I appreciate this response! Question, out of what some see as the 3 major paid RAW processors (say, Adobe LR, Capture One, and DxO) do any of them do better in this regard? (ie writing metadata in a standard way readable by others, keywords in particular)
Since Adobe is the driving force behind XMP metadata and the
go-to in the professional imaging world, their XMP metadata support in products Like Photoshop, Lightroom, InDesign etc. is very good.
I have not used C1 and DxO for a while, so I'm no expert. From my past experiences, these applications are not as good as Lr when it comes to metadata management.
I know that many IMatch users work with DxO and C1, because they get better image processing results from these applications. And they have IMatch to properly deal with metadata. If you want detailed answers, feel free to post a question in the
IMatch user community. You can also use the community search for terms like DxO and C1 and see past discussions. This comes up frequently.
There are of course always some
gray areas, where to store what and how in XMP metadata, leaving room for
interpretation.
For example, how to
flatten hierarchical keywords like
"location|beach|Daytona Beach" into the single word XMP subject or legacy IPTC keywords. Some applications write only "Daytona Beach", some write 3 keywords "location", "beach, "Daytona Beach" and some write the hierarchical keyword as is.
This often causes
friction for other applications which don't handle the original hierarchical keywords or interpret the flat keywords written by the other software in the wrong way.
Some software stores keywords not in the official XMP standard subject / lr hierarchical subject fields, but inside a
proprietary XMP namespace, which makes the data inaccessible for most other applications.
Some software stores
user-created keywords in the official XMP subject fields, but AI-generated keywords in a proprietary and undocumented XMP namespace. This makes them less portable to other applications.
If you're happy with simple keywords like "beach" or "dog", you're probably fine.
If you work with more complex keywords, hierarchies, scientific taxonomies etc., make sure all applications in your workflow agree on how to deal with hierarchical keywords in XMP, or have at least options/features to deal with the varieties in use.
Another real
minefield is the area of
synchronizing data between different standards.
For example, EXIF metadata in images has counterparts in XMP.
When the user changes a field covered by EXIF and XMP (maybe a timestamp, because the camera clock was wrong), these changes must be saved into
both EXIF and XMP - else he ends up with
two sources of truth, and this is never good.
Same goes for other EXIF data, descriptions, GPS coordinates and suchlike. A surprising number of applications, even popular ones, lack in this area.
For images created in professional environments (press, stock), legacy IPTC metadata, if available, must be synchronized with the corresponding counterparts in XMP. The IPTC committee even has a
test suite for this, enabling applications to test and verify their compliance. Here are
the results for IMatch, for example.
This is what I mean when I say you should check/test before committing to a free or paid DAM system or RAW processor.
If you look into purchasing one of the three products you mentioned, I would recommend to install their trial versions, process some of your images, add the metadata you need and want to preserve, and then open these images in the two other applications.
Check if the metadata is still complete. keywords are there etc. If you use a language that has special characters like German umlauts or accents, check if they survived.
Testing this will maybe cost you a day, but it's worth it, I believe.
If you put in the work in to rate, label, describe your files, to add keywords, maybe GPS coordinates, location data etc., you should
have to do this only once.
And from then on the data should safely
travel with the image and be accessible for all capable and XMP-compliant applications.
If users do the work, they either need it for professional reasons or because they want to preserve the info for future generations.
I ramble about all this (and care so much) because new users coming to IMatch often face exactly these problems with their existing image collections:
- Metadata is missing in their files because their old system tucked it away in some proprietary catalog.
- XMP metadata is incomplete and not standard-compliant.
- EXIF metadata is out-of-synch with XMP.
- Legacy IPTC metadata is broken, out-of-sync or in parts unreadable, because the wrong character set was used or not specified correctly.
- Flat keywords out-of-sync with hierarchical keywords
- Different timestamps in EXIF, XMP and IPTC
- GPS coordinates only in EXIF or XMP
- Different GPS coordinates in EXIF and XMP
- ...
See
Metadata for Beginners for more information about all this.
Even if IMatch has special tools to support users with identifying and remedying this
metadata mess, it's still a lot of work and unpleasant. And data lost has to be re-entered completely.
Spending a bit of time to test such issues before you commit to a product is worth your time. Also check the results you can produce for your RAW files in these applications. There will be differences more or less important for you.
Users (often) switch RAW processors when they find a product that produces better results / is cheaper / non-subscription.
But if this means that they'll loose much (or all) of the metadata they have meticulously entered over the years, it's
much harder to switch. Proprietary metadata management can indeed be used as a customer lock-in / retention tool :-(
