kerravonsen: (Default)
Kathryn A. ([personal profile] kerravonsen) wrote in [site community profile] dw_suggestions2017-02-16 02:44 pm

Take image descriptions from the image meta-data

Take image descriptions from the image meta-data

image hosting

Save time in creating image descriptions by taking them from the image meta-data, using that to pre-fill the image description form.

When one uploads an image to the DW image hosting (yay!) you are able to drag-and-drop or select an image to upload, and then you get presented with a form which has things like "Title" and "Description" in it, which you have to fill in. But a lot of my images already have descriptions in the meta-data (e.g. the "Comment" field in a JPEG file). It's a pain to have to type all of that in again when I already did it once. What I would like to suggest is that the image uploader read the meta-data from the image, and use that to pre-fill the form. The user then can edit that as they like, but if they're already happy with what's in their meta-data, they can just save what's there without changing it. This would also be useful if someone is using the old "upload by email" interface which used to be the only way of uploading images to DW. That interface could use the meta-data of the image to fill in the Title and Description information, which the user could edit later on the DW website. As for what meta-data to use, I think one could use the filename for the Title (that's what LJ does) and use the JPEG "Comment" field, or the "Caption-Abstract" field from the EXIF data for the Description.

Poll #18047 Take image descriptions from the image meta-data
Open to: Registered Users, detailed results viewable to: All, participants: 32

This suggestion:

View Answers

Should be implemented as-is.
16 (50.0%)

Should be implemented with changes. (please comment)
4 (12.5%)

Shouldn't be implemented.
6 (18.8%)

(I have no opinion)
5 (15.6%)

(Other: please comment)
1 (3.1%)

[personal profile] swaldman 2017-02-27 08:12 am (UTC)(link)
AIUI these don't have any way to embed metadata such as captions. (well, SVG is XML, so it certainly could contain metadata in a custom field, but I'm not aware of provision being in the standard. IANAExpert.)