I assume you meant 489 Gb
Problem is backing it all up means you need a terra- byte or risk your collection being lost as soon as the disk goes belly up
Have had 3 external drives go over last 3 years (all within warranty but so what if data is lost)
Beware MediaMonkey. I downloaded it because of this entry and after tagging many files with the Amazon search I noticed that it was renaming the track titles. Now I don’t know how many tracks I have which are mistitled 
So turn off auto tagging the tracks!
The Library view in Windows Media Player shows the full file path of each music file. The File Name field can be dragged to the left so that it is immediately viewable.
And think of the problems when you get to 50k tracks or more which is easily (ahem) achievable given current technology …
Most programs like MusicMonkey that maintain internal databases of the tags and tracks (for performance) start to crawl at that speed.
Windows Media Connect which I use for my Roku Soundbridge iterates over the entire collection every time I restart my server. Ouch!
re: Musicbrainz.
“The audio signature matching is the only unique feature”? The audio signature matching is in flux and Picard worked happily for a long time without it.
“the only unique feature” of musicbrainz is surely the massive, reliable, user-moderated and regularly updated databse of albums and artists that blows cddb away for reliability. (PS try the old pre-picard tagger application. Some people still prefer it to Picard)
My library includes about 900,00 songs. (3TB) on a 3 box network.
I too am fanatical about tagging songs well. To do that I use DrTag (from NL). It is the most intuitive at extracting data from the file structure and doing swaps of title\album, etc. There are several config feature too once you get used to it. The best $20 I ever spent. BTW - I swap all names library style e.g… Beatles, The, Lennon, John, etc. This gets real tedious with all the “THE” bands out these days :).
Generally I divide all music into Genres (Metal, Folk, Electronic, etc.) until a band has 4-5 CDs or about a gig of material. At that point they go into the A-Z by artist section.
I also have about 700 DVDs of backed up music data.
Too keep track of all that and to find a individual tune across all the discs (fixed and removable) I use MyMp3Organizer. With the database itself running at over 1 gig, it is the only program that will handle such a large number of records. MS Media player simply stalls out at around 150k records.
In this program, once you have run a search (on any of the ID3 tag fields), you can simply click and play, drag into Nero, or modify the tags. It will also move and restructure as needed.
the TPOS tag, … does not appear anywhere in MediaMonkey
It does in MM V3 - as Disc#.
Why have you all gone quiet? It’s been really interesting and i’ll be following up loads of ideas from so many of you.
My takes are:
A hell of a lot of thought by some mega-brains went into ID3 v.2 BUT most of the organisations and coders that supported it were pin-head-brains (e.g. only one genre per track). God, i worked with a bunch of shit-hot coders (i.e. ADHDs still wearing diapers) and despair they’ll ever do anything useful or robust.
We all want different stuff from music databases. For myself, i want to focus on works or songs that may be performed (sometimes more than once) by different artists. Have given up expecting something commercial or even shareware to fit the bill, so i wrote myself a database using Access - NOT the best, but the quickest 
After all these years, why the hell aren’t all media players and interfaces ALL UNICODE? One of my favourite songs has versions in Thai, Vietnamese and Cambodian.
Sorry, just had to get these off my chest.
MyWeirdDefinitionOfGenre\Artist - Album\Artist - Album - Track - Title.mp3
iPods use their iTunesDB to organize the files. This means that it’s possible to have duplicates of albums. When you have an album twice, it just lists every song in the album twice. That, I hate. So much, in fact, that I set out to hack my damn iPod. I want to use the iPod like the USB disk it is, then run a small utility to update the blasted iTunesDB.
I find that organizing by file name is more reliable for detecting duplicates - there’s no room for two times a Sepultura - Roots folder, for example. Inside that folder, I cannot have two times 01 - Roots Bloody Roots.mp3.
I use mp3tag to convert the filenames to tags. Also, I use a homegrown file renaming tool called Filename Facist that uses super-sed (I’m a programmer) for renaming files. Sadly, it stopped working on Vista. So now I have to use Total Commander multi-rename-tool for applying my facism to the file names.
I’m still looking for a program for downloading release years and album covers from the interwebs. MediaMonkey does a nice job, but it isn’t always fetching the correct release year.
I’ve been thinking of writing one myself using Google Image Search and some computer vision techniques…!
Very nice XD
Wondering though, what’s with the seemingly random pic of you? and David Lee Roth? Which is, fantastic BTW.
Have you ever tried mp3tag?
This tool is very user friendly. Especially when mass tagging! It auto-focusses on the right fields when performing actions and can also use CDDB queries to tag files on a artist/album search!
The only disadvantage is that it rebuids it’s index at every start. This really takes some time if you have a large music collection.
To be frank, you are underestimating Chinese songs, J-POP, and K-POP. Put together it’s very likely they have more than 50% market share.