[Foundation-l] Verifiability: Constitution?

[Erik Moeller]

On 9/17/06, Andre Engels <andreengels at gmail.com> wrote:

> But how do you define 'can be sourced'? The only way that you can show
> that something can be sourced is by sourcing it. Does this mean that
> we should remove all unsourced statements from all articles? If so,
> there will be little Wikipedia left. If not, then what do we accept
> without source and what not?

There is currently a poll on the German Wikipedia whether new articles
that cite no sources should be deleted:
http://de.wikipedia.org/wiki/Wikipedia:Meinungsbilder/Quellenpflicht_f%C3%BCr_neue_Artikel

The proposal, translated literally: "New articles may only be created
if they cite sources.These citations should be listed in the edit
summary [*] and/or in the article itself. To cite the main sources for
an article, the sections 'Literature' or 'Weblinks' should be used.
(...) New articles without sources can be deleted without further
discussion through a speedy deletion request."

[*] The German Wikipedia calls the standard edit summary field
"Zusammenfassung und Quellen," i.e. "Summary and Sources".

The poll is currently 2:1 against. In this discussion,
[[m:Verifiability]] has been cited to argue that verifiability is, in
fact, not negotiable and that the poll result itself should be
ignored; articles without sources should be deleted anyway. In fact,
one administrator added it to the criteria for speedy deletion already
while the poll was running. I reverted that change, and oppose this
notion of verifiability.

While citing sources is crucial, so is NPOV, and so is the consistency
of the flow of argument within an article. Articles are gradually
improved, and problems are pointed out and identified, to fix them
systematically later. The idea that articles can magically appear as
perfect, feature quality texts is anathema to the wiki principle of
working and contributing in small chunks. I strongly oppose any
foundation-level verifiability policy that would make it impossible to
contribute small pieces of work. I would also suggest reading
Anthere's comments when the French Wikipedia reached 50,000 articles:

http://mail.wikipedia.org/pipermail/wikifr-l/2004-August/001911.html

Verifiability, "no original research", and so on, exist as policies to
help us create high quality encyclopedia articles. They are not and
should never be unquestionable dogma. When policy flies in the face of
reason and common sense, policy needs to be questioned, and quite
often, revised. I am frankly disappointed whenever people want to
support their arguments by asking for the help of Jimmy Wales or the
Wikimedia Board. This suggests to me that their arguments are either
wrong, or that they have not spent the effort to make them clear and
understandable.
-- 
Peace & Love,
Erik