ferrbea
commented
8 years ago Best practices:
Use a relevant image that matches your text Pick the right file name for your image Make sure image dimension match the image size as displayed Reduce file size for faster loading Add a caption for easier scanning of the page Use an image alt text, title text is optional Add OpenGraph and Twitter Card tags for the image Don’t break the left reading line using an image Use images in your XML sitemaps
So in our case file name and alt tag are the most important data. Title is optional. I haven't found any issues about duplicated content.
official suggestions from google https://support.google.com/webmasters/answer/114016?hl=en
The best practice recommends to write title and alt tag with a real description of the image. This means that is unlikely that you find duplicates as each image has something different in it. This also means that our method of compiling data with post title works for featured image only. All the other images should have customised attributes.
In this discussion here https://moz.com/community/q/alt-tag-labels-that-use-near-duplicate-text-seo-no-no (even if not really new it's from 2013) they say it can be seen as keyword stuffing even if you add numbers at the end.
So, the best practice would be having the user fill the alt tag manually with some valuable text. If we want to follow this rule I think we should leave the automation for the featured image only and let all the others to be filled manually. We could anyway add some warnings when alt tags are missing or duplicated. And add an explanation of this change of mind. Or as this is not a huge SEO error we could leave everything as it is and add some warnings when alt tags are not customised by the user.
Determine which is the best practice: to have unique attributes with progressive numbering of the possible duplicates or to keep duplicates? Once determined which is the best practice, implement it directly into the code (and not via user settings).