takes precedence; I wouldn't hesitate to go with Coursera if it means 10x the number of students (potential Wikipedians) enrolling.--Ragesoss (talk) 11:54, 12 February 2013 (UTC)[reply]

Thanks for your response. I stand by my "anything but Coursera" comment, but am interested in what you say about Class2Go, with whom I am not familiar. Though I do also wonder why Wikipedia needs to go through any such body: it's not as though the world's number 7 (or whatever it currently is) website lacks visibility to attract potential students. If we use existing open-source components, and put the rest on wiki, what does even Class2Go offer? --Jbmurray (talk) 22:21, 12 February 2013 (UTC)[reply]

In some sense, we do lack visibility for this purpose. People don't come here to take classes, and using (say) banners to try to recruit students for a MOOC we were running ourselves would be pretty inefficient and generate some blowback. (Once we have good content and a good system for delivering it, that approach might still be worthwhile. But doing all that independently is a much bigger project.)

To your final question, Class2Go (which I've only investigated a little bit myself) offers, in essence, an open-source way to patch together other open-source technical components into a decent user experience for a large online course. (Class2Go isn't an organization we'd work with/through, it's a technical project to build an open-source MOOC platform.) Coursera offers, on top of that, a very large userbase likely to be rich in people with the disposition to become Wikipedians, technical infrastructure, and considerably more features (flexible peer evaluation, most critically).--Ragesoss (talk) 03:03, 15 February 2013 (UTC)[reply]

Unlike Jbmurray, I would only support this if Coursera was used, or possibly edX (with which I'm less familiar). If the WMF is going to be spending money on this project, the money has to be spent where it will be useful. We'll get more future value from every dollar spent using Coursera than any other site. Don't ask me for facts to support that, I don't have any, it's my personal opinion.Ryan Vesey (talk) 02:15, 19 February 2013 (UTC)[reply]
- - And, if I may say so, a rather odd opinion, too. --Jbmurray (talk) 20:01, 19 February 2013 (UTC)[reply]
- Good news from your perspective, then. As I just updated the grant proposal to reflect, it looks like we're likely to go with Coursera. The professor who wants to lead this is at University of Pittsburgh, which is already partnered with Coursera. We're interested in porting course materials to other platforms as well, though, later on. My feeling is that some people will find the structured timelines and student cohorts of Coursera-style classes work for them, but others will get much more out of the go-at-your-own-pace environments like Khan Academy. Having Wikipedia materials on as many online learning platforms as possible is a medium-term goal for me.--Ragesoss (talk) 03:39, 19 February 2013 (UTC)[reply]
  - I think this is a great pity. Sadly, however, it looks as though this is the direction Wikipedia is going. --Jbmurray (talk) 20:01, 19 February 2013 (UTC)[reply]

I have some experience running a MOOC[edit]

Hi, I'm an active member of WMFr and I'm starting a project management MOOC. First french xMOOC, enrollements started early january.

The problem is I'm very busy as you can guess, but if you need some advice get in touch... or enroll if you speak french. This MOOC's stats will be open data --Rémi Bachelet (talk) 11:31, 12 February 2013 (UTC)[reply]

Thanks! Good luck with your MOOC! If/when we get to the point of planning a specific course, I'd love to get in touch for your advice.--Ragesoss (talk) 11:38, 12 February 2013 (UTC)[reply]

Really?[edit]

I don't know much about MOOCs, but the idea of 100,000 people turning up on English Wikipedia because of an online course like Coursera seems... rather a lot. I'm very skeptical of this project, although I can't quite put into words the reasons why I'm skeptical.

There lacks a list of measures of success, for a start. And the 'sustainability' section doesn't deal with the increase in workload on the Wikipedia community. If you dump 100,000 people onto Wikipedia for a semester... who is going to clean up after them? —Tom Morris (talk) 13:22, 12 February 2013 (UTC)[reply]

Tom sums up my own concerns. The upside potential is clear, but isn't the downside fairly disastrous, if things go badly? Thousands of users overwhelming a small number of articles with limited existing dedicated editors, and little incentive to learn the rules? Even if they are, as you suggest, better motivated than the average student, their desire for a grade is still likely to be stronger than their desire to become a Wikipedian. How can the downside risk be managed? Mike Christie (talk) 14:28, 12 February 2013 (UTC)[reply]

[edit-conflicted] Thanks, Tom. You've cut straight to the central risks and challenges of this idea. (There ought to be a Risks and challenges section on these grant proposals; I'll add one to this one when I get a chance.)

Re:100,000, let me unpack that a little. A Coursera course with 100,000 nominally enrolled students would see more like 50,000 who actually participated in any way (i.e., watching the first lecture and/or doing the first assignment of, say, creating an account), and then the first few weeks have a rapid fall-off from there. 10-20% of enrolled students actually watching a significant portion of the lectures and doing the quizzes and assignments is at the high end, from what I've seen. I haven't seen any good data about this, but the few data points I've seen suggest that completion rates ranges from a few percent to the high teens (for courses with minimal requirements). So we'd be looking at more like 20,000 new editors, and may a few thousand who follow through to the end. Still, that's rather a lot.

The concept with the course design will be to focus on teaching the students about Wikipedia, without assigning them to do things that will require extra cleanup. So early on, there would be nothing outside of sandboxes that students are assigned to do on Wikipedia. There would be quizzes about various key aspects of Wikipedia that would form part of the basis of their grades, and peer evaluation of sandbox work that would form most of the rest of their grades. By the end (when the less serious students had stopped participating), we would start to introduce mainspace editing, focusing on cleanup of stuff that's already bad. So the main idea is to teach people what they need to know to become Wikipedians—if they so desire—without requiring anything as part of the course that will create big messes.--Ragesoss (talk) 14:40, 12 February 2013 (UTC)[reply]

As for measures of success, I see the central one as pretty simple: the number of students who become/remain active editors after the course ends.--Ragesoss (talk) 14:42, 12 February 2013 (UTC)[reply]

Do you have any numbers in mind for Wikipedian-to-student ratio? And admin-to-student ratio too. I think people are slightly worried given how much of a clusterfuck Education Program things with 30 or 50 students in one class can cause... a MOOC could likely increase that problem a thousandfold. —Tom Morris (talk) 15:08, 12 February 2013 (UTC)[reply]

In short, no I don't have numbers in mind. We operate generally on the assumption that new users (unless they are, say, being required to write major content to earn a grade like in the Education Program) have a net positive effect, that they will clean up more than they mess up (on average). Since these students would actually get a lot more background and training and explanation as they get started than the typical newcomer (but won't be assigned to do things beyond their skills in mainspace), the goal would be to actually reduce existing problems while not creating significant new ones. Looking out for major problems (and reorienting or stopping the course if any start to appear) would be a key task of the 'running the course' phase. Basically, it can't be done in a way that has net zero or negative excess cleanup burden, then it's failing.--Ragesoss (talk) 15:19, 12 February 2013 (UTC)[reply]

I do rather share these concerns. This is a quantum jump even from the current EP, and quite a risk. (On the other hand, I agree that if anyone might do it, it'd be Sage.) But this might be a reason to start small? Hundreds, rather than thousands? --Jbmurray (talk) 22:25, 12 February 2013 (UTC)[reply]

The more new users, the greater the benefits, provided they come slowly enough that we can teach them--either by the normal editing procedure, or AfC , or class instruction. I agree that, as for any new user, it will be much more practical for them to correct articles and add references than to write articles, with the hope that they might continue and graduate to article-writing. But this only applies if they do edit mainspace; I do not see the benefit to us of having them only edit sandboxes if they are not going to go on to edit articles; the only reason to write in sandboxes is in order to practice before writing in mainspace, so I do not see the reason for a course that does not have this goal. There is one exception--the ability to find references will help their general research skills, and I can see this as a stand-alone project--but wouldn't it be better if we actually made use of the material? Nor do I see the purpose in an academic setting of learning how to make use of WP-- this should be just a component of learning how to make use of information resources--particularly more reliable ones than WP. In any case, I really doubt that learning the mechanics of our edit functions will help in any other environment, so I hope it is intended to wait until the visual editor becomes fully functional. DGG (talk) 03:00, 13 February 2013 (UTC)[reply]

DGG, the idea is more that the course would prepare them for article editing, but let them move to editing on their own (those who wish to) rather than trying to have everyone taking the class edit. It's not so much (not much at all, really) about learning the mechanics of editing. It's about learning how Wikipedia works, what's going on when people argue over X or Y, and so on. Many people taking a class like this will not even be interested in editing, and are instead trying to understand Wikipedia for other reasons. But some portion will be 'natural-born-Wikipedians', and a class like this will let them learn enough about it (and have a supportive group of peers to learn with) so that they don't get scared off by what's become a rather hostile environment for newcomers.

I think it will be worthwhile later on to experiment with more editing assignments (starting small, and figuring out how to do incrementally more significant editing assignments without causing messes), but for a first MOOC I think that introduces too much risk.--Ragesoss (talk) 02:01, 15 February 2013 (UTC)[reply]

Page on meta[edit]

See also Massive open online course using Wikipedia. I will comment more on this proposal later. --Ezalvarenga (talk) 13:42, 13 February 2013 (UTC)[reply]

Thanks! Sorry I never started hacking on that page earlier... it took the grant deadline to finally get me fleshing this out on-wiki.--Ragesoss (talk) 21:46, 16 February 2013 (UTC)[reply]

Similarity to the VLE project[edit]

Is this similar to Wikimedia UK's Virtual Learning Environment? If it's markedly different, could there be some overlap - perhaps some content can be re-used to save on costs? Chase me ladies, I'm the Cavalry (talk) 14:45, 13 February 2013 (UTC)[reply]

I just learned about this the other day when someone pointed it out in response to this proposal. Certainly, if some content could be re-used, that'd be great. Is any of the VLE content live and browse-able? I wasn't able to get much of an idea of what the content will actually look like from that page. Also, what's the current timeline for it?--Ragesoss (talk) 02:04, 15 February 2013 (UTC)[reply]

All of the content (and the software) can be re-used because WMUK is making it available under a free licence. It's effectively in a closed beta stage right now, but I'm sure it could be arranged for you to have an account on the current site if you wanted to browse around the content. I should warn you that one of the results of integrating a wiki with Moodle is that the content is intended to be collaboratively improved and extended, so it may be subject to change until a version is 'frozen' for live deployment. The timescale is not fixed yet for going live, as the VLE will be a closer match for your future goal of courses where students work through modules independently at their own pace - so there's no deadline of a term/semester start to constrain it. It would be good to explore the possible synergies between this proposal and the VLE. --RexxS (talk) 20:19, 21 February 2013 (UTC)[reply]

Great! I'd like to get an account to explore how it's coming along.--Ragesoss (talk) 12:40, 22 February 2013 (UTC)[reply]

Tightening plan and scope a bit[edit]

I'm interested to learn if a MOOC would work well to attract, engage, and retain new editors! I think tightening the scope and plan a bit might help you really test if this strategy works, and improve your chances:

It sounds like you're thinking the project will take a team to pull off, not just one individual. Any grantee accepting a portion of the funds (you mention dividing based on hours) will need to be named as definitely "in", so even if you don't know the professor yet, it would be best to figure out who would do what on the support team, and whether they would be included as a volunteer participant (which is quite flexible) vs member of the grantee team (commitment that you would be requesting funds for to plan the course, then produce it...potentially including a professor fee for the unknown instructor?) as soon as you can. [Siko]
- I don't think it's feasible to lock in a team before knowing which professor and platform we'd be working with. (The professor—or rather, their institution—will likely determine the platform, since the big MOOC organizations associate with universities rather than individual professors individually.) Would it be possible to sort of put this proposal on hold for a few months and make a decision once we know those things?--Ragesoss (talk) 02:48, 15 February 2013 (UTC)[reply]
That would then help you figure out the budget a bit better, I'm guessing :-) Time and Spirit of the grantee team is not quite enough information to make a funding decision on, so you might look up the cost of the camera and any other supplies you'll need to produce the course (are there any fees associated with the platforms?), and based on a baseline guess of each grantee's time commitments divide up time and spirit a little more. [Siko]
- I could start laying out the details of the several different possible paths forward, which roles would be needed for each, and what sorts of time commitments for each role could fit within the budget. These vary widely depending on how the platform, the technical resources the professor's institution has, and the length of the course. (None of the platforms under consideration have fees, although going the indie route with Class2Go or the like would potentially involve some additional technical costs, or at least having a versatile sysadmin on the team and having servers we could use.)--Ragesoss (talk) 02:48, 15 February 2013 (UTC)[reply]
One of the things that I see makes it hard to lay out a real-ish budget is not being sure of the platform, duration of the course, or audience size you'll be dealing with. Why not pilot the first WP MOOC on a small scale to learn if it helps retain more editors than current methods, and if the return on cost makes it worth doing more and bigger MOOCs? Deciding to start with a short course and smaller audience also helps alleviate the risk of overflooding with newcomers if something unexpected happens, which would be nice to avoid. [Siko]
- The scaling aspect makes that tricky: cost and size sort of run in opposite directions. The platforms with the largest userbase are also the most well-developed and integrated, and so could be bootstrapped most easily. Short course length and less ambitious content (ie, fewer videos to write and produce) is one way to start small, but that's in part dependent on the professor and how the course fits into their teaching schedule. Going the asynchronous route (Khan Academy-like courses, where students go through at their own pace independently, which has been brought up on the Wikimedia Education mailing list) would be another way to start gradually, focusing only on a few basic modules to begin with. (That would probably also have significant overlap with the WMUK VLE project noted above, if that project is close to what I'm guessing.) If we can't find a professor who is a good fit, that's the route I'm in inclined to take. But I'm hoping we can find a professor.--Ragesoss (talk) 02:48, 15 February 2013 (UTC)[reply]

Looking forward to seeing more details when you've given them some further thought, I'm curious to see where this idea goes. Siko (WMF) (talk) 05:44, 14 February 2013 (UTC)[reply]

And Wikiversity...?[edit]

I really don't understand why not develop this on Wikiversity. Rodrigo Tetsuo Argenton^m 21:45, 14 February 2013 (UTC)[reply]

The main reasons are that MOOCs are where the people are, and that wikis are simply not the right tools for a MOOC. The leading MOOC systems have hundreds of thousands of students or millions, and good ways for those people to browse available courses and sign up for them. Bringing new users to Wikiversity in the first place would require it's own major outreach project. The technical limitations of building a video-, quiz- and peer evaluation-based course on a wiki are also significant; all those things are possible on Wikiversity, but not in a sufficiently user-friendly way.--Ragesoss (talk) 02:18, 15 February 2013 (UTC)[reply]

The politically incorrect answer: because that would require that Wikiversity actually have a purpose other than being the last chance saloon of trolls. —Tom Morris (talk) 07:31, 23 February 2013 (UTC)[reply]

Too many people editing at the same time[edit]

Repeating what I said here, I do think we could use a etherpad as a test in such cases. I have already worked with 5 people in one document and it was extremely agile and productive. I am going to try this to a crowsourced translation on some articles and, after finihed the text, some wikipedians with more experience on Wikipedia syntax and rules, can wikify the results from the pad.

I also recommend an analysis of this research on what Wikipedia can tell us about the future of news, since when we have a lot of cases on the English Wikipedia where articles are edited hundreds of times an hour. On the Portuguese Wikipedia we have this on a smaller scale, but it also happens and the community deal considerably well with it. --Ezalvarenga (talk) 12:44, 15 February 2013 (UTC)[reply]

I think that, even with etherpad, having large numbers of people focused on the same article at the same time doesn't scale. Except perhaps at the level of small groups, I don't anticipate realtime collaboration to be a focus of this first MOOC. It's an interesting concept, and worth experimenting with later on, though. I do think the fun and excitement of realtime collaboration is worth taking advantage of, but in my experience it works well with like 5 people (rather than 50 or 500, not that I've tried). Breaking news of particularly wide interest (with a lot of independent coverage) is a different matter, and one that turns out pretty well with people each taking on one small bit of the emerging story, and some people working to weave it all together, but I think that would be hard build a class around.--Ragesoss (talk) 21:45, 16 February 2013 (UTC)[reply]

Beyond etherpad, I think we could experiment google docs. We have to begin to see how many people editing at the same time collaboratively can work well. And I think using something similar to google docs or etherpad can help us to set a good number on when too many is too many. And using such technologies (or maybe others I don't know or that will come) won't cause harm to Wikipedia if after the content worked by the students is brought to Wikipedia by experienced wikipedians. --Ezalvarenga (talk) 21:28, 18 February 2013 (UTC)[reply]

Etherpad is pretty good tech for collaborative editing... I suspect it could handle all the people you'd actually want working on something at the same time. And trying to find the right range of number of collaborators is certainly worth some attention. (Not directly relevant, but I was trying to take a Coursera course about online learning, and the professor was having students organize into groups on Google Docs. The course has been indefinitely postponed because there were so many students that it was crashing Google Docs when so many people were trying to edit a doc at the same time.)--Ragesoss (talk) 21:36, 18 February 2013 (UTC)[reply]

Educational Technologists[edit]

BTW, I 100% recommend getting some educational technologists on board. (Sadly, the EP doesn't even seem to know such people exist.) People like Jim Groom, Brian Lamb, for instance. --Jbmurray (talk) 18:34, 15 February 2013 (UTC)[reply]

Thanks. Sounds like a good suggestion. Could you give me an introduction to those two, if you know them? (sage

rageoss

com).--Ragesoss (talk) 18:38, 15 February 2013 (UTC)[reply]

You have mail. --Jbmurray (talk) 00:22, 16 February 2013 (UTC)[reply]

Well, you would have mail if that address didn't bounce... Do you mean sage

ragesoss

com? I'll try that. --Jbmurray (talk) 00:23, 16 February 2013 (UTC)[reply]

Yep, my bad. Typing with cold fingers. Thanks for the introduction!--Ragesoss (talk) 21:05, 16 February 2013 (UTC)[reply]

P2PU[edit]

Hi, and thanks for this very neat project idea. I'm curious if you've learned anything from the P2PU MOOC focused on bringing an article up to featured status: https://p2pu.org/en/groups/wikipedia-aim-for-featured-article/. It might make sense to review the approach and success of that course since although more advanced it seems to be the closest attempt at what you're trying to do. Cheers, Ocaasi (talk) 18:32, 18 February 2013 (UTC)[reply]

That course never really got off the ground. Matt Senate started it, and I was co-facilitating it for a short time, but I think both of us lost enthusiasm because of the combination of limited audience and limited platform. It's had a slow stream of people sign up for it in the meantime, but it doesn't seem to have had any measurable success that I've seen aside from a few dozen new users. That's not surprising, because we never really developed a plan for moving beyond the 'collectively choose an article to work on' step. The userbase of P2PU seems to be relatively small, although not insignificant; I haven't checked out how the platform has progressed since 2011, so it's probably worth exploring again. I think adapting a well-developed course to P2PU down the road makes sense.

In terms of things I've learned, I guess it's relevant that newcomers are unlikely to self-organize for a collaborative project on their own, without some active facilitation. (Whether they will do so even with active facilitation in a repeatable way is an open question.)--Ragesoss (talk) 20:06, 18 February 2013 (UTC)[reply]

Hi, I hope my feedback is not inappropriate here. As some may know we are currently running a P2PU/School of Open course on Writing Wikipedia Articles through through the Communicate OER project - happy to share whatever feedback might be helpful. We had 100 sign-ups or so, but only about 10 committed students on a weekly basis. This was never supposed to be a MOOC of course! We will run a second version for other time zones when this round concludes. - Snarfa (talk) 16:25, 1 April 2013 (UTC)[reply]

Copyright/plagiarism[edit]

Hi; I don't usually comment on proposals like this, but I saw it linked from en.wiki. Feel free to shoot me down if I'm being silly :) Now, from what I know of mass engagement projects in the past, one of the key problems has been copyright and plagiarism and I haven't yet seen a discussion of these. I understand that the motivations of editors taking this course would be good; however, plenty of good-faith editors have the ideas of copyright and plagiarism wrong in their heads - i.e. what is acceptable and what is not. There was a note on en that this was about "basic" tasks, but since I don't know exactly, could someone explain what copyright/plagiarism risks this course might face and how they are being minimised and/or dealt with? Grandiose (talk) 10:18, 20 February 2013 (UTC)[reply]

Aye, I know what you're referring to. I've seen some students say "I can copy this as long as I cite it, right?" or "I found this image on Google, so I can upload it, right?". In my eyes, I think it would be valuable to dedicate a lesson early on in the course to the proper use of sources and images. It's not just about minimizing risk -- the ability to concisely paraphrase a source without losing meaning is a skill that takes time tom develop. If this gets incorporated into the lesson plan, the students who do go on to contribute would be much more capable because of it. --Cryptic C62 (talk) 13:35, 20 February 2013 (UTC)[reply]

I agree with Cryptic: this is a valid concern, and certainly something to take very seriously in light of the India Education Program problems. I think I may have a partial solution: I imagine one of the major themes in the class (pending discussion with Dr. Farzan) will be understanding the social implications of the knowledge we construct on Wikipedia, covering the problems of: accuracy vs. slander, copyright, plagiarism and the relative value of new and minority opinions on Wikipedia. I have already taught a couple class sessions about these issues and have a list of readings, etc that work well for demonstrating the problems. Even if we don't turn all of our students into editors, at least they will understand what it means to create public knowledge and communicate Wikipedia's values about creating public knowledge. That seems a good enough (though hard to measure) outcome, especially if we are imagining this course as the equivilant of a college class: It gives them practical critical thinking outcomes which they will most definitely use outside the course alongside specialized knowledge outcomes, which they hopefully will use, but may not. Sadads (talk) 14:27, 20 February 2013 (UTC)[reply]

[edit conflict] In terms of the mechanics of the course, Coursera supports peer evaluation for assignment grading. So the way I expect we'll structure the assignments is something like this:

(After having seen videos and written warnings against plagiarism, threats of removal from the class, etc.) Students are assigned to put together in a sandbox a paragraph of well-cited material that could be added to an incomplete article.
Students go through a series of training exercises applying a grading rubric that involves explicit checks for plagiarism to example content. (I.e, they get shown various wikitext, and they need to identify which is good and which isn't.)
Once they are able to pass the training exercises, they apply the same grading rubric to 3-5 of their classmates' sandbox assignments.

So, every student's assigned edits get checked for plagiarism by other students, and every student learns to check for plagiarism in the work of others so that they understand the issue more fully.--Ragesoss (talk) 14:33, 20 February 2013 (UTC)[reply]

I suspect that it would be more productive to ask (as part of the course) why there is so much plagiarism on Wikipedia, and why it seems that plagiarism is particularly rampant among students who are editing Wikipedia as part of a course. (I'm not sure by the way that the latter claim is true; but constant discussion over the history of the Education Program certainly suggests that it seems to be true.) At present, too often Wikipedia treats plagiarism much as the academy traditionally regards Wikipedia itself: as something to be denigrated, indeed exorcised if at all possible (and it isn't entirely possible), but never fully understood.

Incidentally, I've said this before, but if one day there is to be a proper dialogue between and about Wikipedia and academia, then plagiarism is probably a pretty good topic from which to launch that dialogue. --Jbmurray (talk) 17:33, 20 February 2013 (UTC)[reply]

+1. A thoughtful examination of plagiarism, Wikipedia, and internet culture would probably make a good course module.--Ragesoss (talk) 17:59, 20 February 2013 (UTC)[reply]

Turnitin[edit]

I've been building relationships with the plagiarism/copyvio detection company Turnitin W:EN:WP:Turnitin. We currently have full access to their API to check any article for copyright issues and to generate full analytic reports. On English Wikipedia wider usage is dependent on an RfC and a proper trial demonstrating efficacy, but there may be some areas where we could collaborate in time for this course.

Cool. With my WMF hat on, I was planning to touch base about that Turnitin project. The education program team may be doing a plagiarism study soon (although using a different plagiarism identification tool). For the MOOC, I'd be interested in exploring the possibilities. One interesting possibility, building on Jbmurray's suggestion above, would be to have an assignment that involves students searching for and analyzing pre-existing plagiarism on Wikipedia.--Ragesoss (talk) 14:44, 21 February 2013 (UTC)[reply]

I'm interested in the tool you're planning on using for ESEP. Part of the evaluation for using Turnitin is the evaluation that it is among the best available options. If you have a good tool I'd be very curious to look into it, at least for comparison. I'll shoot you an email to kick around some ideas here. Ocaasi (talk) 14:17, 22 February 2013 (UTC)[reply]