Must all the parent keywords be clicked to be searchable and exported flat?

[Damit]

I was continuing my experimentation with keywording and noticed that when I assigned a child keyword to a file using the Hierarchical portion of the Keyword Panel by clicking its box and creating a check mark, that the parent keywords did not also automatically check. Their boxes remained empty.

I did some tests and I see that even if this is the case, the file can be found using the parent keyword in the search tool, even though it is not checked in the hierarchical portion of the keyword panel. So, I think IMatch searches all the components of the keyword hierarchy (except for, perhaps, groups) when trying to find the keyword in the search bar.

But I am wondering how this would work if I flattened the keywords.  Would there be an individual keyword associated with the file for each level of the hierarchy?  Lets say I wanted to flatten a hierarchical Keyword WHAT|Art|Artist|Vincent van Gogh, and only Vincent Van Gogh is checked when I look at it in the Herarchical Keyword portion of the Keyword panel, would the export create flat keywords titled "WHAT" "Art" "Artist" & "Vincent van Gogh" if I exported the keywords in flat format? I assume so, but I want to confirm.

I assume that checking all the boxes would create keywords in the keyword panel of "What", "What|Art", "What|Art|Artists", and "What|Art|Artists|Vincent van Gogh" which would just muddy up the Keyword panel, so it is best that only the final child keyword be clicked to reduce the amount of repetition of terms, as there is no reason to have those 4 listings of hierarchical keywords if one would suffice for proper association and search-ability, but I just want to confirm this is the case.

[Lukas52]

I assume that checking all the boxes would create keywords in the keyword panel of "What", "What|Art", "What|Art|Artists", and "What|Art|Artists|Vincent van Gogh" which would just muddy up the Keyword panel, so it is best that only the final child keyword be clicked to reduce the amount of repetition of terms, as there is no reason to have those 4 listings of hierarchical keywords if one would suffice for proper association and search-ability, but I just want to confirm this is the case.

You are correct on that

But I am wondering how this would work if I flattened the keywords.  Would there be an individual keyword associated with the file for each level of the hierarchy?  Lets say I wanted to flatten a hierarchical Keyword WHAT|Art|Artist|Vincent van Gogh, and only Vincent Van Gogh is checked when I look at it in the Herarchical Keyword portion of the Keyword panel, would the export create flat keywords titled "WHAT" "Art" "Artist" & "Vincent van Gogh" if I exported the keywords in flat format? I assume so, but I want to confirm.

This depends on how you want to flatten them. I do write my Hierarchy Keywords back into "flat" IPTC Keywords for compatibility with my web server (stupid thing can't handle XMP...). Flat however just means "string of characters" in that case. It still writes "What|Art|Artists|Vincent van Gogh" as one Keyword. But thats the point, this is now a single Keyword. In my case this means if i want to search for all Artists on my web server i have to look for "What|Art|Artists|*" otherwise it finds nothing since you wouldn't use the "Artist" level of your structure.

When it comes to flat vs Hierarchy it depends a lot on how your individual applications handle and treat them. IMatch can however help you here a bit. What exactly it is you need to do depends on what you are trying to achive.
Maybe looking at some raw Metadata would help you understand the difference between Hierarchy and flat Keywords further. Just load a file into the Metadata analyst app and look at how it creates the Tree that makes up the Hierarchy vs. how a flat keyword really is just a string of text.

[Mario]

that the parent keywords did not also automatically check. Their boxes remained empty.

This is how it must be!
Consider a hierarchy like

LOCATION
|- Beaches
    |- Daytona

You click on Daytona. This adds the hierarchical keyword LOCATION|Beaches|Daytona to your file.
Please do that in the Keywords Panel so you can see which keywords you add. This should make things pretty clear.

Now click on each level. This adds three (!) hierarchical keywords to your file (and you don't want that):

LOCATION
LOCATION|Beaches
LOCATION|Beaches|Daytona

You see this in the Keywords Panel immediately.

Note: When you make LOCATION a group level, it will not become part of the keyword. You see that also in the Keywords Panel. And when you search for or filter by keywords, LOCATION will not be considered. It only "exists" in the Thesaurus. IMatch considers it when importing files which contain flat keywords, to map from flat to hierarchical keywords.

IMatch works only with hierarchical keywords. So it will search the entire hierarchical keyword (it will search what you see in the Keywords Panel for a file). All other features dealing with keywords also work with the full hierarchy.

Since the XMP::Subject and IPTC::keywords tags don't support hierarchies, hierarchical keywords must be converted to flat keywords during write-back. Files contain usually two sets of keywords in XMP (hierarchical and flat) and optionally a 3rd set in the legacy IIM IPTC data.

See Write hierarchical keywords in the IMatch Help which explains the options available.
You may have to change the default settings (after careful experimentation) if other applications in your workflow don't handle hierarchical keywords or handle them wrong or handle only flat keywords but expect them to be in a specific format etc.

I would also strongly recommend to read Metadata for Beginners for a good overview about various metadata and how IMatch reads, processes, maps and writes it.

[Damit]

Great! It seems I understood how things worked. At first, I thought I needed to add checks on all the levels but then realized I did not want that, just as you confirmed. I am also glad I understood that groups would not be written; at least I am starting to understand how some things work. After re-reading that Metadata help section you linked, I realize all the options I have, so if I need to export I can tailor how it needs to look for whichever application I use to easily digest it. After working with hierarchical keywords for a bit the information contained in the Metadata preferences was better digested.

I have read Metadata for Beginners several times and will read it again many times, starting with right after I write this. It was the first thing I read after starting the trial and one of the things that convinced me to buy the program. It is a great document. Unfortunately its a lot of information, with lots of links, and coupled with everything else I am reading about this program, much less anything else, and it does not always stick as much as I would like. I believe it has to do with my over familiarity with the subject. I am sure with more familiarity things will gain more meaning, which has been directly related to retention, things will stick. I know I have read this Metadata preferences section as well, and tried to find this information with a search before posting this question, as I had a faint recollection of reading it, but I was not successful. But please be aware that I have probably read 90% (if not all) of the help section and probably 70% of it more than once, even though it may not seem that way.

Thank you so much for your help!