How to organize collections

[Culmor]

I have a serious collection of geological specimens amassed over many years that I intend to bequeath to a university. One thing universities don't like is a solely computerised catalogue system, technology changes too fast. A dab of white paint on each specimen, then a unique number in pencil and seal it with clear nail polish. Re the cataloguing, I use LibreOfficeCalc and print hard copies regularly.

Perhaps related, a east European guy came to read my electricity meter recently. He said 'You are professori?'
I said ' No. why do you think that?'
'Ah. the specimens, the books'.

[Like Ra]

I have a serious collection of geological specimens amassed over many years

Same 😊 Not a "serious" but still a collection. Quite unorganized I must say.

A dab of white paint on each specimen, then a unique number in pencil and seal it with clear nail polish.

That's something I definitely do not like. I prefer to keep my rocks as "untouched" as possible out of both aesthetics and "respect for the stone". A plastic box with a label - ideal. But I'm too far from this as well. And I hate glue.

the cataloguing, I use LibreOfficeCalc and print hard copies regularly.

Still have to do that for the rock collection...

[Culmor]

That's something what I definitely do not like.

I don't like doing it either but once your collection gets into three or four figures it's the only option. I'm as discreet as I can be but there's no other way I could keep track of stuff.

[cjtl]

For music I use MusicBrainz Picard to tag and put the files in the correct folders according to my own sorting method. Picard is cross platform. For listening I use Strawberry player. If you really want a nerd/geeks/OCD dream have a look at the terminal program beets https://beets.io/ You can write your own Python plugins for it too. I would use it myself but the initial setup and library importing would take me some time and I'd be constantly referencing the manual and tweaking the config file until I got it just right.

Photos all go in folders using this basic pattern YYYY/YYYY-MM-DD - Name_or_subject/photo_number_and_or_name.extension

For films/TV I use the Kodi naming conventions so it can look them up online. Films go in their own folder:
Films/Film Name (2021)/Film Name (2021).mp4
Films/Film Name (2021)/Film Name (2021)/Subs/Eng.sub
Films/Film Name (2021)/Film Name (2021)/Extras/Trailer.mp4
You need the Extras plugin and to change a setting by hand in a config file to use the Extras folder scheme.

And TV shows go in:
TV Shows/TV Show Name/S01.E01. Episode Name.mp4
TV Shows/TV Show Name/S01.E02. Episode Name.mp4
And so on.

They need to be separate in Films and TV shows folders so when you add the Films and TV Shows folders to Kodi the correct scraping method is used or you get really messed up results.

[Like Ra]

For music I use MusicBrainz Picard to tag

I'm disappointed with Picard as it crashes on my collection if used on more, than one album, does not support CUE, APE, WV, does not have the bit rate field in the GUI. E.g. Clementine and Quod Libet can do this.

Photos all go in folders using this basic pattern YYYY/YYYY-MM-DD

Same until this point 😊 Sometimes (for travel photos) I add the location. For tagging I use Digikam, which can do a reasonable job at face detection and tagging. QNAP Photo Magic can even recognize objects (but very unreliable) and even supports Google Coral TPU module, but I hate the interface.

[cjtl]

For music I use MusicBrainz Picard to tag

I'm disappointed with Picard as it crashes on my collection if used on more, than one album, does not support CUE, APE, WV, does not have the bit rate field in the GUI. E.g. Clementine and Quod Libet can do this.

Photos all go in folders using this basic pattern YYYY/YYYY-MM-DD

Same until this point 😊 Sometimes (for travel photos) I add the location. For tagging I use Digikam, which can do a reasonable job at face detection and tagging. QNAP Photo Magic can even recognize objects (but very unreliable) and even supports Google Coral TPU module, but I hate the interface.

My needs are simple so Picard is enough. I've not had it crash on multiple albums though. If I have a .cue and an ape,flac,wv file I use a splitter to process the .cue and spit out stand alone tracks. I still keep the original files.
Beets sounded just the ticket for me, until I found out how involved it would be to initially setup. What I have works and I don't have a lot of music so it's easy to manage.
So I'm not motivated enough to go through all the hassle. The cost would far out weight any benefit. I understand that some people with massive collections do find beets very useful though.

I wrote a simple Python script to copy photos off a camera using the path I mentioned, then hash the files and append the new files hashes to a list.
I wrote it before I started using a fully checksumming filesystem so as I mirrored the files to different PCs I would always know if one became corrupt by checking the hashes. Sometime the corruption in photos is not always obvious unless you overlay then and subtract the difference between them. And that's not counting metadata corruption. Having a list of hashes I can check them against does away with all that. My script isn't portable so I doubt it would be of value to anyone else. You could also use tools from the md5deep and hashdeep collection http://md5deep.sourceforge.net/

My Python script outputs hash files which are compatible with those tools, so to check the photos I just run a batch file or shell script which in turn runs those hash tools automatically for me. Writing that part as a shell script is preferable to trying to remember the switches as I seem to recall the order is important or something like that.
I still use that Python script as I still mirror my archive of photos on to a Windows PC, so having the hashes is still useful.