User talk:Harej

From Wikipedia, the free encyclopedia
Jump to: navigation, search
Archived Discussions since 2004—
IIIIIIIVV
VIVIIVIIIIXX
XIXIIXIIIXIVXV
XVIXVIIXVIII

re: WikiProject_X[edit]

That's a nice idea. In fact I proposed a somewhat similar research piece two years ago (https://meta.wikimedia.org/wiki/Wikimedia_Fellowships/Project_Ideas/Wikiprojects_Success,_Failure_and_Impact_on_Content_and_Community) through I didn't have enough pull to get it through. Let me know if you'd be interesting in collaborating on this; I'd be happy to help out. As you may know, I am an active researcher of Wikipedia; my newest paper got published just now ([1]) and I have a piece about why editors burn out and retire in review (I'd be happy to share a copy with you as it seems related to the wikiproject research somewhat). You can find more papers of mine on Google Scholar. --Piotr Konieczny aka Prokonsul Piotrus| reply here 10:58, 2 October 2014 (UTC)

Piotrus, I am very pleased to see that we both are interested in this matter. How much experience do you have in querying Wikipedia's database directly and doing research based on the retrieved data? Harej (talk) 19:50, 2 October 2014 (UTC)
While I have worked with some larger datasets, my primary focus in on theory and qualitative understanding of the studied phenomena. The short answer to your question is 1) little and 2) some. Longer is as follows: if I need to query the database, I ask someone to help me with it, while I focus on the theory, lit review, and later, analysis. I am also a bit wary or relying too much on database queries. Big data is great, but micro scale qualitative studies (such as interviews or surveys) shouldn't be ignored. It is my experience (and/or opinion) that way too much of Wikipedia studies come from well-meaning computer scientists/other "hard" sciences majors, who crunch numbers, play with big data, use impressive statistics and formulas, and fail to either ground their studies in proper theory, or produce a proper analysis of what their number means, or both (just like the paper I criticized in my last Wikimedia Research Newsletter, which is a prime example of what I consider to be useless and unreadable work). Too often they seem like a "proof of concept" studies - proving that some data can be analyzed, but not explaining why it was done at all. The end result is often a paper presented at one of the CS conferences, or worse, a conference poster, and that's it - little benefit for the community (not sure about the scientists careers, I hear in CS field such conferences do mean a ot). In my field, conferences count for nothing career-wise, we have to publish papers, usually in SSCI, which means that they go through much more of a review, and I also try to write up my work so it is useful for the wider populace. Now, don't take what I write as criticism of big data or quantitative approach - I respect statistics (and have used regressions and such in some of my papers). Rather, it's a long reply to your question: if you are looking for a quantitative big data coding expert, I am not going to help you much. The assistance I am offering comes from 1) being an active participant of Wikipedia community, up to and including running several WikiProjects for many years and 2) having a record of publishing SSCI-journal Wikipedia research that often merges qualitative and quantitative aspects. I'd love to help with studying WikiProjects, as I think it's a very interesting, important and understudied area with which I have some first-hand experience, but if you are already covered in those expertises, and just need a big data coder to round up your team, than I am afraid I will not be able to help you with that. Cheers, --Piotr Konieczny aka Prokonsul Piotrus| reply here 02:58, 3 October 2014 (UTC)
Piotrus, though I am looking for a "hard" science person, I am happy to get all the help I possibly can. Obviously you understand the issue and you have a background that would help prevent me from making amateur research mistakes. Would you be interested in serving as an advisor to the project? Harej (talk) 03:42, 3 October 2014 (UTC)
With my Wikipedia volunteer hat, I am more than happy to provide any assistance in the spirit of helping Wikipedia. I'll try to read your project page carefully and give some suggestions over the next few days in that vein. With my researcher hat, I would like to ask what are the project goals in the realm of publishing? Are you thinking of a conference paper, or a journal paper, or both? As I noted, I have a decent track record of getting things published, so I'd like to help with that as well. This kind of research should be made as visible as possible; it would be a shame to keep it just on wiki pages (in the past some WMF research, while quite interesting and useful, got poor visibility because it was restricted to wiki pages only). --Piotr Konieczny aka Prokonsul Piotrus| reply here 13:14, 3 October 2014 (UTC)
That is a good question, Piotrus. I had thought of research as being done more for the sake of developing a product, rather than something that could be published on its own. But I think getting the research published in a journal would be an excellent idea—the issues we broach are relevant to broader questions of online crowdsourcing and community building. As part of that, I'd like to make sure that our approach to researching this issue is sound. Please let me know once you've read through the grant proposal and if you have any specific concerns. Harej (talk) 19:24, 4 October 2014 (UTC)
Given that you're asking for almost 25,000 dollars in grant money, unless those t-shirts you're raffling are gold-lined, I would expect nothing less than academic presentation in the wrap-up. Your starting point should be an extensive literature review of the many articles and papers already written on the topic of editor retention; being a decentralized open-source project, Wikipedia suffers extensively from duplication of thought, and your current proposal seems to me to be in danger of simply spinning one aspect of this particular wheel yet another time without substantial results. See for instance this paper, or (self-promotion alert) this essay. ResMar 23:46, 5 October 2014 (UTC)
User:Resident Mario: for that much money I would certainly expect serious effort into academic publication (I published all my research without any grants, not counting few hundred $ over the years for conference reimbursements, and such...). If the Wikipedia community is forking the money, they should receive as much as possible for their (our...) expenditure, and that should include a serious attempt to get this research seen by other academics. --Piotr Konieczny aka Prokonsul Piotrus| reply here 08:48, 8 October 2014 (UTC)
Harej, you wrote of "developing a product"; I'd assume "Design a prototype workflow and interface that can be deployed to improve other WikiProjects". I've finally found time to read your proposal carefully, and I think it is a valuable proposal. As an editor heavily engaged with numerous WikiProjects, I certainly agree this is an area that has been neglected by developers, and yet has potential to help with editor's retention. I do in fact would have quite a few ideas on how to do so (I've, in the past, redesigned the pages of WikiProjects Sociology and Poland, so I've tried to do what you are talking about in practice, too). Some additional comments: " hypothesis that WikiProjects help facilitate Wikipedia-editing in a given subject area by organizing contributors around a cause and by providing resources and social support" - this is a compound hypothesis and should be split into at least two if not three, and its components need to be operationalized. 2) You seem to omit a good source of data - Signpost WikiProject reports. 3) I'd suggest adding a metric of increased engagement with WikiProject tools and/or satisfaction (with WikiProject or volunteering at Wikipedia in general). Overall, I support this project, and I hope you get a grant. And as I noted before, I offer my help in getting the data published in a peer reviewed outlet (and no cost; I don't need any of the grant money, just your willingness to share the data (survey, interviews and developed quantitative/qualitative models) and preferably, include me in your discussions and co-write parts of the paper dealing with aspects I won't be working on but others will). --Piotr Konieczny aka Prokonsul Piotrus| reply here 08:48, 8 October 2014 (UTC)
Hello Piotrus! Thank you for reviewing the proposal and for participating on its talk page. How did the re-designs of WikiProject Sociology and WikiProject Poland go?
1) Good point regarding the hypothesis. I'll see if I can come up with more proper hypotheses, including corresponding null hypotheses.
2) I'll take a look at the Signpost reports. I imagine they factor into the qualitative aspect of the research, or do you have other ideas?
3) Increased engagement with WikiProject tools would be a worthwhile metric. It'd even be interesting to see if WikiProjects promote more editing of Wikipedia in general; that would be cool.
I am happy to include you as a participant and to share data with you. I would also like you to review any papers we write as a project, especially since you would know more about that than I would. Would it be published through your university or would you find another venue? Harej (talk) 23:28, 14 October 2014 (UTC)
Re 1) Nobody ever commented on WP:POLAND as redesigned by me, but nobody complained, either. I've never asked... could be worth a mini-survey. For WP:SOCIO, I was reverted after few months by another regular who thought my design was too confusing (even through I intended it to be very newbie friendly). What the newbies, or anyone else thought, is again unknown, as nobody else ever commented on this. 2) Yes, a good source of qualitative data. 3) Yes, but difficult to measure outside surveysinterviews, and even that is difficult - if you asked me if a WikiProject made me edit something more, I'd honestly say I don't know. Maybe, but there are so many other factors. Also, are we looking at "more total edits" or "shifting pattern of edits", i.e. a friendly interaction at WikiProjet XYZ may make me want to edit XYZ topics for a while, but I doubt they make me want to edit more total. But maybe I am wrong, after all I don't know what makes me edit exactly... that's a very subjective question. I am happy to be on board; publication-wise, I would aim for an SSCI-indexed journal if possible, with draft paper versions serving as a conference fodder in the meantime (Wikimania/sym, etc.). I hope to finish my current major project this year, so I could help with serious writing around, hopefully, XMAS/winter. --Piotr Konieczny aka Prokonsul Piotrus| reply here 04:22, 15 October 2014 (UTC)

This Month in GLAM: September 2014[edit]

This month in GLAM logo.png




Headlines
Read this edition in fullSingle-page

To assist with preparing the newsletter, please visit the newsroom. Past editions may be viewed here.

Subscribe/Unsubscribe · Global message delivery · Romaine 17:31, 11 October 2014 (UTC)

Wikidata weekly summary #128[edit]

Learning Quarterly: October 2014[edit]

L&E Newsletter / Volume 1 / Issue 2 / October 2014
Learning Quarterly

Frontpage: #DataCollection #Learning4Sustainability

Stay tuned
blogs, events & more!

Grow the Awesome:
WMRS Education Program

Leave your mark
on Meta!

If this message is not on your home wiki's talk page, update your subscription.

María Cruz, Community coordinator, Program Evaluation & Design (WMF) MediaWiki message delivery (talk) 17:59, 17 October 2014 (UTC)

Wikidata weekly summary #129[edit]