Commons talk:Structured data/Computer-aided tagging/Testing

Juandev[edit]

Cool feature!

it would be nice to merge manual adding of the tags with automatic selection. e.g. I got this image and automatic tag didn't come with the cross. So it would be great to have the opportunity to add it in this stage manualy
Are you sure, you don't want to elaborate this technology more. E.g. for this image it does not provide a specific item for Buckingham palace, even we have it in the filename.
will we provide any data to Google, that they can improve their intelligence?
would it be possible to move in front and back by using keyboard arrows?
the AI sometimes surprises, but very often I would not choose one of the tags. I think we need to provide images and train our/their AI better and/or mix it with some text recognition - often the file has the subject of depiction in filename or description and also you can retrieve a lot of valuable information from the coordinates.
I got the error message "Something went wrong and the tags cannot be...Please try again later" multiple times. Web-version FF.
I am getting system messages and tags in three languages. Would it be possible in the beginning to provide links to the sites, it could be translated?--Juandev (talk) 01:46, 15 November 2019 (UTC)[reply]

web version, FF

Do Popular images appear for you? - yes
Do tags change states from suggested to checked tags? - yes
Do they publish/ buttons work properly? - yes, but sometimes it won't safe (see above)
If you uploaded any images, do they appear for you in the popular images tab? - not tested yet
Can you X out of the confirm tags dialog? -yes
Does the skip button work? -yes
Are any suggested tags confusing, surprising, or inaccurate? - the confusion may arise at the moment I don't know to which wd value the tag will be translated. Surprising are tags like "home". --Juandev (talk) 01:46, 15 November 2019 (UTC)[reply]