Loading AI tools
Wikipedia editing bot run by Alex Bakharev From Wikipedia, the free encyclopedia
The bot looks at New Articles and puts relevant articles into the New Articles lists of Portals and Projects. The goal is to help editors find new articles related to their areas of interest.
This user account is a bot operated by Alex Bakharev (talk). It is used to make repetitive automated or semi-automated edits that would be extremely tedious to do manually, in accordance with the bot policy. This bot was previously approved but has since been retired, and therefore shouldn't be making edits that appear to be unassisted except in the operator's or its own user and user talk space. Administrators: if this bot is malfunctioning or causing harm, please block it. |
If you are interested in patrolling new pages, add importScript( "User:Fred Gandt/getUnpatrolledOfAlexNewArtBotResultsPages.js" );
to your account's common.js file before reading these lists. Requires the user to have the patrol right.
The bot is superseded by InceptionBot run by User:Bamyers99. InceptionBot still uses this page.
The bot currently supports:
When the bot runs, it scores a new article using the rules and if the score matches or exceeds the threshold, the article is added to the top of the new article list for a Portal/WikiProject. Articles already on a new article list are removed 14 days after the creation date. Articles moved into the Main/Draft namespace from another namespace will be reported as new articles. The new pages feed is checked in addition to the recent changes feed. Articles tagged with "mw-removed-redirect" are also checked.
The following namespaces are scanned for new articles/pages: Main (Article), Draft, Category, Template. New non-Main namespace articles/pages will only be displayed when transcluded into the Wikipedia and User namespaces. They will not be displayed when transcluded into other namespaces such as Portal. New non-Main namespace articles/pages can be suppressed from displaying when transcluded into the Wikipedia namespace by using the SuppressNS option in the rule page. See 'Create the rules' below.
The bot does not create an archive of previous new articles. The history for a SearchResult page can be used to view previous results. Viewing every 14th history item will yield the desired result. Example: JapanSearchResult. Click on the date for a history item to view the page asof that date.
Yes, you can add new feeds for the bot yourself. Here I describe how to do it. It is a little bit tricky, so if you are unsure what you are doing, ask the bot's owner.
Steps to create bot feed:
The name should not be used for any other feeds. It should not contain spaces and other non-letter symbols. It should be disambigous. E.G. UK is not good; is it Ukraine or United Kingdom? It should be reasonably short and you should be able to spell it uniformly throughout a few steps. E.g. Pneumonoultramicroscopicsilicovolcanoconiosis is not a good name.
Put the template {{Subst:User:AlexNewArtBot/NewFeed|FeedName|Portal Name}} in the appropriate subsection of the #Currently supported section of this page. Here the FeedName stands for the name you have selected and the Portal Name is the name of a Portal page with the feed (Portal:### or Wikipedia:WikiProject ###). The feed would create redlinks for the Rules, Search Results and Log of the new feed. The Portal Name will be linked automatically, so there is no need to put it in square brackets.
Well this is the most tricky part: You have to provide some rules for the bot. Each rule has some numerical value (might be negative). All the values from the rules applicable to the article are added together to get a score. If a rule matches the lead of the article the points for the rule are doubled. If the final score is at least as much as the threshold, then the article is in.
All rules must follow the requirements of PHP regular expressions.
Both threshold and rules are written in the rules page. One line per rule (and one line per threshold).
Threshold is specified as
means 14 points threshold
Rules has format:
The Points is the number of points for the rule, negative numbers are allowed. If omitted than by default every rule costs 10 points. Do not forget that if the lead is matched points are doubled. Pattern that we should have is a Regular expression as in Perl PHP that should be matched in the text of an article so the rule is counted. The inhibitors are the patterns that "inhibit" the rule. Making it inactive even if the Pattern that we should have is matched. E.g when creating the rules for the Russia related articles, I want to include Saint Petersburg - the second largest city in Russia. On the other hand many American articles mention St. Petersburg, Florida and other American cities. Thus, I could want to decrease the value of the rule and inhibit it completely if Florida mentioned:
In general the names of a country or of its capital are often mentioned in unrelated articles (e.g. somebody traveled there, etc.) But lead rarely mentions unrelated articles, thus usually we would want to have the name of the country to be below the threshold but above the half-threshold.
Categories are usually friends of the bot (if only all the new article writers used them!) so they deserve the cost above the threshold.
Note that \W (uppercase only) is needed to mark a word boundary. Without this by default a rule can match any part of a word.
The following symbols
{}[]()^$.|*+?\
must be preceded by \ to be taken literally. Otherwise they have special functions: * - wildcard, x? - optionality of the preceding symbol or bracketed string, (xy) - scope marking (e.g. for the purpose of | or ?), (x|y) or [xy] - alternatives, etc.
The pattern matches are case-insensitive (the letter case of the pattern and the article text does not matter). To turn off case-insensitivity, use (?-i:<text to match>)
ie. /\W(?-i:L)eena\W/
to only match Leena with an uppercase L.
For other inspirations look in the rules for similar newsfeeds.
There are a few magical words in the rule file:
Use the page at rule test to test the rules after they have been saved. This will check for rule errors. A wiki page name can also be entered to test the scoring.
Put into User:AlexNewArtBot/Master a new line with the name of the newsfeed (the same as the name of the rules file after the /). And that is all. The next time the bot works it will work on your feed.
Add {{User:AlexNewArtBot/<insert rulename>SearchResult}} to the new articles section of a Portal/WikiProject. Example: {{User:AlexNewArtBot/RussiaSearchResult}}
To display as a collapsed list use the following:
{{show |Bot detected new articles | {{User:AlexNewArtBot/RussiaSearchResult}} }}
For a scrolling list use the following:
{{scrolling window|link=User:AlexNewArtBot/RussiaSearchResult|height=300px|title=New articles}}
The display of Categories, Drafts and Templates can be customized by each user. This is done via CSS in the user wikipedia skin. See Help:User style for more information.
Copy and paste one of the following boxes of CSS code into the wikipedia skin. These can be customized to change the colors or not hide all or not indent all. .IncBotC = new Categories, .IncBotD = new Drafts, .IncBotT = new Templates.
CSS code
/* Hide InceptionBot new Categories, Drafts and Templates */
.IncBotC, .IncBotD, .IncBotT {
display: none;
}
CSS code
/* Indent InceptionBot new Categories, Drafts and Templates */
.IncBotC, .IncBotD, .IncBotT {
padding-left: 1em;
}
Sample
CSS code
/* Highlight InceptionBot new Categories, Drafts and Templates */
.IncBotC { /* Categories */
background-color: #cff;
}
.IncBotD { /* Drafts */
background-color: #fcf;
}
.IncBotT { /* Templates */
background-color: #cf6;
}
Sample
Seamless Wikipedia browsing. On steroids.
Every time you click a link to Wikipedia, Wiktionary or Wikiquote in your browser's search results, it will show the modern Wikiwand interface.
Wikiwand extension is a five stars, simple, with minimum permission required to keep your browsing private, safe and transparent.