Gregory Maxwell:
Why are people without computers or reasonable access to computers considered potential audience for editing a website?
Why are people whom are effectively illiterate considered potential audience for editing an encyclopedia?
Your second question is redundant, few illiterate people will possess a computer.
Agreed, participation level does not tell the whole story,
I'll quote my blog post: "The fact that age, literacy, standard of living (internet access), and political climate all influence the real potential of possible editors in a language are left out of consideration. For many of these factors good numbers are not available (worldwide demographics per language), or hard to quantify (effect of political climate) or changing rapidly (internet access in 3rd world)." Especially the last one relates to your first point (think mobile phones).
So the metric is far from ideal. But still rather practical. It paints vividly how we are not waiting for a few minor languages to catch up, but how we have not reached large parts of the world yet. See also strategy wiki: there are proposals to broaden our reach beyond online access, and of course important initiatives are ongoing already, like OLPC and other projects, were Wikipedia plays a part.
And the same argument of being too blunt and non discriminating can be used for most metrics. In fact I made it in so many words against article count, which treats all articles as equally important.
Erik Zachte
On Wed, Sep 23, 2009 at 12:23 AM, Erik Zachte erikzachte@infodisiac.com wrote:
Gregory Maxwell:
Why are people without computers or reasonable access to computers considered potential audience for editing a website?
Why are people whom are effectively illiterate considered potential audience for editing an encyclopedia?
Your second question is redundant, few illiterate people will possess a computer.
Not really, they are both factors and one or the other may dominate in some places. Possessing a computer isn't the criteria I'd impose, it's having reasonable access to one and there are many people who are illiterate who could be reasonably counted as having access even though they can't make much use of it.
Agreed, participation level does not tell the whole story,
[snip]
So the metric is far from ideal. But still rather practical. It paints vividly how we are not waiting for a few minor languages to catch up, but how we have not reached large parts of the world yet.
[snip]
It does but that vivid painting is not a relevant one, and as such it is potentially deceptive.
The reason "how we have not reached large parts of the world yet" is because access to Wikipedia is significantly influenced by things outside of Wikimedia's control and scope.
I think a reasonable argument can be made that Wikimedia's actions could not produce a statistically significant improvement in the penetration vs population metric; simply because the causative factors outside of our ability to influence are so large.
You sound like you've heard me making this point and you acknowledge that this way of measuring has weaknesses, but perhaps I did not state this bluntly enough: I think you're measuring something which is largely irrelevant to the activities of the WMF.
If you instead took each language as it was and simply said that we'd be doing well to increase the users of each language, I'd agree. But saying that the WMF base its decision making on a measurement of "not reached large parts of the world yet" is effectively equal to measuring our performance by observing that Wikipedia is only used on 1 of ~382 planets known to exist in the universe.
Failure to consider this leads to bizarre conclusions like "the lower birth rates in the developed world compared to the developing world decrease Wikimedia's success over time". Which is simply silly— we shouldn't ignore our progress just because it may be dwarfed by shifting populations.
Surely someone must have a respectable count of internet users by language that we could use for comparison? That would be a much better metric for our success today; while raw literate speaker numbers would be a useful comparison for what we could start reaching with non-internet mechanisms.
[snip]
And the same argument of being too blunt and non discriminating can be used for most metrics. In fact I made it in so many words against article count, which treats all articles as equally important.
I fully agree with you that article count stinks as a metric in many ways— but at least it is measuring something which is pretty much fully under our control.
And this is really the point I was trying to make: While article count is a bad metric because not all articles are equally helpful. Penetration vs Population isn't even bad, because it's quite possibly not measuring something which is generally useful to our decision making. (Except perhaps in the decision to chase non-internet distribution; but without literacy calibration and a comparison to a measure of online reach it's even of questionable value there)
If we had a goal to double the number of articles in some reasonable period of time we could do it. If we had a goal to double the penetration, ... well, unless the WMF changes its mission nothing it could do would get us there.
2009/9/23 Gregory Maxwell gmaxwell@gmail.com:
The reason "how we have not reached large parts of the world yet" is because access to Wikipedia is significantly influenced by things outside of Wikimedia's control and scope.
A dramatic demonstration of this: if someone in Beijing flips a switch tomorrow, and zh.wp becomes blocked, our potential audience changes by three hundred million (internet users) or a billion (speakers) overnight (depending if you count population or internet users) and our nominal penetration among Chinese-speakers would presumably collapse as a result.
Surely someone must have a respectable count of internet users by language that we could use for comparison? That would be a much better metric for our success today; while raw literate speaker numbers would be a useful comparison for what we could start reaching with non-internet mechanisms.
There's a couple of estimates on:
http://en.wikipedia.org/wiki/Global_Internet_usage
though they look a little dated.
Alternatively, users by country is reasonably well estimated, I think, and you could try estimating based on languages from that.
wikimedia-l@lists.wikimedia.org