Erik Moeller wrote:

...

It has been suggested on wikitech-l that the article count on the main

page should only contain articles that have been edited at least once after their creation. The implicit assumption is that the primary use of the count is as a measure of our collaborative, human success. By eliminating unedited articles, we would

- exclude most articles created by bots - exclude articles that have not had any kind of quality review by others.

This would not be too hard to implement. It's not a big issue like certification (more on that later), but it would still be nice to get some feedback before proceeding.

The quick and even easier alternative is to manually subtract articles from the count that have been generated by the bots we are aware of. <<

-----

I couldn't agree more, and I think this is actually rather pressing, given that the number appears to be approaching our goal. It *appears*, albeit misleadingly, that we have nearly 100,000 articles on the website. I know that's not what it says, but the number just jumps out at the casual reader. When that number ticks over 100,000 then I suspect people are going to start passing judgment on Wikipedia's level of quality: "They say they've reached their goal of 100,000 articles, but just look at the articles! Another waste of bandwidth!"

Of course not everyone will say that, and *I* wouldn't say that, but plenty of people might, e.g., on Slashdot or K5 or other influential forums. Let's not give them the opportunity.

Let's use Erik's modifications to cut the number of reported articles down. This will give the Wikipedia peer review project, and possibly other article selection projects, some time to start working. Then, if someone feels inclined to say, "You might have 100,000 articles but most of them aren't worth reading," we can reply, "There are review projects --join them and stop complaining."

Larry