User:West.andrew.g/2014 Popular pages

__NOTOC__

These 10,000 pages were the most accessed on the English Wikipedia during 2014.{{efn|This includes data from the year as defined by UTC time. This list is derived and aggregated from the raw data available [http://dumps.wikimedia.org/other/pagecounts-all-sites/ here]. See the README on that page for a more detailed description of how view counts are calculated.}} The first column is the numerical ranking.{{efn|Certain articles are inherently popular. Others map to cultural phenomena and news events. When an "unusual" topic appears prominently, sometimes a Google Doodle or Reddit thread is to blame. Automated views (i.e., non-human accesses) can also be a factor, especially when DDoS attacks occur or a script has been misconfigured. A Signpost article describes these catalysts in much greater detail.}} The second column is the article title; sometimes the "article" is not really an article, but a script/image/etc.{{efn|There is a best effort to link to the article, character encoding issues notwithstanding. Other times, a server message/error code will be erroneously "blue-linked". For example, the high traffic at "undefined" is the result of non-existent page requests – not actually visits to the undefined article. It is not possible to automatically determine this intersection of naming conventions.}} A number of subsequent columns enumerate article designations, as described in the legend at right below.{{efn|An article may be classified differently by multiple WikiProjects, in which case multiple class icons will be displayed. The presence of an x icon should be interpreted to mean "one or more WikiProjects have classified this article at level x".}} Aggregate statistics related to these designations appear below the table. The next-to-last column is the number of page views.{{efn|Note this includes only "desktop" views. Mobile view counts were enabled on SEP-28, but were not included in the view totals for this report (but were used in the analysis mentioned in a subsequent footnote). It was determined including counts from such a small portion of the year could be confusing, and extrapolating yearly estimates based on short-term and non-representative data would be inappropriate. Mobile views will be in next year's summary.}}{{efn|Those looking for statistics on a particular page may look [http://stats.grok.se/ here] (though [http://en.wikipedia.org/w/index.php?title{{=}}User_talk:West.andrew.g/Popular_pages/Archive_1&oldid{{=}}537389549#Differences_with_Henrik.27s_tool problems] have been noted).}} The final column is an estimate of the percentage of "mobile" (i.e., cellphones, tablets, etc.) traffic (versus "desktop") arriving at the article.{{efn|This metric answers, "how much of the article's traffic was mobile from SEP-28 (first mobile traffic availability) until the year's end". This metric is intended for determining "illegitimate" entries in the list. One can generally conclude articles with very low mobile percentages are unlikely to be the result of organic/human traffic. However, given only partial year data, our interpretation must be more cautious. For example, an article whose views were inflated via a bot in August would have an inflated view total, but the mobile-% might report normal numbers.}}

This report is produced in a manner similar to that of the weekly "top 5000". However, this report will omit redlinks and articles that were deleted during the year.{{efn|The list of titles for consideration were all NS0 "article namespace" titles obtained shortly after the new year. There are approximately 11 million articles in this set, versus the computationally vast 150 million that would need to be aggregated with redlink inclusion.}}. This report is generated by User:West.andrew.g, or [http://www.andrew-g-west.com Andrew G. West, in real life]. He is unable to address statistical queries regarding other languages/projects, but is particularly interested in academic collaboration regarding this English Wikipedia dataset.

A similar aggregation identified the most popular articles in 2013.