Re: [cvn-l] Help test new edit review tools

Hi all, Wikimedia Foundation's Collaboration team[1] is working on new tools that may make it easier to review edits, detect and fight vandalism, and be more welcoming to new users. The interface will be based on recent changes, but enriched with information from ORES [2], the machine-learning system that predicts things like "Does this edit have problems / did it damage the page?" and "Is this edit likely made in good-faith or bad-faith?". If you're familiar with ClueBot, ORES is a little bit like that. Based on that information, the idea is to provide different feeds for different purposes. For example, a feed for good-faith edits (e.g. the "welcome" feed), or one for bad-faith edits (e.g. the "Vandalism"). Details are still open for discussion, but we could for example hide edits from the Vandalism feed if they are damaging but likely made in good-faith (to avoid speedy-reverting a new user). Or perhaps to add them to the Vandalism review after a short delay instead of in real-time (to give users of the good-faith feed some time). Right now these tools are still in early stages and in the end we all want the same: High quality wiki content, happy editors, happy vandalism fighters. If you'd like to share your ideas and get an early look at what they're working on, e-mail Daisy with the following details: * Username. * City/Country or time zone. * Best time to talk to you. * Short description of your primary reviewing or patrolling activity. The testing should take about an hour. To participate, send your details to dchen(a)wikimedia.org with subject line "Counter-Vandalism". -- Timo [1] https://www.mediawiki.org/wiki/Collaboration [2] https://meta.wikimedia.org/wiki/ORES _______________________________________________ CVN mailing list CVN(a)lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/cvn