Hi, I’m using the LR Publisher together with Jeffrey Friedl’s Metadata Crawler. It doesn’t seem to work as I expect. I want to strip some metadata, e.g. remove some specific keywords and also the IPTC Title.
This works when just doing an export from LR, so the Metadata Crawler preset is generally correct. But when I publish my images the metadata won’t get changed, it’s still the same in the downloaded image.
Am I doing something wrong or does the publisher simply ignore Metadata Crawler?
Unfortunately, our publisher doesn’t work with Lightroom’s export plugins. There are good technical reasons for this.
Lightroom is designed such that an export of 100 images is a single task, and it either succeeds or fails entirely. Plugins hook into this task at some point to operate on the images being exported.
Our publisher does not export your images in a single task, but as a series of isolated tasks per image, which allows us to publish multiple rendition sizes of each image (thumbnail, full-size, and possibly more depending upon your settings), and allows you to resume your export, if it should fail, without having to restart entirely.
We think these are good tradeoffs.
Your best workaround would be to stop using the Publisher altogether.
Instead, export your images from Lightroom to desktop, using whatever toolings you like (Metadata Crawler, etc.). Then upload those files directly to Backlight with your web-browser. Backlight’s native publishing features are great.
Thanks for your explanations. I understand your point completely and I think I’ll go without the Publisher as I don’t want my serial numbers and other metadata in the images. Nevertheless I like the integration of the Publisher because it allows a much easier workflow.
I know nothing about Lightroom Plugin development (yet) but I think it should be possible to take a plugin having been added to the Publisher into account!?
I also think that I’m not the only person who would like to strip metadata in published images? What about extending your Publisher to make usage of Jeffrey Friedls Metadata Crawler directly or even provide your own functionality for stripping metadata?
Last I looked at it, Lightroom’s plugin API was very restrictive. It’s one of the reasons we spun Backlight out into a standalone app, rather than continuing to make plugins for Lightroom’s Web module.
Also, not all of our customers use Lightroom. We have limited time, and would prefer to spend it on features benefiting all of our customers, not only the segment that uses Lightroom. I’m a Capture One Pro user, for example.
We’ve built out Backlight’s native publisher to improve the publishing experience for all. And if it’s important to export images using these third-party plugins, then I recommend – as before – exporting to desktop, then uploading to Backlight.
As @Matthew has said, the LR plugin API is very restrictive.
The default publishing loop produces one rendition per image. That process also calls other LR plugins. This is handled outside of our own plugin and we have no control over whether or not these calls are made.
Our approach explicitly creates multiple renditions per photo. This is necessarily done outside of the default lifecycle of publishing. Unfortunately, this misses out on the automatic elements of the default publishing loop and we have no mechanism to perform the same calls to other plugins that LR provides automatically in the default publishing loop.