…it all depends on who you and your expectations are. Some people value other people’s rubbish.
Anyway, electronic archives need to be migrated from time to time as systems and technologies evolve, no way around it. Today’s formats might well be lost after tomorrow, and as far as my personal experience goes, you best select the optimal data format when you migrate to the next archive technology - and accept that some info might still be lost in the process. All that is not migrated will be lost, unless we use human readable formats (prints) and store them accordingly.
Good old time when my ancestors draw pictures on rocks 1000’s of years ago. They are still found and can be seen. All those wonderful paintings from a few hundred years ago. The only tool we need are our eyes.
My images are mine. I don’t have the illusion anybody will take care of them after I’m gone. If I want to be remembered later the only solution will be the printed version and hope somebody will find them sometime in the attick or in the basement in a shoe box. For the time being.
In the Nikon software it was common to write the keywords in the raw file. Nikon left that part of commerce since the Nik-software was sold to Google. Camera’s up to the D750 are still supported in that software. Keywords added that way or using Exiftools in the iptc section are read by DxO. They are not written to that section but to the xmp section. I think it’s the dominance of Adobe.
True, however if your keywords have been written back to the image, then any file management software capable of searching file contents can search for images that contain particular keywords.
True, but it seems that while some apps can and do store IPTC data in the original raw file the process is not supported by the owners of the file formats, when not Olympus, Panasonic or Nikon. Also editing a single keyword will mean that the large raw has to be written to back up. So you pays your money…
Which software? I ask because the few that I have looked at seem to write to sidecar. Mind you I could have missed a preference setting or other command. PhotoMechanic does, but is not yet a full blown DAM application.
Thanks for that. Only Photo Supreme runs on MacOS and it uses a non-standard interface but it does have preference which appears to mean that it writes keywords into the raw files.
However, I’ve had a brief play with Photo Supreme and I don’t think its for me. The use of non standard terminology is unhelpful and I suspect that while in the past the use of “catalog tags” was an improvement over other applications this advantage has been implemented by other apps in the form of hierarchical keywords.
Also, try as I might I am unable to get it to write back to Panasonic raw files; it just updates the .xmp file. Looking at the release notes I see a reference to macros with ExifTool so my guess is that PhotoSupreme uses ExifTool to do any writing to raws although as I say I can’t get to do the write despite ticking the preference setting.
Photo Supreme writes directly to raw files for “heritage” files only. I was in touch with the developer and he told me that this decision was made a few years ago for those file types but for newer raw-file types it is only .xmp. He said this is the standard and that PSU will stick to this standard.
Embedded metadata writing in PSU is supported for
NEF
NRW
TIF (RAW versions some Canon older cams)
ORF
CR2
PEF
George - I am not the developer, so I can not answer that. I only ran into it because I added a few RAF file from my son’s camera into PSU and noticed that xmp files are created. Before I only had NEF and CR2 in the database. After asking support of PSU I got the above answer. If you need more info about it please contact PSU directly.
Standard way for handling image metadata according the Metadata Working Group guidelines, interpreted by the developer of Photo Supreme (write only to XMP sidecar). (And the safer way for me after having troubles with CR3 and Photolab.)
The developer of IMatch on the other hand says, he has to write at least some information to the RAW files and uses Exiftool for it.
I haven’t tested Lightroom in the respect of writing to RAW files directly, but after you set the catalog to write also to files, it writes to XMP sidecars.
There is an option somewhere to write directly to the raw files which I haven’t used/read about.
Exceptions are DNG format raw files where you have to write to the files directly.
I remember reading about someone having also problems with some of these files with Photolab in the past.
Section 1.1.1.1 deals with raw formats. Just a warning not to deal with tif-based raw files as just being tif-files.
The iptc and xmp sections are just added to the raw file. The only change that has to be made in the raw file deals with the adressing of the different sections.
I write raw file but that could be more files, video, images, pdf etc. See the link.
In the same section I do read that acr can write xmp directly to the raw. I don’t know if lr, using acr, is giving that possibility to the user.
My memories were not totally correct.
Lightroom can only write date or time changes directly to the raw files, see last marking. (This is not tested by me and it is an older version.)
The upper marking shows the non standard setting of writing every change directly to the sidecar (not only to the catalog).
The setting before the upper mark tells Lightroom to write also the changes made in Lightroom to some specific formats.
As far as I understood the nature of ACR, it uses the same engine as LR, but then puts the result into Photoshop. Don’t know about metadata handling.
Anyway: writing metadata directly to the RAW files is possible, but you have to be very careful to use the right tool for it.
The (not very user friendly command line tool) very powerful Exiftool is probably your best bet and everything that uses this tool.
BUT: if you use Photolab, you should be even more careful then in other tools. After I run into the problem with my CR3 files many people stated to me that they were writing everything to the raw files for years and didn’t have any problems.
A little slumpy part of the Photolab CR3 reader code was enough to not let me open my CR3 files in Photolab and it will take some more time until the fix will get into the main version, although it is already solved by the programmers.
So be very careful of writing data directly to the RAW files and always keep the unchanged files backed up.
I can’t remember if you are a Win or Mac user, but this part of the IMatch (Windows only DAM) Help can help you understand some not that basic facts about Metadata in image files. Part of IMatch Online Help about metadata storage
Even if I decided not to use IMatch (because it only works correctly if you let it modify the RAW files and I don’t want to do it for the uppermentioned reasons) I appreciate the things I have learned through the evaluation of that software and the hunt for the CR3 error about metadata in image files and how they are/can/should be handled.
Thanks for the extra information about DAMs. This seems to confirm that writing IPTC data into raws requires a lot of effort on the part of software authors. They have to reverse engineer and test all new raw formats on a camera model by model basis. Its hard work especially when compared to read/writing xmp files. So its far simpler to write one set of routines to work just with xmp.
Not sure what you mean by ‘catalog tags’ - I use Photo Supreme and it definitely writes hierarchical keywords (or optionally delimited keywords, though that’s deprecated) - see options below:
Having checked (there is a thread here), I can confirm that despite having the option to write to RAW files (see above settings), Sigi is correct; in practice it only supports selected RAW files (NEF, NRW, ORF, CR2 and PEF + DNG), not all RAW files. It also writes to JPG & TIF. From that thread, it seems that extending support hasn’t entirely been ruled out - I’d certainly support it’s extension.
No, it doesn’t use ExifTool. It has its own set of internal macro commands (plus the ability to write scripts in Pascal if the macros don’t do what you want).
IPTC-IIM is an superseded 1990’s standard. However the ‘IPTC fields’ are still present in XMP, which superseded IPTC-IIM, so no surprise that XMP is preferred.
To clarify I’m basing my comments on a trial of PhotoSupreme. When looking at the documentation there are references to catalog tags and that these may be written as keywords in meta data.
Not what I meant: writing meta data of any sort into a raw file will probably change the length of the file. This means that all the pointers stored in the IFDs must be modified otherwise the file will be corrupted. This is how a Tiff file works, however raw files contain undocumented IFDs and tags (not the tags from my first comment) which mean that each and every version of a raw file has to be reverse engineered to determine how to edit it. Each raw is different so every raw the software edits requires special treatment and testing. With 400 odd raw formats out there it is no wonder that xml xmp sidecar files have become popular as just maintaining the details of 400 read and 400 write rules is a major challenge and is most likely the reason that PhotoSupreme does not write to more recent raw files.
