[Foundation-l] Handholding for new articles (Was: Re: 80% of our projects are failing)

[Gerard Meijssen]

Hoi,
So if the question to any of these questions is not positive a person should
not contribute ?

I would strongly argue that when a valid subject is identified and a plain
text of one or two paragraphs has been written we already have a winner. You
still want wikification, you still want interwiki links, you still want
illustrations and you still want references. We call this a stub and stubs
are good and can be improved at a later date.

I understand where you are coming from, you want to see Pallas Athena rise
fully armoured from Zeus's head. The Greek Gods do not exist. Not all
contributors write perfect articles in one go. Requiring Gods to participate
will drive ordinary people away. The process of writing the perfect
Wikipedia article is not obvious and it takes time for people to become
comfortable with it. Some time ago I was asked to write an article on the
English Wikipedia on imho a valid topic. I decided against it because I am
uncomfortable with the straight jacket that is imposed on me.

So the conceptual question is, how do we want what to achieve and do we want
other people to participate ?
Thanks,
      GerardM

2008/12/2 George Herbert <george.herbert at gmail.com>

> On Mon, Dec 1, 2008 at 9:39 AM, Gerard Meijssen
> <gerard.meijssen at gmail.com> wrote:
> > Hoi,
> > The software has been tested but not all extensions are considered ready
> for
> > WMF production. I am establishing contacts with, among others, people at
> > UNICEF to make sure that we identify the outstanding issues carefully and
> > fix them efficiently. Given that the CreatePage extension requires
> changes
> > to the skin, it may make sense to consider using a superset of monobook
> (I
> > do not know how feasible this is).
> >
> > Given that the software is already being localised at Betawiki, we do not
> > need to restrict ourselves to English. I understand that UNICEF uses some
> of
> > their software in Swahili :) I would love to consider Swahili for this
> ...
> > Kennisnet is interested in this functionality, that would make Dutch an
> > option. It needs to be clear that it is not only Wikipedia projects that
> > will benefit.
> >
> > The benefits from a more useable interface have little to do with a
> "simple"
> > approach. Newbies are not able to contribute. Our need for more
> contributors
> > and content is most dire in our smallest projects. Personally I am not
> that
> > interested in using "simple" as a test environment. From my perspective,
> it
> > should be there for all the projects that want it. Obviously, when this
> > extension is localised first, it will be more effective.
> >
> > When we are to test this in a Wikimedia Wiki, we need to get involvement
> > from Brion. It would help a lot when the WMF actively takes part in this
> > collaboration and make usability a priority.
> > Thanks,
> >      GerardM
>
>
> Thinking about this... (and catching up in thread...)
>
> There are two levels of failure with new pages on enwiki now.
>
> Level one is technical - UNICEF study pointed that out, your
> extensions are approaching that problem.
>
> Level two is more conceptual.  Does a person who wants to create a
> page understand all that a "well done" page in Wikipedia should have?
> Can they explain what the idea is, and why it should have a page?  Do
> they understand references and think about how to provide some?
>
> To be really useful, a toolset that structures a "create page" button
> response should address some or all of these questions.
>
> Have the output be not just a page, but a series of pages, which
> provide short inputs and do some useful things with them.  Perhaps,
> for example:
>
> "Wikipedia is an encyclopedia.  It exists to collect useful general
> information about all topics and make it freely available.  But there
> are lots of things which don't belong in encyclopedias.  Are you sure
> that the topic / article you want to create is really an encyclopedia
> article?  Is it a word definition instead (link to Wictionary), or an
> image of some sort (link to commons), or (fill in some more).  If your
> idea for an article is really an encyclopedia entry, click 'Yes' below
> to continue."
>
> "Can you explain what this page / article will be about?  What's the
> topic?  Where did you learn about it?  Please fill in the text box
> below with your idea of what this new article is about.  This will be
> posted on the article's talk page to explain the purpose of the
> article."
>
> "Wikipedia relies on outside references to verify information people
> post here.  Can you provide the titles of some books or magazine
> articles, website URLs, or other sources which confirm what you are
> saying in the new article, in the text box below?"
>
> "Wikipedia would like to have articles about all important and useful
> topics, but some topics (normal people, most small businesses, etc)
> just aren't important enough.  Is your article something which people
> in other states or countries will find interesting and useful?
> Wikipedia has some policies on what we recommend as being notable
> enough for articles (link to policies).  If you think this article
> idea is notable enough, please click 'Yes' below to continue."
>
> "Wikipedia likes to have links from article to article.  Are there
> other existing articles which you think this new article should
> connect to?  List them below if you know of any."
>
> "Wikipedia article start with a short introduction, then more details.
>  Can you summarize what this article is about in one to three
> sentences, to start the article's introduction?  Think about it and
> then fill in the introduction below if you can.  Then click on
> 'Continue'."
>
> "Ok, now let's create the actual article contents.... " (filled in
> template article, with introduction section inserted, and slightly
> textually processed references and see also sections).
>
> And the final step drops the article rationale entry into the talk
> page as well, on article creation.
>
>
>
> Does this process make sense?
>
>
>
> --
> -george william herbert
> george.herbert at gmail.com
>