JJDear83 / BHL-WIKI-Working-Group

A project space for the BHL-Wiki working group
MIT License
0 stars 0 forks source link

BHL Flickypedia Documentation #11

Open JJDear83 opened 3 months ago

JJDear83 commented 3 months ago

Story: As a BHL Image Curator, I would like to have comprehensive documentation on how to use the Flickypedia tool with BHL images so that I can populate the requisite metadata following best practices and aiding in downstream user search and retrieval.

Doc Link: https://docs.google.com/document/d/1Q-GRyE0GeCLDx_CSaMJbOLg7N068QalZgo-pZg-If4I/edit

User Acceptance: BHL-WIKI WG/Flickr Foundation Staff.

alexwlchan commented 3 months ago

Because this is part of the Flickypedia data model, I added an entry for BHL Page ID to the list of mapped properties in our documentation: https://commons.wikimedia.org/wiki/Commons:Flickypedia/Data_Modeling#P687

JJDear83 commented 2 months ago

Thanks Alex! I've updated this stub ticket with more info.

JJDear83 commented 2 months ago

At the JULY 15 meeting we decided we need an FAQ section that includes some of these example questions:

  1. I massively imported bad data, what do I do?
  2. Where can I go to ask a real person a question?
  3. What other policies or documentation exist that can help me?

@jorlowitz, @SiobhanLeachman , @trnstlntk please weigh in on questions and where we might point to answers. Thanks!

trnstlntk commented 2 months ago

I would do a slightly different approach and FAQ, if this will be published(*). Wikimedia Commons is not Wikipedia and it has other customs. With Flickypedia it's very difficult to do large scale uploads, as the tool really asks you to do granular and precise manual one-by-one edits on a small scale. So the questions should take that situation into account.

Before a FAQ, it's best to make sure that the "positive" tutorial is brief and clear on best practices and gives a few clear "best practice" examples. I've had good experiences with that with the OpenRefine and Wikimedia Commons howtos - I've seen less mistakes done by uploaders after these howtos have been published.

If folks don't read those / miss some instructions, I expect the following problems to occur most frequently:


(*) P.S. I have missed the previous discussions and decisions around Flickypedia for BHL. I think Flickypedia, as it is currently designed, is a suitable tool for very small-scale transfers of typical Flickr photos to Wikimedia Commons (i.e. contemporary digital-born photographs created and posted by Flickr users). It is, on purpose, not built for larger-scale uploads beyond, say, half a dozen images at the time. The tool encourages deliberate choices for specific individual images. It does so by introducing friction, nudging/forcing uploaders to write and actively look up new file names, captions and descriptions (not copying from Flickr itself), and manually entering categories. All one by one. Even uploads of a dozen of files thus take quite a bit of time, and I don't expect a lot of people to go through this process at scale.

If performed for BHL illustrations, Flickypedia-to-Commons uploads will not come with the accurate structured data that this project would like to have (i.e. taxonomic names matched to Wikidata items).

So, my thoughts around the role of Flickypedia in this project: I think it's good to offer Flickypedia as one of the possible options for very small scale "correction" and "addition" uploads after large batch uploads have been done by the WiR (with more accurate data and at scale). But I would not promote it in this situation as a tool to be widely used by the BHL or Commons communities. I think there are better tools for this particular task. But maybe I misunderstood the intentions behind the role and involvement of Flickypedia here?