I’ve never used tif for my scanned family photos. But looks like you are right. I’ve read a bit more about tif and understand now that tif is better for this purpose.
But is it really the best way? I write all necessary data directly on the images - above/below them. In this case anyone can see who is on the photo, when it created, places, etc… As for metadata - how many people will read something there except the genealogists? I know, nobody of my family can read metadata. And more… they even dont know what is it. Its just interesting question for discussion. Who uses metadata, who writes data directly on the photos, …
I used jpg format to keep things simple for me. I scanned most of the pictures at 300dpi so it kept the files size to about 1meg each. Size was not an issue since I have Terabytes of space available.
I would not try to edit the picture and put text on it since my goal was to preserve the picture as best I could for historical purposes. Editing them to change any mistakes would have messed up the image too I think. Numerous times I went back to correct a name or add found info.
Adding the info as meta data is very simple with the windows Photo app. You can then use the file search to find the files with the data you are looking for.
This is also accepted on the photo sharing platform Flickr where the person viewing can see the info. Again it can be searchable.
I wouldn’t use j-peg if there is any other option. The compression artifacts are just too destructive. Worse if the content is high-contrast or has sharp lines.
If you’ve got lots of storage, don’t use a lossy format.
These photo were taken from the early 1900 through until we started using slide film about 1980. None of them are of high quality, many are blurred as the camera was moved as the picture was snapped. Some are so faded it is hard to make out the image.
I appreciate all the comments and I may change my methods for the future but I will not be editing these and I don’t want to manage 2 copies (tiff & jpg) of each image, so jpg is good enough.
I never write on photos unless it is mine and it is a signature…
Any data for a scanned photo I add to metadata… but if you like to write under or over the image on a white frame, that is all up to you, I will not tell you have to do something in one way or the other…
But text in images is not searchable, you cannot sort on them, organize on the information etc.
If you have a few hundred or maybe a few thousands of scans, it will at some point be important to be able to sort and organize them…
Just for a simple task as to give the relevant images to relatives…
If you use a software like Darktable, (edited 6 days later: or DigiKam of course) you can still print out images with name on them, both to a file (e.g. pdf or jpg) or to a paper copy…
The tiff files are the “digital originals” all you do with them is to add the metadata… if you want to do anything else with them you eiter use a software with nondestructive edit, or you make a copy of the file, that copy when you are finished editing, will stay with the original… (need to find a good file name strategy that fits your workflow).
And you can still link your photos in Gramps and any other software where you want to use them…
I’ve used Tropy for archival research (not family history related) but in very small (a handful or two) photograph collections and it worked well. One thing I see on their website is that it will connect with Omeka. If you wanted to make the images available online and have control over them, you could install your own Omeka (both Tropy and Omeka open source) and connect the two. I’ve used Omeka quite a bit a few years ago for archival collections and though text could be a problem with large PDF files, the images always worked pretty well. Never tried connecting Tropy to Omeka though.
Tropy has export to both Omeka and Zotero (CSL-JSON).
You can also export your research to both CSV and JSON-LD.
Both Tropy and Zotero is great tools for genealogy if the user just thinks a little outside the box on how to use them… Same goes for Omeka.
There is a lot of open source and free software out there, but sadly it is often difficult to use them in combination with e.g. Gramps or other genealogy software, because none of the genealogy software support any of the file formats commonly used in other type of historical research…
I think that originals should be saved in high quality and information about it should be in the exif-data in the image itself. In Gramps you probably don’t want hundred 25MB images of a person if you generate a report or a website.
I will now do it like this:
-Only load the very old family photos (around 300 images) in PNG format into Gramps and link them
-The originals are stored in TIFF format in my archive
BTW, IPTC was designed for news organizations sharing images (and maybe other media). EXIF and XMP are more general purpose and might be a better choice for longevity?
Yes, you’re right. In EXIF there is also a field for a description of the image. With the GPS data, date and time, these are the fields I need.
I will test the add-on. Thanks
I’m reading this thread a little late, so I’m going to answer the topics raised a little out of order in relation to my practices.
About referencing all of his photos in Gramps
I did it, a few hundred photos, and I didn’t encounter any particular performance problems in Gramps.
The interest for me is to be able to tag, not with tags but with attributes, each of the media with pre-established attributes:
Identified individuals : number of persons on the photo vs number identified. I have a filter to be able to find all media whose count of recognized people / unrecognized people in the photos is different.
Photographer: when named, to filter media with the same photographer
Type of document and Description : for general use
Tag : To describe photo content with keywords and be able to search on same content type
Interest of attributes is to be able to add notes or citations telling why I’ve said I recognize someone (even if I use it too in notes and citations of top part of person/family-media relations).
About Tropy
I usually use it to transcribe old sources but also for postcards. The media in Tropy and Gramps are linked to the same Windows file. Tropy is interesting because in addition to this transcription capability, it allows you to join several media as a single document (both sides of a postcard, all the pages of a notarial contract, etc.). Like Gramps, it allows you to select a part of the image to transcribe only this part for example. When I have finished transcribing the document, I export the text of this transcription into a Gramps note. I also extract a JSON from this Tropy entry containing all the source information, files and transcriptions, and I attach it to the source in Gramps as another media named Source Transcription.
And, if you want some examples, I’ve wrote some blog posts, probably not all translated from French to English, on my media usage with these tools (at least two are translated: one about Tropy, and one about media usage).
