Genealogy

[BarryC]

Hello from someone with a trial version. Mario, congrats on what seems quite amazing software, never seen so many positive responses. I will be purchasing and am certain it will be great. My wife has a passion for genealogy and is amassing way too many scans, photos, documents etc. She has genealogy software where she links somewhat lower res images within the application. No problem there, but outside it creates a versioning issue as there may be an original image, a restored high res version and then the lower res version for her software. I have been reading and looking at Hierarchy, keywords, labels and templates - much more to learn. I am trying to understand the best method of Hierarchy. Does a top level start with say the family surname, then certificates (eg birth or death), images, documents etc. Does one start perhaps with Certificates then surname or Images then surname etc. Are keywords automatically generated from the Hierarchy or are keywords populated seperately. I figure the best way to update metadata in files is by way of a template and as I read somwhere here, 5 good categories is better than 20 bad ones. I just need to know the best structure for us to quickly find images or documents. I hope you can see my starting dilemna. Most images bear the contents within the file name at the moment. Looking forward to the IMatch journey and some starting help. Barry

[Mario]

Hi, BarryC

welcome to the community 

A small tip for your first post: It helps a lot if you begin a new paragraph after two or three sentences. Especially users who read this on small devices see a 'wall of text' otherwise 
Just hit <Enter> once in a while, thanks.

I cannot comment on Genealogy topics because I don't use that.

Does a top level start with say the family surname (...)

You have basically 3 ways to organize your files in IMatch:

1. IMatch Categories
Very fast, quick and easy to re-organize (especially while you try things out) stored safely in the database.
Categories are an IMatch concept and not directly (without an extra step) accessible for other software).
Advanced features like formulas (to combine categories in many ways, like (Family Miller AND Family Smith), Alias categories to "link" the same sets of files (family members) into different places in your hierarchy etc.

2. Hierarchical Keywords

The XMP metadata format supports hierarchical keywords like "Families|Miller|Paul" and IMatch supports that fully, in the Keyword Panel.
The special @Keywords category automatically reflects the hierarchical keywords in your files. This gives you the best of both worlds. I explain this in detail and with examples in the IMatch help for the @Keywords category.

3. Both

If you want to make your family hierarchy accessible for other software via XMP keywords, create your family hierarchy in the IMatch Thesaurus. This gives you a controlled vocabulary (good) and allows you to pick keywords from a list. Very fast input, zero errors. See also this related know-how article in the IMatch Knowledge Base.

When you add or change hierarchical keywords in your files, the special @Keywords category reflect that automatically. Here you can also re-arrange files "between keywords" or hierarchy levels using drag and drop or copypaste.

For additional options, to combine categories dynamically, aliasing categories etc. you can use other category functions.

Note:

IMatch categories and XMP keywords are hierarchical, top-down.
This is ideal for all typical use cases, except n:m relations. n:m here meaning that a category/keyword can have more than one parent.
Genealogy software is specialized and supports n:m relations of course.
Paul has both a Mom (Anna) and a Dad (Michael) and hence has two "parents" or ancestors.

This is not possible in IMatch or other standard DAM products.

I'm sure other IMatch others will chime in and give you some tips about how to manage this in IMatch.

[jch2103]

... My wife has a passion for genealogy and is amassing way too many scans, photos, documents etc. She has genealogy software where she links somewhat lower res images within the application. No problem there, but outside it creates a versioning issue as there may be an original image, a restored high res version and then the lower res version for her software.
...

IMatch is indeed a very useful support tool for genealogy. The fact that it's so flexible and provides so many options is excellent, but of course deciding how best to use all this may not be so simple and depends very much on your needs.

I use IMatch to track all my photos, including my genealogy images and documents. (I use Legacy Family Tree for genealogy, but the principles should apply to any genealogy program.) I should also note that my genealogy efforts are a work in progress, in that I've changed how I track information more than once. Fortunately, IMatch's flexibility makes it fairly easy to implement changes.

Here's an overview of what I'm currently doing:

- Photos of people are named 'Last name, First name explanation.extension' (most extensions are jpg but may also be pdf, etc). 'Explanation' is fairly broad, and may include things like description (e.g., 'home' or subject age/date or 'wedding' or 'crop'). I personally have made only limited use of versioning for genealogy photos (I have more work to do on editing/restoring my images) but the feature should help you deal with different versions of the same image.

- My metadata layout includes some essential fields. Description is fairly obvious; it allows an extended description what's in the images. Creator is useful for things like photos of paintings (who painted the original) or the name of the portrait studio (some of which no longer exist). Hierarchical keywords are very useful for several purposes: I use the keyword 'Genealogy' for genealogical images; I've also started a hierarchical set of keywords for 'Photo Type' (e.g., 'Tintype', 'Carte de Visite', 'Cabinet Card', 'Postcard'). You may want to use a special variation for the lower res images used in your genealogy program. Where known, I include Country, State, City and Location and in a few situations add GPS coordinates (e.g., photo of a house). Very important are dates: Original Date (XMP::photoshop\DateCreated\DateCreated\0) and Digitized Date (XMP::xmp\CreateDate\CreateDate\0). Note the exact metadata names; there are quite a few very similar sounding date/time metadata tags, and it's important to choose the right ones. The Original Date is the date/time the original photo/painting/etc. was created, which allows use of the Timeline feature of IMatch. The Digitized Date is of course the date/time the current image was scanned or digitized. Note that inexact dates are somewhat of a problem because these metadata fields don't provide for approximate dates; most folks use a convention such as '1/1/1901 12:00:00 AM' to show an approximate date of 1901. Search the Forum for more examples of handling approximate dates.

- I've made some preliminary use of face tagging. This can be a useful supplement to including 'left-right' descriptions of who the folks are in photos, especially group photos. See the Forum and IMatch Help for more information on this topic.

These are just some starting thoughts. As I said earlier, my own implementation is a work in progress. Let us know if you have questions or other thoughts on the subject. Good luck!

[BarryC]

Thankyou for the feedback and apologies for the wall of words. I'll dig deeper into keywords and the thesaurus functions. I'm told I need to tag images with one of 8 colors as well. Labels to the rescue I suspect. More play, more questions, back soon. thanks Barry

[Mario]

I'm told I need to tag images with one of 8 colors as well

For that I would look into XMP labels.
An XML label is a text, like "Red" or "For Review" or "Final". There is no real standard.

Capable DAM applications allow you to associate colors with label texts, e.g. the a red color with the label "Red" or a green color for files with the label "Final".
In IMatch, you can define your own labels and associate colors under Edit > Preferences > Metadata.

You can then pick the label (and see the colors) from the rating & label bar under the thumbnail in the file window.
IMatch writes the label text into the XMP label metadata tag and other applications can see that and work with it.

Files are automatically grouped by their color (label) in the Collection View.

The Filter panel allows you to quickly find files with specific color labels.

Tip: You can configure a file window layout to show the label color on the thumbnail backdrop (one of the standard layouts included in IMatch does that). This is very helpful, especially if you use a thumbnail only layout with small thumbnails on a big monitor. You can see hundreds of files and their colors at once.

[jch2103]

One more note on genealogy-related metadata: See http://cv.iptc.org/newscodes/digitalsourcetype/ for use of a metadata tag that describes the source type of digitized information.

[ubacher]

I have only dabbled with genealogy a little.
I see the following needs/principles:

You need to be able to enter into the genealogy program the location of associated documents: photos etc.
I suggest a unique file name - this way you find the file easily in Imatch (but also with Windows if need be.)

You want  to link the documents in Imatch to a unique person in the genealogy program. I think most use some Id number for each person.
Since a document can link to many persons you need to be able to record a list of Ids.  I suggest using an attribute.

And the problem of synchronizing: When you enter new documents in Imatch you will want to have some way of indicating that
the information has also been added in the genealogy program: a category you set when this has been done.

The problem of uncertain dates for documents has been mentioned in this forum several times in association with scanned images.
There is no known solution.

[BanjoTom]

This has been discussed previously, but the best solution I've found for dealing with uncertain dates is to follow the guidelines in the Library of Congress's Extended Date/Time format (EDTF), which is detailed here:

          http://www.loc.gov/standards/datetime/pre-submission.html

While the EDTF system cannot be used to fill in dates in the "Date Created" metadata field ({File.MD.XMP::photoshop\DateCreated\DateCreated\0}), one can put EDTF dates into an otherwise-free or unused field such as the Dublin Core "Coverage" field {File.MD.XMP::dc\coverage\Coverage\0}, or the Dublin Core Event Verbatim Event Date field {File.MD.DarwinCore::Main\EventVerbatimEventDate\EventVerbatimEventDate\0}. 

When I have an uncertain date, I put the earliest LIKELY date into "Date Created" in a format that uses a standardized date and time of 12:00 a.m.  This makes that image appear in an appropriate, although approximate place in the IMatch Timeline.

So, for instance: with a scanned photo that was PROBABLY made sometime in spring or summer of 1940, I would enter "1940:06:01 00:00:00+00:00" in the "Date Created" field AND would also put an EDTF entry into the Dublin Core Coverage field: "1940-(06)?~" or the alternative "1940-uu-uu".  Before I began to follow the EDTF guidelines I used any notation I chose for a known year with an unknown date, e.g. "1940, Spring/Summer" or just "c. 1940," or a phrase that might indicate a limited (but known) date range, like "June 12-21, 1940."

Then, in the File Window tip and in any printout/export formats I set up (for instance, in the Design/Print module and in the Image Batch Processor), I use a formula that uses the "Date Created" field UNLESS there is an entry in "Coverage," like this:

{File.MD.XMP::dc\coverage\Coverage\0|default:{File.MD.XMP::photoshop\DateCreated\DateCreated\0|format:MMMM D, YYYY};hasvalue:{File.MD.XMP::dc\coverage\Coverage\0}}.

Still an imperfect system, but I find it works very well for me.

Of course:  "Your mileage may vary . . . "   

[ubacher]

Code Select Expand

{File.MD.XMP::dc\coverage\Coverage\0|default:{File.MD.XMP::photoshop\DateCreated\DateCreated\0|format:MMMM D, YYYY};hasvalue:{File.MD.XMP::dc\coverage\Coverage\0}}.

Fantastic solution!!! Thanks for sharing!
Shows how versatile Imatch is.