My iTunes music collection has over 1400 albums in it (no doubt plenty of people have more, but still, that’s quite a lot of music).

A couple of years ago I decided to re-encoded all my CDs as Apple Lossless, since the physical discs were going into storage and I wasn’t sure when I’d see them again.

I store my collection on a mac mini, which is my media server. However, I also had a copy on my laptop, which I kept as AAC 256k, to save space (and because, frankly, it’s hard to tell the difference).

I like using the meta data tags properly, and I hate it when people abuse the Album tag to indicate what disc it is (by adding “[Disc 1]” on the end), when they mark an album as “Compilation” when it’s a collection of songs by the same artist (for me that tag is only supposed to be used to group together tracks by different artists), or when they use the Artist tag to name check collaborators (by adding “feat. Joe Blogs” or whatever), so you end up with one album scattered across multiple artists (if you’re going to do that, use the Album Artist tag to unify the album under the main artist).

Over the time that I’d built up the collection, I had spent a lot of time editing the meta data to get it into the format I wanted it. Unfortunately, re-encoding everything undid a lot of this work, and left me with quite a few duplicates - some with my “correct” meta data, some with the rubbish tags from the internet.

I managed to clean up a lot of the problems on the mini, but of course that didn’t fix the laptop. Worse, thanks to iTunes match, the problems started to multiply again. iTunes Match on the mini has become confused on multiple occasions and “forgotten” me, so I’ve had to add the entire collection to it again. At which point it started adding duplicate AAC copies of every album where I’d edited the meta data.

I guess that this is because it had matched different encodings of the same tracks on different machines, in some cases with different meta data. The upshot is that now on my mini I’ve ended up with two copies of a lot of stuff, with both the correct and the incorrect metadata - and my iTunes match collection is now in an almost unbearable mess as a result.

I try to remove the duplicate, but they’re not always easy to spot unless I fix the meta data problems first, because they get filed in different places. I attempt to delete these duplicates from the cloud at the same time, but I’m quite scared that I’ll end up removing the only copy of something from the cloud too, by accident.

It seems to me that the root of this problem is that there’s no ‘authoritative’ place to view your match collection as it exists in the cloud. That and the fact that match seems to take the approach of avoiding touching your meta data whenever possible - which sounds sensible but isn’t if you end up with the sort of mess I’ve got.

What I badly need is a way to view the collection on the web, remove duplicates, clean up meta data, and then sync these changes down onto all my machines. I’m really not entirely sure how it deals with meta data changes right now - I suspect that it basically does nothing, which means that if you ever re-sync your collection, you end up with duplicates again.

For now, things are so messy that what I’d really like to do is delete everything from all but one machine and from the cloud, and start again. Except that I don’t really trust that everything that needs to be stored in the cloud is, and that it’s in the correct format, with the correct data, and that deleting it all from most places wouldn’t end up with me losing stuff.

In a word: “aaaaaaaarrrrrghhhh!”.