It’s just work. Doable. Leaning into metadata, though, puts it on the user to have fully and properly tagged their images with metadata, which puts a lot of work on them. I’ve always wanted Backlight to do more, so that the user need do less. And it’s never about wanting to limit the user’s control, but about relieving them of unnecessary burdens in making proper use of the software.
Granted, over time, complexity creeps in as we continue to expand the feature set. But I always try – and the structured data is, I think, a good example of this – to get away with as much simplicity as I am able. Then user requests inevitably lead to added complexity. But then, at least, I’m only adding what’s needed and being asked for, rather than littering the UI with too many assumptions.
Understood. But here’s another perspective - from the perspective of someone licensing images through agencies as well as my own site.
One of the most important things to do is embed metadata into all images you put online anywhere, especially those you submit to agencies. You do the work once and wherever the image goes, the metadata goes with it. In the multiple agency scenario, it is actively bad to add metadata on the site (or in this case in Backlight) because that information is not available when you put the image somewhere else later on (you can copy and paste, but that gets unreasonable as the collection size gets large).
Agencies and sites come and go; the images live a long time
With keywords, for example, it can be hard to remember all the details accurately long after you initially took the image, so researching and entering those once, up front, is a win.
I realize there are many different situations with different requirements and priorities. Only wanted to flesh out one of the many.
Thanks for the perspective. It’s extremely helpful to us to hear how other photographers actually do things, as without being told, Backlight is built only around how Ben and I like to do things, as well as how we imagine others might like to do the things that we do not personally need to do.
So that appears to work - although I don’t understand why I see two (apparently identical) licensable images on one page (using the Rich Results Test tool). I think I set things up using your code as you intended:
/* jas 09/21/2021 to try & get Google licensable tag on images */
function single_bottom( ){
define("__AUTHOR__", "Jo Ann Snover");
define("__COPYHOLDER__", "Jo Ann Snover");
define("__LICENSE__", "/Standard-License-Commercial.pdf");
....
';
}
A separate but related question. A number of the agency links with the licensable tag use one of my images (and credit me as the copyrightholder) but the link is to a search versus to the image page.
For my own site, images that show up in Google images (no licensable icon) have a URL for the gallery, not an individual image. For example:
versus the image it shows from that gallery
Can your code be adapted so that $ph_path is a gallery?
I did not disable the structured data in Backlight - should I have? I thought I understood your instructions, but perhaps not.
I’ll find some of the agency examples that link to searches versus individual images. I think I have examples of both types from more than one agency - i.e. they are not uniform in what they do
All the examples I saw link to the image. From a usability perspective, I think it makes more sense to link to the single-image page. But this shouldn’t have any impact on the badge at all.
Google didn’t put a licensable badge on anything from my site - no real clue as to what the issue is.
I updated to Backlight 4.1 and have disabled PHPlugins, plus re-enabled structured data in backlight.
I want to see if Google will be happy if the images have embedded metadata, so I deleted a small album from my site (just a dozen or so images) and re-uploaded them so I could get the metadata into the watermarked image. I’ll see what shows up in image searches after a day or two.
I used IPTC’s inspection tool to see the embedded metadata that Backlight 4.1 Publisher copied over from my originals. I don’t know where the extra characters in the Copyright Notice (IIM) and Creator (XMP) are coming from - they’re not in my original.
The screen capture shows the tool’s output and the Tool Inspector (Mac’s Preview app) window for my original JPEG - don’t know if the extra characters are significant (i.e. will they hinder Google) or not.
Hi @joannsnover, thanks for trying this out. Please let us know if the badge shows up in Google.
I don’t know if that extra character will affect anything. I’ll try to reproduce it and put in a fix. It’s likely related to handling of unicode characters in either reading the value from the source image or writing it to the resized image.
Hi @joannsnover, I’ve worked out the issue with the Copyright symbol. The character encoding is correct. What was missing was that the embedded IPTC data needed an encoding flag set to declare the type of encoding. This was tricky to work out, because the metadata reader that I use, the command line exiv2 automatically worked out the encoding so didn’t show this issue.
I’ll put in a fix shortly for this.
hi @joannsnover, I’ve pushed a silent update to address the encoding of unicode symbols including the copyright symbol. To install this, visit the Backlight Modules page and click on reinstall for the entry for module-publisher. You’ll need to re-upload the images for the embedded metadata in the renditions to be updated.
Ben, I have installed the silent update and re-uploaded some files to test the change. I now see the copyright information in the watermarked gallery images as it was in the original JPEG, without the extra character - your fix looks good to me (but all my images were generated in Photoshop, and if there’s anyone else using other tools to create the JPEGs they upload, it’d be good to hear from them too).
In looking through the various image searches, I haven’t seen any licensable badges appear yet, but I’ll keep an eye on it.
It is a good thing that my copyright and contact information is in the watermarked images - if anyone lifts one and later wants to find out where it came from to purchase a license, that’s now much easier. At least for the newly uploaded files. It really would be wonderful if Publisher could regenerate watermarked images from the already-uploaded originals…
I’m still a bit puzzled about how Google is deciding what to index. When I look at the search console, before I made and uploaded a site map (17 Sep 2021) it appears Google only indexed the galleries, not individual images. Since then, Google is making slow progress indexing the permalinks (the xxx.php URLs).
The boats in Boston Harbor image is one that Google has indexed, but it still doesn’t show the licensable icon or my copyright information from the image search page. I will worry about getting the licensable icon on the gallery URLs once I’ve seen them show up for the individual image links
