User:Mu301/Learning blog/2016-02

Promoting and improving Wikiversity
Note: this started as a discussion on my talk page. I've copied it here to continue the conversation. --mikeu talk 21:23, 8 February 2016 (UTC)

Guy vandegrift
In retrospect, I think I should have answered Miked's comment about "promoting" WV's tolerance for weirdness here and not on the Colloquium. My father was a writer, not a great one, but good enough to coach my mother who wanted to write a book about my autistic brother. I grew up hearing about what makes for good writing, even though I myself struggled with all my writing courses. The collaborative writing on Wikipedia is at best mediocre, and almost always lacks focus. The articles are encyclopedic. Wikiversity does tolerate parallel efforts, forking, POV and in the past, so-called "research" that is unrefereed.

But let me come to the point and argue that you and I are almost certainly in agreement on the important points. You and Dave want a cleaner WV, with less attention to the clutter and perhaps even less clutter. Let me pose and answer number of questions about your efforts:
 * 1) Is it harmful, even if it means deleting pages? ... I say no (and you say no).
 * 2) Is it necessary? ...I don't think so (and you perhaps you think it is).
 * 3) Is it sufficient? ... I say no (and perhaps you say yes or maybe).

Even if we disagree, none of our disagreements have any relevance because I have no intention of interfering with your efforts, which I consider to be unambiguously beneficial. Now let's reverse roles. I want to do two things: (1) organize student writing on Wikiversity and (2) establish an online refereed Journal that credits authors.
 * 1) Are my ideas harmful? Not unless I promote them with too much hyperbole (which I probably do on occasion).
 * 2) Are they necessary? I think so, but I am not sure. I just can't think of how a smaller version of Wikipedia can do anything Wikipedia can't do (unless your goal is to make WV "ultra-organized", but good luck with that!)
 * 3) Are they sufficient? I have no idea.

A couple of details:
 * 1) Most WV resources have single authors, which make the bylines easy to write. This is not the case with the WP articles.  Refereed Journals all have this problem.  We can attribute the editor who condenses the WP article, gives it focus, and submits it to the Journal. People can argue and complain about how many original authors deserve to be mentioned.  If the Journal is a resource page equal to all others on WV, the complainers can create their own Journal.
 * 2) Of the three wikis (WP, WV, and WBooks), WV seems to make the least effort to highlight quality resources. I just say we do this democratically, in keeping with the sole wiki that permits POV in mainspace.  In other words we make the Journal an ordinary resource page, creating an environment where anybody can start a journal.
 * 3) I'm still struggling with how the referees will "vote". Perhaps it needs to be secret and off wiki.  Don't forget that if people don't like it they can create their own journal.

--Yours truly, Guy vandegrift (discuss • contribs) 04:59, 9 January 2016 (UTC)

Mu301
and - anyone/everyone else is also welcome to jump in

Re: clutter. Yes, I do feel that there are some important reasons for managing, reducing, or otherwise placing constraints on the "clutter." The first reason is to optimize the functionality of our internal search engine. It does little good for a new user trying to find a resource to get pointed to an empty page with a note that says "welcome, please help us create a resource about this ." It is human nature to get frustrated and give up when the results are unhelpful. There are even objective (scientific) studies of how to improved the usefulness of a website's search and/or navigation. I'll look into a about that to inform our efforts. But, my gut feeling is that "we've been doing it wrong." (See my blog post about the wormhole page, which may have even damaged our Google search ranking. It certainly didn't help.)

Kipple drives out nonkipple is a phrase used to describe how clutter attracts more clutter. While Wikiversity is perpetually a work in progress... we really need to do more to highlight our Featured resources and improve the Initial experiences that visitors have here. One possibility is to identify a carefully curated short list of the best recent resources and use templates on wikipedia to direct users here. Think of these efforts as an attempt to flip the kipple quote to "quality attracts more quality." But, it's difficult to find quality learning materials when there is an overabundance of kipple flooding the search results.

These are just a couple of my thought on this, I'll write more about it as I think it is an important topic to explore. --mikeu talk 22:17, 8 February 2016 (UTC)

Guy vandegrift (redux)
Comment added after I wrote this: Oops! I somehow found myself on this page and noticed the January 8 date. I failed to note the year. I have accidentally continued with a discussion we had almost exactly one TWO year(S) ago !!! Sorry. --Guy vandegrift (discuss • contribs) 02:46, 8 January 2018 (UTC)
 * Your concern about Google search is correct. That is how I search the wikis, and I have noticed a distinct prejudice against Wikiversity, even though some of our resources are better than their Wikipedia counterparts.  I am too busy to find a specific example, but if you ever need some, I can find you several in the field of physics education.
 * The idea of highlighting quality resources is good, but only if we think strategically. Our goal is to impress the maximum number of potential "customers", and focus on what they call the "high-end" customers (i.e. those with good taste.)  To maximize this impact, we should keep in mind the following:
 * Most of our these high-end customers won't go to the Main Page or any similar resource because they will not expect to see anything of interest. Is it possible to put a banner announcement at the top of each page they see, not unlike those used by Wikipedia requesting donations?
 * Our target clientele consists of specialists in wildly diverse fields. There is little chance that we will pick something of interest to any given reader.  Therefore, a mathematics resource needs to look attractive to history buffs, and vice versa.  I once heard that the friendliest referees of a submitted paper are those who are close enough to the field to partially understand the paper, but not so close that they can see the faults.
 * While we want to impress our readers with the fact that we recognize the need to clean up, we should present only one or two of our best resources. For that reason, I suggest only one or two links.  One would be sufficient, but two would emphasize that there is more than one quality resource.  The links need to be rotated as often as possible, but not so often that we display anything but the best and most attractive.  But, let's not burden them with a long "reading list".--Guy vandegrift (discuss • contribs) 02:30, 8 January 2018 (UTC)


 * Welcome back! Let's continue the conversation that we started so long ago.


 * No one but The Google truly knows exactly how their search results get ranked. But, I suspect the so-called "prejudice" that you've noticed is a direct consequence of the large quantity of low quality pages that we host. I'm simplifying it a bit, but that is basically how site ranking works. It ain't rocket surgery to make that connection. You keep asking "what's the harm of those pages?" The answer is right in front of us. See the graph at right? I just downloaded those datapoints from the Google Webmaster dashboard. Google has cut the number of our pages that it is indexing to ½ of what it was just one year ago. I project we'll hit zero sometime around the middle of the fourth quarter 2018. How much longer do you want to continue debating value of these pages to our mission?


 * We had about 225 clicks to our entire site from Google search in the last 90 days. That is a rather pathetic average of 2.5 per day. Ohio Youth Problems, Functioning and Satisfaction Scales (Ohio Scales) appears to have shown up the most in search results with 250 something impressions, but with only 47 clicks during that time frame. Our other pages don't come close to that. That one page accounted for 1/5 of our incoming search traffic. Google doesn't even register any external links to our site. Apparently there are too few for Google to even bother counting them. It's like we've been sucked into a wormhole where no activity can reach us.


 * As to your bullet points:
 * That is probably true, but still our main page should be up to date and showcase our best work. Someone landing at Parkinson's Symptoms (one of the most linked to pages last time I checked in 2015) might then click on the Main page to learn about our site. At some point we're probably going to need to make some changes anyway. There are a couple of things that break on mobile devices.
 * Thinking... I'm not sure this is an urgent consideration.
 * I've been playing around with generating dynamic content. See "Today's featured resources" in Public humanities for an example. There are seven sets of three quotes which rotate depending on the day of the week. Instead of flooding the reader with two dozen examples it selectively displays carefully choosen puil-quotes . This is one way to address your last concern and it also "spreads the load" of page views across multiple pages. I used the same undefined function that the Main page uses to rotate Featured content.
 * Any ideas on bringing our featured resources to the front of the store would be greatly appreciated. --mikeu talk 07:16, 8 January 2018 (UTC)
 * Update: in 2009 no resource on our site had more than 400 hits. The Main Page was an order of magnitude more popular.
 * 40,581 [32.98 %]: Special:AutoLogin
 * 4,465 [3.63 %]: Wikiversity:Main Page
 * 1,043 [0.85 %]: Special:Search
 * 392 [0.32 %]: Favicon.ico (incl. icon requests)
 * 315 [0.26 %]: Wikiversity:Browse
 * I don't have more recent data, but clearly it is a high priority given the attention it receives. One out of every 30 page views hit the Main Page. --mikeu talk 17:08, 8 January 2018 (UTC)
 * More recent update from Statistics/2017/12.
 * 112,704 - Wikiversity:Main Page
 * 52,907 - Special:Search
 * 22,971 - Special:CreateAccount
 * 19,280 - Principles of Management
 * The main page is still getting an order of magnitude more activity than any other page. --mikeu talk 00:40, 11 January 2018 (UTC)
 * With those hit numbers on the Main Page, we definitely need to ensure that that page is of high quality, no argument from me there. I surmise that your interest in Wikiversity is not based on what Wikiversity is today, but what it could be.  That certainly describes my interest.  If that is true, our goal is not so much to promote Wikiversity, but the use of wikitext in a manner that extends beyond its use Wikipedia.
 * As you do your Google investigations, you might want to compare Wikiversity with the w:Wikipedia:Education_program. It seems to be a parallel effort that is aimed exclusively at Wikipedia.  I do not see them as our competition, but a collaborators working to achieve the same goal, which IMHO is to promote wikitext.  My semi-humorous reference to "Making Wikiversity Great", actually referred to a strong belief that the academic world should be largely based on wikitext.  The ability to store all edits, include sister-links to vocabulary (and perhaps someday to quality focused articles on any given subject) is only part of my enthusiasm for wikitext.  I also like the convenience with which the CC licensing and the simplicity of its markup language permits one to recycle and reuse the work of others.  My (and possibly your) goal is to "Make wikitext ubiquitous" (link to wiktionary not intended to help you with your vocabulary, but to emphasize how wikitext permits us to write in higher level prose that does not require us to stop and explain every word or topic to which we refer.)
 * I see our poor Google stats as more of a symptom than the disease itself. The disease is that we have failed to make Wikiversity useful.  That is why my initial interest in the "cleanup" was based more on dividing Wikiversity into useful and useless portions.  I am, however coming around to the idea of simply cleaning it up.--Guy vandegrift (discuss • contribs) 02:09, 11 January 2018 (UTC)

Yes, I've been in some contact with WikiEd, but we really should collaborate more with them. They are very focused on wp, specifically getting students to write about topics that most regular contributors don't like women in science or the history of third world countries. I wholly agree that the Google statistics are a symptom, rather than the problem. However, improving our visibility with search engines could draw more attention and participation on our site leading to growth. The current situation is detrimental to that. --mikeu talk 04:20, 11 January 2018 (UTC)


 * I agree, but have inserted a much needed comma into your nearly flawless prose because I believe they do like women ; ) Guy vandegrift (discuss • contribs) 09:33, 11 January 2018 (UTC)