[WikiEN-l] Newbie friendliness, markup hell, and editing structure

[Peter Ansell]

On 17/03/2008, Steve Bennett <stevagewp at gmail.com> wrote:
> On 3/13/08, Magnus Manske <magnusmanske at googlemail.com> wrote:
>  Since there is very little consensus over the "correct" ordering of
>  metadata, any tool which reformats metadata in some rigid format is
>  bound to step on some toes. Which is probably just an argument for
>  *reaching* some consensus on metadata formatting, of course.

There is very little consensus about people making a point of having
categories or interwiki links in a specific order or in separate parts
of the document. What ordered metadata can you describe, other than
the example with the comment that is used solely for ease of freetext
editing and could be better defined in a permanent section on the
discussion page? Unless there is any actual reason for wanting to have
one category declared between a template and something else, when
infact the position of declaration is completely meaningless, it seems
like a distraction to talk about it being a lack of consensus.

>  * Edit box for header stuff, containing {{hatnote}}
>  * Edit box for main text, containing "Some text <<ref>> / Foo"
>  * List box for categories, containing blah
>  * Edit box for footer stuff, containing {{template}}
>  * List box for categories, containing foo
>  * List box for references
>
>  That is, the wikitext is effectively decomposed into sections, and
>  edit boxes of the appropriate types assembled in order.
>
>  Someone is bound to mention that since it's javascript, anyone can
>  customise it however they want, but I think that's the wrong approach.
>  We should be trying to find a good solution that satisfies almost
>  everyone, and end up with almost all users of Wikipedia using it.

Still don't quite see what your point is with two interleaved list
boxes to accommodate a preference to define categories in a
meaningless order. As far as the rest is concerned, it would be nice
to have a place just for references, which may facilitate a movement
away from the unwieldy system of intext full citations that we
currently endure because there is no better alternative.

Overall looks promising to me.

Peter