Datasets:

sil-ai

/

bloom-captioning

like 12

So I kind of need to get straight all of the following in my head:
what to call that thing we call a "book" thus far, which has its own page on Bloom, and when we download it it has a folder with a .htm and meta.json, and a title like "Chuskit goes to school"
what to call one of the languages in that "book". a "translation"?
what to call the same image-caption pairs, but located in two "books". e.g. the original english "Chuskit goes to school" but also the english translation which is included in the Russian "book"
This is all getting mixed up in my head because I originally just parsed everything to VIST format because I thought that was our only goal, before the scope expanded. WIthin VIST we've got...
albums which are sets of images, no captions. These have a uuid.
"stories", which are created when volunteers come along and add captions to an album. You can have multiple stories generated for each set of images. These also have uuids.
And I made some choices with uuids and such which is confusing all of this. uuid for the htm file, uuid for the album, uuid for the story