I've finally gone ahead and just moved my MIDI information (and related data) to a SQL database (using duckdb, which is more like PostgreSQL - apparently).

It's something I've been sorely in need of for a while. Most new information I gather about MIDIs is circuitous - if it wasn't, it would have been made available already. No documented name for this MIDI creator? What if I look at their friends, their friends' friends, find some general community pages... can I find their name then? A new URL? Maybe even a MIDI? That's generally my route. (The NIFTY Serve files are another matter entirely.)

The problem is that, to find new information from a dead virtual space, it's vital I record even meaningless, guestbook-level entries of names, emails, and info (not for public distribution or anything). It's exhaustive - Excel gets laggier and more cumbersome the more I add to it.

I don't have any confidence with code. I have a very limited understanding of the basics that I employ liberally when needed, and I can make a janky bash script when I need one. But nothing like SQL.

I spoke recently with Matt Sephton as I saw his username pop up on a couple spots of interest to this search. His area is specifically archiving old technology, largely Macbook-related material from Japan. I asked him for some guidance and tips and received some excellent (if troubling) information; but one big tip from him was just to keep looking for documentation, information, links etc. in similar software (as I've been doing with KISS dress up software and some programs specifically for NIFTY). SQL seems to me to be the most reasonable way to map this information as its scale increases.

So far, so good so what? It seems to be working well enough for now, but only time will tell. (I still need to add byte size...)

Of all things I expected to be well preserved, it was not the BBS boards and guestbooks slathered all over geocities and like-sites, but a lot of them have survived remarkably well (and contain a lot of really specific information). I'm hoping to target this information as best I can - we will see.

The good news is, I have more MIDIs to add!! I found one on a garbled page from an old employee of Monolith productions (and from the MUCK RP era, no less?) - a new cover of Rashiku Ikimasho by Yoshikazu Hirai, composed on the Roland SC-88, most likely. It's very exciting news... at least for me lol.

And just under a year ago, some were donated to me that were from a very unique site named Prose'n Poetry. For some reason, the site had poems dedicated to sailing, and then after having garnered a certain amount of Sailor Moon attention, simply pivoted to including it with a Sailor Moon section. That really tickled me! That website has some particularly unique later MIDIs, specifically from PGSM.

The last missing MIDI of Rini's also appears to have been donated as well!!

I'm hopeful that using SQL I'll be able to note easily whether something's been contributed, by who, and if it was original to the collection or added later.

There's the bad news, though. When I spoke to Matt, who was a massive help to me, I knew that distributing MIDIs is not legal (without paying hefty distribution fees) under Japan's copyright laws. What I didn't know was that this extends even to things like print media of and around software, and worse yet - that this extends to even things like scans of magazines. Where people can skirt these laws by using specific photocopies or sending photos privately, distributing a MIDI is distributing a file directly and people are still reticent to do it, for obvious reasons...

It's frustrating news. I spoke to him about whether similar archival efforts regarding older software are going on in Japan - where a lot of the old web was archived using toolbars like Alexa, this wasn't particularly popular in Japan, and many Japanese sites hosting MIDIs or "copyrighted" materials are outright excluded from the Wayback Machine anyway.

I don't think all hope is lost or anything, but I've never held out hope that all these files will be found. I do believe 99% of the files ever made are sitting out there somewhere, sleeping inside someone's Sharp X68K, waiting to be resurrected by the glare of an electron beam... but nothing lasts forever.