[lesrhorer]

I wasn't suggesting Plex itself was being discussed just alternate sharing solutions to centralize your artwork.

I am sorry.  'My mistake.  I thought you were talking about centralizing them in the cloud or some similarly horrible idea, not merely the notion of a user owned and defined repository.  As poor as I consider the idea to be, even a proprietary cloud-based solution would be OK-ish as long as it is optional and configurable by the user.

It looks like you've moved your artwork handling which is similar to how Mezzmo does it, just without all the knobs, whistles and bells (and Windows   Wink )

Please don't curse in front of me.

Some people like whistles and bells.  I like knobs and toggle switches.  Whistles and bells... not so much.  The vast majority of systems here in my house are headless.  Ignoring the TVs attached to the KODI hosts and TiVos, the touch screens on some of my appliances, and one laptop I virtually never use, there are only two active monitors in my house, and both are attached to one host.  You might guess I am sitting in front of them right now.

The question now is how to populate your missing artwork.

I presume you mean getting the .jpg files onto my server?  That is pretty easy, really.  First of all, I installed Tiny Media Manager on one of my servers.  It does a nice job of creating KODI compatible .nfo files.  I created a short Python script that yanks all the actor names and links for actors with artwork on TMDB from all the .nfo files created by TMM.  These get dumped into a shell file like the following:

#! /usr/bin/bash
wget -O "Jack Lemmon.jpg" https://image.tmdb.org/t/p/h632/7jWVWLr1...TCIHRw.jpg
wget -O "Tony Curtis.jpg" https://image.tmdb.org/t/p/h632/6Pj1XkYd...8PPwlY.jpg
wget -O "Natalie Wood.jpg" https://image.tmdb.org/t/p/h632/6me9oS7G...y7fVA1.jpg
wget -O "Peter Falk.jpg" https://image.tmdb.org/t/p/h632/5HhzU8Er...4n26C0.jpg
wget -O "Keenan Wynn.jpg" https://image.tmdb.org/t/p/h632/n5iLeYpe...ArIQZk.jpg
wget -O "Arthur OConnell.jpg" https://image.tmdb.org/t/p/h632/jJh2Fxbr...cWNQdc.jpg
wget -O "Vivian Vance.jpg" https://image.tmdb.org/t/p/h632/yQLSRUQO...FuQTrl.jpg
wget -O "Dorothy Provine.jpg" https://image.tmdb.org/t/p/h632/9HS7y9bM...1kjXT8.jpg
wget -O "Larry Storch.jpg" https://image.tmdb.org/t/p/h632/jg4W842F...B59Ebp.jpg
wget -O "Ross Martin.jpg" https://image.tmdb.org/t/p/h632/xthQH149...cyrTWd.jpg
...

which gets saved under the movie name.  I make the file executable and then run it.  I don't intend to pull the art for every single actor in every single movie.  A very nice feature in KODI is the ability to sort the actors by the number of movies in which they appear.  With a few exceptions, I am only going to focus on those actors who appear in at least three or four of my movies.  The rest are of little concern to me.

For the KODI boxes themselves, I wrote a little scraper that pulls the actor names off the existing database used by my TiVos, which has itself been built and continues to be augmented the TiVos themselves for shows recorded by them and by Metagenerator for those not recorded by them.  The scraper checks to see if the resources are extant on the main server and if so inserts them into the <actor> group in the .nfo file.  This may sound like a lot of work and a lot of moving parts, but it isn't, really.  For adding new movies from one of the TiVos, it's automatic in the transfer process.  For adding new movies from elsewhere, Metagenerator is considerably easier to use than TMM (mot that TMM is all that bad), and the structure of the TiVo metafiles is considerably easier to handle than XML (againm not that XML is all that bad).  Once the metafile is in the state I like, I run the scraper and the collection script if it is needed, and then run the wget script.

[lesrhorer]

i dont feel i can explain anything to you, it takes entirely more effort than it should

Exactly how much would that be?  I really don't know how much more or less effort it takes WRT me as compared to anyone else, but answers that are both pertinent to the query and complete are essential irrespective of their brevity or lack thereof.

and resources given go unread

This is untrue.  I read every one of them, unless I missed one somewhere.  In point of fact, I had already read the majority of them, but I went back and read them again to be sure I had already seen it and then went back again and read them to make sure the answers were not there.  They were not, or rather the information was not comprehensive and not specific enough to my use case to provide an appropriate level of certainty.  For example, I still do not know what the <tag> tag implies.  I suspect it implies nothing whatsoever, but nothing I have read or been told confirms that.  IK see tons of examples as well, but still they do not inform me if there is some mechanism somewhere that actually does something internally or if it is just a label with no meaning.  I have a notion this may be the case, but if so I am a bit puzzled why it exists at all, unless the <genre> tag is somehow entangled in the guts of KODI in some more obscure way making it unwise to create <genre> tags capriciously.

if you have a library currently, export the library "to separate files" and inspect how kodi writes everything to your disk, duplicate this and kodi will work for you

I have read this at least a dozen times, and still is not clear to me what you mean.  I did export the library once, but almost all the directories were empty and the ones that are not totally empty only have one or two .jpg files.  That is not very informative.  More importantly, it has absolutely nothing to do with the question at hand.  I could duplicate that struture all over the face of the Earth and it would do nothing.  What is it you think I am trying to accomplish?  What is it you think following your suggestion will accomplish?  I cannot see how it would accomplish anything.  The fact the suggested actions are incomplete  I am sure is not helping.

do it on a smaller library section from a portable install if you dont want a massive amount of files written side by side your media

This is nearly gibberish.  What does "portable install" mean?  I can't get KODI to install, let alone run on my desktop machine (it is quite sick), and I rather suspect it won't run on my servers.  I am not sure what I could do with it on any of my servers if I did install it.  What constitutes "a massive amount of files" in this context?  My main servers both have 22 TB of free space, which is vastly more than enough for me to be unconcerned about data volume, and I don't see any more than a handful of directories on the KODI hosts, so what would / should be my concern?  What does "side by side" mean in this context?

[jbinkley60]

I presume you mean getting the .jpg files onto my server?  That is pretty easy, really.  First of all, I installed Tiny Media Manager on one of my servers.  It does a nice job of creating KODI compatible .nfo files.  I created a short Python script that yanks all the actor names and links for actors with artwork on TMDB from all the .nfo files created by TMM.  These get dumped into a shell file like the following:

This was the part I was questioning, how you planned to get the actor metadata back into the Kodi database.  TMM will do it.   I personally am not a fan of the NFO approach and cluttering up my drives with them.  I prefer a direct database update, which is how I do it with the Mezzmo Kodi addon sync process but your approach will certainly work fine. 

I don't have any idea how TMM tracks artwork and will it go back and check for new images, which often get posted later.  At the point I am at with my library, having over 85K actors / actresses and 60K images,  I don't get a high percentage of new images nor new actors /actresses from new movies.


Thanks,

Jeff

[lesrhorer]

This was the part I was questioning, how you planned to get the actor metadata back into the Kodi database.  TMM will do it.

It will?  That, I did not know.  I am at the very least curious what mechanism it uses to do so.  Right now TMM knows nothing whatsoever about any of my KODI hosts and they know nothing of it.  I am skeptical I would really care to have TMM stuff the data onto the KODI hosts.  For one thing, running TMM is a bit inconvenient.  Not horribly so, but not something I want to do very often.  I would rather have more of a server utility.  Something that more or less purrs along in the background waiting for instructions and queries.

I personally am not a fan of the NFO approach and cluttering up my drives with them.  I prefer a direct database update,

Point taken.  I am of two minds on the issue.  I don't mind the clutter so much since there are only a small handful of directories involved.  Endlessly tripping across directory boundaries is a massive pain, but having them all collected into a small-ish number of directories creates less angst for me.  I also am not a fan of keeping all my eggs in one basket.  Having a monolithic database is definitely convenient, but when something whacks it, watch out!  Having everything contained in a centralized database, whether it consists of discrete elements or not, would be very attractive.  The KODI hosts would never have to scan anything to build their library.  They would just request information from the server.  That sounds like a lot of software work to me, however.

 which is how I do it with the Mezzmo Kodi addon sync process but your approach will certainly work fine.

Well, at the moment it is the only avenue I know is open to me, and it does work.  I am not so sure about "fine".

I don't have any idea how TMM tracks artwork and will it go back and check for new images, which often get posted later.

It doesn't as far as I know.  If we are talking about updated images, re-running the handful of scripts my system creates is trivial.  If we are talking about images added for actors who previously had none, I am not really concerned about it.

At the point I am at with my library, having over 85K actors / actresses and 60K images,  I don't get a high percentage of new images nor new actors /actresses from new movies.

Wow!  That is a lot of movies, unless maybe you have lots of Cecil B, DeMille movies with complete cast lists, or something.  Yeah, precisely.  In the movie database, I think I will wind up with perhaps 30K actors or so when the dust settles.  By definition, popular actors are in lots of movies, and consequently actors that are not in a lot of movies are not really what one would term "popular".  Most movies have actors who have already been in some number of previous movies. It is fairly uncommon for a movie to introduce a new soon-to-be-popular actor.  Thus, while my actor database may increase by a handful of names every year, most of them will sail below my radar, for a while anyway.

[jbinkley60]

It will?  That, I did not know.  I am at the very least curious what mechanism it uses to do so.  Right now TMM knows nothing whatsoever about any of my KODI hosts and they know nothing of it.  I am skeptical I would really care to have TMM stuff the data onto the KODI hosts.  For one thing, running TMM is a bit inconvenient.  Not horribly so, but not something I want to do very often.  I would Sorry, I meant to say I believe it will via NFO files not direct insertion, hence my comments right after about not being a fan of NFO files.rather have more of a server utility.  Something that more or less purrs along in the background waiting for instructions and queries.

Sorry, I meant to say I believe it will via NFO files not direct insertion, hence my comments right after about not being a fan of NFO files.
 

Point taken.  I am of two minds on the issue.  I don't mind the clutter so much since there are only a small handful of directories involved.  Endlessly tripping across directory boundaries is a massive pain, but having them all collected into a small-ish number of directories creates less angst for me.  I also am not a fan of keeping all my eggs in one basket.  Having a monolithic database is definitely convenient, but when something whacks it, watch out!  Having everything contained in a centralized database, whether it consists of discrete elements or not, would be very attractive.  The KODI hosts would never have to scan anything to build their library.  They would just request information from the server.  That sounds like a lot of software work to me, however.

Mezzmo has a great database, config and artwork backup utility.  I've used it for years.  Click a button and everything gets backed up with timestamps.  I generally keep around 10 backups.  Only once have I blown up the database and that was due to my own fault programming.  I am not a fan of all of the folders either.  I keep my movies in simple folders by genre and the artwork is in 2 folders, those which Mezzmo scrapes automatically and the UserPoster folder where I override the scraped images.  Mezzmo also has a lock feature which will lock an object (i.e. movie, episode etc..) so it can't be overridden by rescraping etc. 
 

It doesn't as far as I know.  If we are talking about updated images, re-running the handful of scripts my system creates is trivial.  If we are talking about images added for actors who previously had none, I am not really concerned about it.

I didn't think TMM did but if you aren't concerned about it, then it isn't a big deal.  That's where the Mezzmo Artwork Checker tool comes into play for me.  It continuously checks for missing artwork.

I am open to potentially adapting the artwork checker to your use case but we'd need to agree in the inputs / outputs and what you want to track.  Direct database insertion of actor information isn't really a viable option in a distributed client local Kodi database environment.  The Mezzmo Kodi addon handles all of this for me.  You'd probably be looking at NFO files and importing into each client separately, as you do today.


Thanks,

Jeff

[Hitcher]

That is not a personal criticism, but my first post in this thread made a query about why there were so few actor images in my KODI databases.  The first response (from Hitcher) was:

It's all down to the sites the scraper get the images from ie https://www.themoviedb.org/ and if they don't have them then Kodi can't store them.

Which I am afraid in context is pretty nonsensical.  Then you piled on to that response with the second response:

Like @Hitcher said, it depends on what images are available.

To me, it is completely and instantly obvious this has absolutely nothing whatsoever to do with my first post.  It is completely irrelevant.  This quite obviously has nothing whatsoever to do with actors who have no artwork associated with them in the internet databases.

Completely relevant given the information you provided.

I am not manually scraping the files from the internet...

No where in your first post did you mention this so how we were meant to know?

[lesrhorer]

That is not a personal criticism, but my first post in this thread made a query about why there were so few actor images in my KODI databases.  The first response (from Hitcher) was:

It's all down to the sites the scraper get the images from ie https://www.themoviedb.org/ and if they don't have them then Kodi can't store them.

Which I am afraid in context is pretty nonsensical.  Then you piled on to that response with the second response:

Like @Hitcher said, it depends on what images are available.

To me, it is completely and instantly obvious this has absolutely nothing whatsoever to do with my first post.  It is completely irrelevant.  This quite obviously has nothing whatsoever to do with actors who have no artwork associated with them in the internet databases.

Completely relevant given the information you provided.

No, it is not.  To my knowledge, both then and now, there is no way to change the behavior of the KODI's gathering methods using the standard configuration tools.  I have since learned there is at least one plug-in that does so, but had I installed any such plug-in I would have mentioned it.  Had I in fact installed such a plug-in and failed to mention it, then this misunderstanding would have been totally on me.  I suppose I might have said, "I and totally new at using KODI and all three hosts have default configurations with the exception of the media locations."  The only way this could be relevant is to assume otherwise.  The one probably pertinent piece of information I did leave out is the fact I am using .nfo files, and at this point this seems to be the culprit for the behavior, but it has nothing whatsoever to do with any lack of artwork on TMDB or any other data repository.

Let's put it another way: under what circumstances would it be possible, let alone likely, that TMDB is missing a vast percentage of actor's artwork?  I know of none.  I have looked at the various configuration options, and although it is very possible I might have missed it, but to my knowledge both then and now, there is no way to change the database in use by KODI.  This is a red herring.  If I am correct there is no way to configure KODI to look at any other database, then it is a complete red herring based on assumptions that can never apply.  If I am incorrect - and if so why has no one spoken up even now? - then the conclusion is based upon an assumption that could only apply in a very specific situation.  If the user was required to specify which database was being used by KODI, then there might well be an issue with the fan base itself,  but there still would be no reason to assume this.  This makes the notion at best unlikely to be relevant and as far as I am yet able to tell its is absolutely irrelevant.  I will be very happy to be wrong about this, so if I am, please enlighten me.

I am not manually scraping the files from the internet...

No where in your first post did you mention this so how we were meant to know?

Perhaps that was a poor choice of words on my part.  So far as I know, there is no way for anyone to configure KODI in a way that it thinks the user is manually scraping files from the internet.  The way that one knows the user is not manually scraping files from the internet is the simple fact one is running KODI, QED.  I am thinking about it, and I am hard pressed to think of how this would ever be relevant, given what I think I know about how KODI works.  Could I be wrong about this?  Of course I could!  Once again, if I am, then please enlighten me.  More to the point, however, is in what way would me downloading files from the internet prevent KODI from finding files on the internet?  Once again, putting it another way, which configuration, ostensibly in the settings menu, should one set in order to change KODI's mind about where to look for the files?  As far as I know, there isn't one.

After eliminating all the things that could never have been wrong in the first place, we can talk about what could actually be wrong back here on planet Earth.  Even at this point, it is still slightly speculative to suggest it is the presence of .nfo files.  It does seem rather likely at this point, so let us stipulate it to be the case.  With that assumption in place (but not forgotten), we are left with several possible conclusions:

1. The presence of any .nfo files at all causes KODI to quit its search on the internet.  This seems highly unlikely.  First of all, it would seem a very foolish way from a coding standpoint to address the situation.  Stranger things have been done, but the people who are writing the code for KODI seem far more savvy than that.  Empirically, however, even though every single movie file on my system has a .nfo file associated with it, KODI still spends a very large amount of time building the library database.  If it is not searching the internet, then what is it doing that takes so long?
   
2. The presence of an info file for a particular video causes it to stop searching for data for that video.  While much more likely than conclusion #1, it still seems a rather odd coding choice to me.  Judging by one of Karellen's replies, this might actually be the case, however.  The last question in conclusion #1 might still apply.
   
3. The presence of an <actor> tag in the .nfo file causes KODI to abort the search for artwork data for that particular actor.  This fits very well with the empirical data, that being KODI was finding a reasonable number of photos for less prominent actors.  Since the more prominent actors are well represented in the .nfo files, it seems to make sense the more prominent actors don't have artwork, while many less prominent ones do, because the list of actors in my .nfo files is not exhaustive.  One contradictory bit of data is there are some prominent actors who did have artwork.  Perhaps that is "bleeding over" from the TV database?  In any case, this would seem to be the most likely cause for failure to obtain artwork.  This, too, seems an odd coding choice, but then there could be some very solid reason why the decision was made to handle it this way.
   
4. I am missing something entirely.  It is even possible the .nfo files have nothing directly to do with the issue, although when I specify a link for the actor, KODI certainly propagates the information throughout the database.
   

I still have no firm information about which of the conclusions is correct.  They are all four obvious conclusions based upon the investigation done so far.

[Karellen]

@lesrhorer

I am sorry if I am being disrespectful, but now I am cringing everytime I read your new posts.
It just proves to me how far off base you are because you haven't been listening to us, nor reading any of the links we give you.
If only you put as much effort in reading what we give you, as you do in writing your posts, you would quickly find your posts unnecessary.
Your questions have been answered, yet you still keep implying how useless Kodi's processes are because you can't get it do what you want.

On a personal note, please cut down the overly verbose posts. I give up about half way through. Just keep it to simple questions.

The presence of any .nfo files at all causes KODI to quit its search on the internet.

Correct.

The presence of an info file for a particular video causes it to stop searching for data for that video

I am not sure what you mean by this.

The presence of an <actor> tag in the .nfo file causes KODI to abort the search for artwork data for that particular actor.

I don't really know what you mean by this.
If you have local artwork, and the actor is in the nfo file and either has or has no URL, then Kodi will use the local artwork.
If you don't have local artwork and no URL, no image
If you don't have local artwork but do have a URL, then Kodi will download the image. (if that image at the end of the URL is still available)
Geez, I have explained this to you already.

I am missing something entirely.  It is even possible the .nfo files have nothing directly to do with the issue,

What issue?? Sorry, but I have lost track of your so many issues and complaints.

although when I specify a link for the actor, KODI certainly propagates the information throughout the database.

Then I don't understand your problem.

under what circumstances would it be possible, let alone likely, that TMDB is missing a vast percentage of actor's artwork? 

Quite high. TMDB seems to discard a lot of artwork. One reason is that users upload junk, then it gets purged. Another is duplicates. Then there are the mass deletions that we never understand why it is done.
So if you are unlucky to have one of those links to the deleted artwork in your nfo file, then there is no artwork for Kodi to download.
If you are performing a normal scrape, then it is not a concern because only live artwork links are scraped.
Artwork links in nfo files are perishable. That is why I always insist users save artwork locally.

but to my knowledge both then and now, there is no way to change the database in use by KODI. 

Really?? TMDB, TVDB, TV Maze, IMDB, fanart.tv are all databases Kodi can use.

[lesrhorer]

@lesrhorer

I am sorry if I am being disrespectful, but now I am cringing everytime I read your new posts.

Don't sweat it.  I have not been here long enough to earn any respect.  Furthermore, unless I am fairly close to someone - a friend, family member, lover, whatever - I really could not care less whether someone respects me, or not.

If you have local artwork

You still have not defined "local" in this context.  Technically, nothing on my KODI hosts is local, unless you are considering a network resource to be "local".  That is not the way the term is normally employed, but if you want to use it way, then OK, fine.  It still, however, begs the question.  Via configuration, KODI allows two video database structures.  One expects one video file per video directory.  The other employs numerous video files per directory.  I am using the latter.  From my reading, I take it in the former case, the artwork for each movie is contained in a hidden .actors subdirectory of the individual movie directory.  I would expect this is what you mean by "local"", even though the entire drive system is not local, but rather a network resource.  I am inferring this to be the case, but nothing here so far has actually confirmed it.  Am I correct, or not?

This obviously is not a very efficient way of handling the situation, since any popular actor's artwork is going to appear 20 or even 30 different times throughout the file tree, but c'est la vie.  What, however, constitutes a "local" resource for a structure that has many video files in each directory?  Is there some place(s) where the .actors directory can go to make the artwork "local"?

By the way, it only just occurred to me some folks may actually have KODI physically running on the same machine that has the data structure mounted.  The notion seems rather strange, but it is of course possible.  That might be part of what is confusing me.

Geez, I have explained this to you already.

No, you had not and you still have not.  I cannot think of any definition of "local" that would allow me to infer the operation of KODI in consideration of the results I am seeing.  Every single movie has a .nfo file sitting next to it.  Yet only a comparative handful of actors have images taken from the internet.  I don't see how any of them can in light of your explanation.

What issue?? Sorry, but I have lost track of your so many issues and complaints.

I have no complaints and only one issue.  My three KODI systems are retrieving only a handful of actor thumbnails from the website, EXCLUDING those that have no artwork on the website.  So far I have downloaded the actors' artwork from 52 movies.  TMDB reports a total of 2145 different actors credited in those movies.  Of that number, 814 or 38% have no artwork on the site at this time.  The results you reported on github are similar, and not particularly surprising nor distressing.  Yet considerably less than 5% of the actors on my KODI systems have artwork.  Geez, I have explained this to you already.

     The presence of any .nfo files at all causes KODI to quit its search on the internet.
Correct.

That seems all but insane, to me.  Why would the system completely quit searching just because somewhere there is a .nfo file unrelated to all but one video, if even that?  What's more, that does not seem to be the behavior I am seeing.

I am not sure what you mean by this.

Well, it does seem very odd to me, but it is simply the second of four possible scenarios.  What's more, from what you have said, it sounds like it is possibly how KODI is allegedly supposed to work.  If the individual video file has a .nfo file, KODI doesn't bother to check the internet.  Again, however, that does not seem to be the case.  Every one of my videos has a .nfo file, but there are some actors with artwork.

Quite high. TMDB seems to discard a lot of artwork.

Not really.  Not compared to the level of missing artwork on my KODI hosts.  More than 95% is missing, and almost all of the more popular lead actors, supporting actors, and character actors are missing.  Now, I intend to fill most of this void myself, but that does not mean I do not want to know how KODI works and what options are available  to me.

Really?? TMDB, TVDB, TV Maze, IMDB, fanart.tv are all databases Kodi can use.

How?  I have looked all around KODI for a configuration field to change, or preferably add to, the sites KODI will scan when looking for artwork.  If you are talking about the ability to search for  artwork for a particular movie from inside the information routine, that is virtually no help at all.

[jjd-uk]

Local means anything not fetched from the Internet, so is available on your local device/network.

[jjd-uk]

Btw that's the only thing I got out of your last post as the walls of text in posts are very off-putting, I don't want to be reading a novel.

[lesrhorer]

Local means anything not fetched from the Internet, so is available on your local device/network.

This illustrates my point rather well.  All of the artwork (as well as every video file, every profile, etc.) physically exists on two different RAID arrays attached to two different Linux servers.  There are network shares of every file via HTTP, SMB, NFS, DLNA, SFTP, and Zeroconf, plus probably one or two other network resources I am not remembering at the moment.  None of these is as far as KODI is concerned in any sense "local" to the video files.  Certainly KODI cannot reach them at the moment except via URL links in the .nfo files.  That it does, but per the discussions in this thread, this does not constitute "local", so what does?  One large ~.actors directory sitting somewhere?  A .actors directory in each of the 27 movie directories?  A <moviename>.actors directory?  I could easily create 27 hard links (or symlinks, for that matter) to the /RAID/KODI/VideoDB/Thumbs/ directory, but is that the correct thing to do?  It would be a bit nicer than creating a URL link in the .nfo file to a file in the directory /Backup/KODI/VideoDB/Thumbs for each actor, which is what I am doing now, for a number of reasons.  Even if I choose eventually not to take the tack, I still want to know what the  tack is so I can makes a informed decision, and because if nothing else I want to know.

[lesrhorer]

Btw that's the only thing I got out of your last post as the walls of text in posts are very off-putting, I don't want to be reading a novel.

War and Peace is a novel.  My posts are as brief as I can make them.

This is an extremely annoying attitude.  I blame cell phones and the lousy state of our educational system.  My cousin get her degree from Oxford.  She had to write a 10 page literary paper every day and turn it in to her advisor, who was not even one of her professors.  In my job, I typically had to write anywhere up to 20 pages of technical notes every day, and read 50 - 100 more.  For leisure I read about 250 pages every day if I get the chance.

That said, if someone here would start actually answering my questions, perhaps I could make them briefer.

[Karellen]

War and Peace is a novel.  My posts are as brief as I can make them.

This is an extremely annoying attitude.  I blame cell phones and the lousy state of our educational system.  My cousin get her degree from Oxford.  She had to write a 10 page literary paper every day and turn it in to her advisor, who was not even one of her professors.  In my job, I typically had to write anywhere up to 20 pages of technical notes every day, and read 50 - 100 more.  For leisure I read about 250 pages every day if I get the chance.

That said, if someone here would start actually answering my questions, perhaps I could make them briefer.

None of this is needed. Its just extra text you write that clutters your posts.
Many of us don't want to read a novel as a post. We just want the question succinctly written and we will respond to it.
We don't want to read a thousand words just to figure out you are asking a question that could have been stated in 15 words or less.

this does not constitute "local", so what does? 

Oh geez.
Consider the following an and statement. Both statements must be met.
Local means:
1. On your local drive or a local network that Kodi can access, and
2. Files are saved in the manner designed for Kodi. That is explained in this and related pages... https://kodi.wiki/view/Movie_artwork

If you are hoping that you can dump a whole lot of images somewhere on your drive, and that Kodi will go hunting for them, searching every folder, on your hard drives or network, then you are sadly mistaken.
Do you now understand what local means?

[Hitcher]

All of the artwork (as well as every video file, every profile, etc.) physically exists on two different RAID arrays attached to two different Linux servers.  There are network shares of every file via HTTP, SMB, NFS, DLNA, SFTP, and Zeroconf, plus probably one or two other network resources I am not remembering at the moment.  None of these is as far as KODI is concerned in any sense "local" to the video files.

Are they on your LAN? If so then they're local.

What Is a LAN?

A local area network (LAN) is a collection of devices connected together in one physical location, such as a building, office, or home. A LAN can be small or large, ranging from a home network with one user to an enterprise network with thousands of users and devices in an office or school.

Regardless of size, a LAN's single defining characteristic is that it connects devices that are in a single, limited area. In contrast, a wide area network (WAN) or metropolitan area network (MAN) covers larger geographic areas. Some WANs and MANs connect many LANs together.