[Foundation-l] Newbie recruitment: referencing

[Liam Wyatt]

2011/11/3 David Richfield <davidrichfield at gmail.com>

> A tool which pops up asking for a URL, author and date would be a rich
> source of bad references.  We should rather be looking at ways to get
> references to books and journal articles.  Web references should be
> the exception rather than the rule, because the vast majority of
> websites are not WP:RS.
>
> How about a wizard-like tool which asks "did you read this in a book,
> in a newspaper, a journal article or on the web?" and if the answer is
> "on the web" asks the user how they know it's true.  Compare for
> example Commons's image uploader.  Users who care about references
> should be taught how to extract good refs from Google Books and Google
> Scholar - both quite easy to use.  If you paste the ISBN of a book
> into Citation Expander, it fills in the whole citation for you, and
> the same for pubmed IDs.  Now we just need a tool which will do this
> for major newspapers on the web.
>
> --
> David Richfield
> e^(ði)+1=0


I like this approach, but it is also possible to do this the "other way
around" - that is, on OTHER websites which are frequented by well educated
and (potential) good quality Wikimedians....
Let me explain...
Click on the "cite" button near the top-left here:
http://trove.nla.gov.au/ndp/del/article/628050
Scroll to the bottom here:
http://www.powerhousemuseum.com/collection/database/?irn=207936
The first of these is the National Library of Australia's Digitised
Newspaper collection, the second of these links is the database of objects
in the Powerhouse Museum in Sydney.

They have both now implimented across their entire database a system to
automatically generate correctly formatted citation code to that specific
newspaper article/object that the reader can copy+paste into Wikipedia.
This both legitimises "working on wikipedia" to their users and also
increases the likelihood that their content will be used by us - a win-win
situation.

I believe that the kind of people who are spending their time looking at
museum catalogues and looking at very old newspapers are EXACTLY the kind
of people that we would like to encourage to work on Wikipedia adding
citations. Furthermore, the "generate the correct citation code" clearly
already works (e.g. there are 6 such citations used here
http://en.wikipedia.org/wiki/Sydney_sandstone#References ).

Whilst the effort to move towards WYSIWYG code in MediaWiki continues
behind the scenes, perhaps it would be relatively easy for the WMF (or a
dedicated individual) to produce some documentation (and sample code?) that
clearly and easily explains to other similar organisations how to implement
this system on their own site. I can see this being particularly useful on
newspaper websites too. I imagine that this would be much simpler, cheaper
and faster to do than building a new Wizard/tool ON WIKIPEDIA because that
would require all sorts of community debates, browser testing, localisation
etc. etc.

Just a thought,
-Liam