[Foundation-l] It's not article count, it's editors

George Herbert george.herbert at gmail.com
Wed Sep 23 04:19:55 UTC 2009


On Tue, Sep 22, 2009 at 8:50 PM, Erik Zachte <erikzachte at infodisiac.com> wrote:
> Thanks for all feedback,
>
>
>
> To be sure, I do not think participation level as a metric makes other
> metrics obsolete. My blog post can be read as ‘forget article count, embrace
> participation instead’ but that pushes things too far.
>
>
>
> There are other metrics that bear witness of our accomplishments, but are
> more meaningful than article count (which will always be a nice trivia).
> Examples are: article views per hour, unique visitors, percentage of
> potential audience reached (unique visitors per million speakers). All of
> these seem better than the static 'article count' because they focus on
> whether and how much our content is being used, in other words on the
> relevancy of our work for the readers.
>
>
>
> I can see that for outward communication we still need to emphasize somehow
> that we do matter by telling what we achieved.
>
>
>
> I would hope that internally we focus on metrics that point to the future,
> to what is yet to be achieved, and what can serve as inspiration. In that
> context participation level has its place.
>
>
>
> Percentage of potential audience reached (see above) shares a characteristic
> with participation level, namely that the largest languages don't
> automatically get all attention. At the other hand any ordering scheme that
> lets the 'Volapüks' of this world take top rank is putting the horse behind
> the cart. By the way right now we cannot yet measure unique visitors per
> project ourselves.
>
>
>
> ----
>
>
>
> In summary I would suggest: let us downplay article counts in future
> external communications, and present our achievements in a way that
> emphasizes how we matter to the public.
>
>
>
> For ourselves let us celebrate language communities that thrive, and focus
> on building communities where they are absent, or small compared to their
> potential, the rest will follow.
>
>
>
> ----
>
>
>
> I would of course welcome advanced analysis of relations between metrics. My
> hunch is that advanced analysis will yield interesting but complex
> dependencies, and possibly produce multi-factorial composite metrics, which
> will be less suitable as primary ordering principle, as they are above most
> people heads.
>
>
>
> Compare economics. News media present simple metrics like inflation rate,
> unemployment rate, gross national product. Each of these is too broad to be
> useful for advanced economic analysis, yet apparently best abstraction level
> for general discussions.
>
>
>
> Erik Zachte


It is an interesting and useful statistic, which I think has been
somewhat glossed over in the discussion.

The significance of it - and similar variations - is probably going to
take some time to settle in, as to how much we (the community) and
outsiders (the media, etc) feel it represents a useful metric to
describe Wikipedia and its relationship to the world.  But I see value
in having it and I think it's a significant and useful number.


-- 
-george william herbert
george.herbert at gmail.com




More information about the wikimedia-l mailing list