After we’ve developed a functional prototype of the MediaWiki extension, we’ll drop a link to it here on our blog so that other interested parties can stress-test it.  Once we have a more complete version prepared, we’ll be working with partners from wiki communities that could benefit from LRMI in particular or schema more generally to encourage its adoption.  If you belong to a community that we haven’t reached out to yet and would like to chat with us, drop us a line at inquiry@untrikiwiki.com.

We’ve never made a single edit for which we had a conflict of interest on Wikipedia – ever.  Although we have advertised such a service, we’ve not aggressively pursued it – and we have not accepted any clients interested in on-Wikipedia work.  We’ve set up private internal wikis for organizations, and we’ve provided brief consulting on best practices to organizations who were interested in learning how to engage productively with Wikipedia’s community, but we’ve never made any COI edit on Wikipedia – even in user space.  We’ve made clear to all entities that have talked with us our ethical standards, and made clear that if they can’t accept them we can’t work with them.

We believe – strongly – that there’s nothing inherently wrong with accepting for-profit engagements that involve contributing to Wikipedia, as long as it’s approached in a transparent and ethical fashion.  We understand why it’s a controversial issue, but we believe that it’s a necessary and emerging field and believe that it’s important that people with knowledge of Wikipedia’s ecosystem move in to it and establish standards that protect Wikipedia’s integrity.

Starting now, and lasting indefinitely, we will not accept any paid conflict of interest Wikipedia editing work. To support this statement, we have removed mentions of the services from our website.  This isn’t because we think it’s wrong, but because we think it would serve as an unfortunate distraction to our current work and because we recognize that if we ever pursued paid editing as a service, we need to first publicly develop and declare a process that will be acceptable to Wikipedia’s community.  We think our currently intended process would have been ethical and more than meets the ethical standards of an overwhelming majority of Wikipedia’s community, but we made the error of not publicly talking about our process before posting a graphic about it in public.

We may choose to enter direct COI work eventually, but such projects are on-hold for the near  future.  If we do ever accept COI work, before we ever accept an assignment, we will develop an explicit set of standards by which we will operate, and we’ll post them publicly.  We’ll incorporate community feedback into the development of these standards.

If you wanted to understand our SXSW proposal in an image it you could use the one below:

If you wanted the same understanding as video, you could watch the one below:

And finally, if you wanted to vote for us, and secure this presentation’s actualization you can click on the image below. (You’ll have to make a SXSW account unfortunately).

Thanks
Max Klein @notconfusing Merrilee Proffitt @merrileeiam

What does this mean for the reputation of the biographies online? Their integrity is precarious at best. Let’s take a look at a snippet of the Wikipedia talk page for the Romney article below:

CC-BY-SA Wikipedians http://en.wikipedia.org/wiki/Talk:Mitt_Romney#Page_protection

The Romney article was page protected from anonymous editing which may seem like a bold move, but as the editors are say this is just another day in the life. While not all articles are as well read as Romney’s -  776,512 page views in the last 30 days – all are being vandalized just without the media fanfare.

How do you counter the ever present threat of Vandalism? As reader you should be checking the page history to understand the recent activity on a page. To see a tutorial on how to do so you can watch the “Forensics section” of a Webinar (starting at 31:00) one of our team recently gave to a group of Librarians.

Beware the trolls.

At 556 attendees strong the recent OCLC Research Webinars “Librarians are Wikipedians Too” and  ”Wikipedia and Libraries: The Connection” piqued the progressive, exploratory minds of Librarians worldwide. Conviced tech managers at independent research libraries asked for help to jump onto the Commons mass upload bandwagon. Reference Librarians started to dream up combined workshop / editathons, from the explanation of the two.  As well workshops and edithons the webinars outlined the 5 classical points of collaboration between the two communities, and how to forensically evaluate which areas of Wikipedia are fertile for Library linking.

• For the Public Library focused presentation you can now view the slide deck or webinar recording.
• For the Research Library focused presentation you can now view the slide deck.

A webinar is nothing without it’s audience and their questions.  We answered as many as we could at the time, but there were some more difficult questions to answer, which now clear of time restraints, I’ll answer in full.

Where to go next:

The answer of where to go next is somewhat of a mantra we hope to impose: “the wiki”.  The Wikipedia Loves Libraries portal is a growing base of related materials, ideas, and links to the subject. We recognize that using a wiki to get help with wikis can be somewhat of a contradiction, and have set up a simple form to get paired with Wikipedians in a more traditional way.

Unanswered questions from Chat:

Question from Bob Kosovsky to All Participants (02:54:43 PM):

Max: WP is 6th most used website; but acc. to visualizations I’ve seen, DPpedia is THE most used data source; can you talk about the implications of DPpedia being the MAIN source of data/information for numerous websites?

I think you’re referring to this image,

which shows DBpedia as the center of the Linked Open Data universe. DBpedia is a database of information scraped and infered from Wikipedia. It being this large has the implications that Google searches will be eerily smart, and occasionally possibly wrong. Beyond that it signals that despite some best effort to deride crowdsourcing as untrustworthy, the internet are utilitarian.

Question from Madeline Wagner to All Participant

I would like to know more about how “minority” views on a subject are handled : ie the recent article by a scholar who tried to edit the entry on the Haymarket affair.

This question leads to an advanced and philosophical design choice of Wikipedia. The controversy arond the Haymarkey affair on Wiki (chronicled here) highlights, that Wikipedia is not an encyclopedia of truth but an encyclopedia of proof . That is, by design, the facts that belong on Wikipedia are the ones that can be sourced, and true-but-no-provable statements aren’t valid Wikipedic content. Wikipedia is this way for practical reasons. For a full justfication read the essay “Wikipedia:Truth – A place for minority views.”

Question from Michele Combs to All Participants (02:58:36 PM):

Rule of thumb seems to be “no institutional WP accounts,” only individual ones so that there is a single responsible person for each edit; would you advocate permitting creation of institutional accounts for creation/editing so as to make edits more credible/authoritative?

Let us be pragmatic. It’s highly unlikely that Wikipedia would ever change it’s policy to allow group accounts, because if you are looking to make a user account’s edits more authoritative then we’ve lost the equity granted to anonymous users – a very historic tenet. To achieve a unity and community respect for a library’s editors as whole I’d suggest using a naming scheme in the vein of [name]+[institution]. For instance in my personal life I am User:Maximilianklein but when I edit for OCLC I use User:Maximiliankleinoclc which knots mine and my institution’s reputation.

Question from Kjerste Christensen to All Participants (02:33:07 PM):

If your library has a strong focus in a particular area, what about partnering with a WikiProject related to that subject area to look up information or scan media as needed?

This isn’t really a question at all but a fantastic comment. Click here to view the directory of Wikiprojects.

And remember — it’s  not confusing http://twitter.com/notconfusing.

Watch the video for an overview of the project:

Now the collaboration to edit 250,000 Wikipedia pages is growing closer. With data and permissions obtained from OCLC and VIAF, Wikipedia community approval is all that remains. After a warm reception at the Wikipedia Village Pump, the next step of the process commences – sitewide Request for Comment. It’s requested, so you may as well comment on this Authority control integration proposal.
.

Algorithmic metadata integration project between OCLC data aiding existing Wikipedia efforts.

Member service outreach and education.

That’s because most libraries have devised a way to avoid confusion between the two John Smiths born in 1927 that are still alive. The answer of course is – give everyone exactly one number. This is known as “Authority Control”. Yet building authority control systems did not ensure that two different authority control systems were compatible. It’s not their fault, it was a long time ago. Now that libraries are bandying together to build one number to rule them all, it seems worthwhile to point to on Wikipedia pages.

It’s reasonable not to expect there to be a universal map of everybody ever, which is why the current philosophy is to map relations. VIAF (the Virtual International Authority File) maps large institutional identifiers to large institutional identifiers, and then numbers the clusters. Now zoom out and realize that VIAF itself is only one clustering service. Wikipedia is another. Therfore, in applying the same design pattern, we should map between Wikipedia and VIAF.

That crosswalk is precisely what we plan to create. The proposal – which addresses myriad omitted details in this blog – is live now at Wikipedia’s Village Pump, so have your say.

The Pipeline idea is to integrate with Wikidata so that we can push interesting, authority controlled facts about people across all

language Wikipedias at once. Because after all, we all seem to be able to agree on who George Orwell supposedly was.