It seems that the extended outage caused by November's server move has had an unexpected effect...

From time to time, I use Google's "Advanced Search" feature to comb E2 for common typos and misspellings, which I then submit to the author and/or the editors for correcting. When I tried to do just this yesterday with "aquit", Google stared blankly back at me...

What is this "everything2.com" of which you speak, Gord?

Nonplussed, I tried a few other common misspellings: "definate", "monkies", etc... nuffin'.

Your search - thier site:everything2.com - did not match any documents.

Five months ago, that common error showed up in 134 nodes, and I hadn't yet begun the process of alerting the authorities to their presence. This meant two things:

  1. The content editors had scoured the nodegel for every instance of the typos I mentioned in my August daylog and wiped them out with due diligence.

  2. -OR-
  3. Google forgot they were there.

The second hypothesis being the easier to test, I plugged a few common words and phrases into the search engine:

Your search - jessicapierce site:everything2.com - did not match any documents.
Your search - "mr. t" site:everything2.com - did not match any documents.

Even a search for "monkey" yielded only six results, and as Segnbora-t's "Monkey Tile Box" clearly shows, E2 has a fascination with monkey nodes.

The only conclusion I could come to was that Google crawled the indexed E2 pages during the outage, found them down and removed them from its index. Tens -- perhaps hundreds -- of thousands E2 nodes used to be indexed (across at least four domains), but this number has been reduced to a fraction of that ("Searched pages from everything2.com for "everything2". Results 1 - 10 of about 9,930", even many of these lead to dead links and none of them have cached copies). Checks with other leading search engines have given similar results.

So what does this mean? It means that fewer users will stumble across E2 during the course of research and Web surfing. But what can we do? Two things, really: Submit E2 URLS to Google and other search engines for (re-)indexing and link to E2 nodes from other Web pages. If you have a blog, link to a node you wrote. If you participate in online forums, find ways to link nodes there, too. Eventually, we'll correct Google's temporary amnesia.

Follow-up: arieh pointed out that the site's robots.txt file currently prohibits Web spiders from crawling E2. This would prevent Google from reindexing the site if it found the pages down. I hope this changes soon.