Recovering Wikipedia-deleted articles started
In September last year, I announced the intention of the employees of the open-source wiki project “ALL” to begin recovering data deleted on Russian Wikipedia. Until that moment, deleted articles, pictures, templates, and some pages of interest were downloaded to independent repositories.
Since ALL is an encyclopedic project, administrators save there not every spam or self-PR of contact users, but articles about real people, events, companies. To begin with, the selection of articles was carried out according to a certain algorithm that eliminates obvious vandalism.
For example, many articles on a fictional universe have been restored.
According to the bot programmers, the algorithm was as follows.
* An article with that name is not currently on Wikipedia (i.e. it was not recreated as a separate article - only as a redirect);
* In the comment on deleting pages, there is not one of the keywords like “vandalism” or “copyright infringement”, which show that the page most likely does not represent any value.
The bot worked last fall, selecting a list of about 100 thousand articles satisfying these conditions.
At the beginning of this year, the fill bot was finally launched. He slammed ALL articles - it received more than 2 thousand deleted articles from the Russian Wikipedia. At least several thousand more interesting articles are on the way. You can read their full list at the link above. However, these articles do not exhaust the unique content of ALL: there are many articles aboutpeople , schools and, for example, iconic songs .
The articles have templates, categories and pictures.
Since ALL is an encyclopedic project, administrators save there not every spam or self-PR of contact users, but articles about real people, events, companies. To begin with, the selection of articles was carried out according to a certain algorithm that eliminates obvious vandalism.
For example, many articles on a fictional universe have been restored.
According to the bot programmers, the algorithm was as follows.
* An article with that name is not currently on Wikipedia (i.e. it was not recreated as a separate article - only as a redirect);
* In the comment on deleting pages, there is not one of the keywords like “vandalism” or “copyright infringement”, which show that the page most likely does not represent any value.
The bot worked last fall, selecting a list of about 100 thousand articles satisfying these conditions.
At the beginning of this year, the fill bot was finally launched. He slammed ALL articles - it received more than 2 thousand deleted articles from the Russian Wikipedia. At least several thousand more interesting articles are on the way. You can read their full list at the link above. However, these articles do not exhaust the unique content of ALL: there are many articles aboutpeople , schools and, for example, iconic songs .
The articles have templates, categories and pictures.