Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[bopz]

Matching Lookup data with tracks, using Freedb and/or YADB
and mixing that data together, and updating the servers.

MC looks great and I've been playing with it all day, comparing it to some other tools as I organize a large audio library.

However, the way it works for lookups is (uncharacteristically) simplistic.
One of the tools I've been testing is
http://www.wensoftware.com/MusicLibrary/

Here's an example of the kind of thing that I think might work better, while still allowing proprietary databases to grow.

Let's say we could use a lookup tool that does multiple lookups (shows tabs for YADB, Freedb and others), presents the user with options and allows track and album mapping in various ways before committing the changes to the database and writing the mp3 tags.

Basically your selected tracks could come up on a right-side list, underneath search fields (which should probably default to the current album or artist, if known)

Above that, there could be three tabs, one for YADB, one for Freedb searches, plus other databases under other tabs (AllMusic would be good for webpage reference.) Selecting these tabs allows the search fields to get different results.

The lookup search results produce a list of albums (or tracks.) Double click on one of the items in the list, and the track(s) from that item then fill the left-side fields next to the mp3 tracks you have already selected from your collection in the right-side fields. You can re-sort the right side list, re-order it by moving tracks up or down, match them to the left side tracks by manually clicking or by matching all strings in the titles or by matched track order. You can exclude selected fields from being replaced. You can go back to your lookup-search results list, to try out a different entry that may better suit your preferences (nicer English, better nomenclature, whatever).

When happy, click the commit button and this does what you've set up in the options, for example:
-writing id3 tags with the new info
-updating YADB with the info (this could be especially useful since the data may now be a mix of different databases plus previous user edits and entered data.)
-running the MC file renamer with its template

[jgreen]

FWIW, the only time I found yadb lacking was when I ripped the CD, "White Town/Women In Technology".  Granted, yadb found the tagging info just fine, only it couldn't tell me why I bought the album in the first place.  Live and learn.

I'm fairly new to the forum, at least compared with the fossils, but I can tell you already that the  subect of yadb vs freedb or cddb qualifies as a full-fledged "sore point".  jriver has yadb, and they're stickin' with it.  Originally, yadb was started with some version of freedb, I'm told.  Many of these same fossils spent long hours of their own time cleaning up the tag info, and adding plenty of their own.  The yadb database is extensive, and I've never had an issue with it.  But others occassionally have, mostly with obscure stuff that they later found on freedb.  Why yadb cannot continue to update from freedb is a mystery to me, although I have a hunch it has something to do with open-source issues, etc. 

Nothing's preventing you from ripping to a cddb- or freedb-enabled program, and then importing into MC.  But I wouldn't hold my breath waiting for jriver to embrace freedb, or pay for cddb.  Although if you're determined, you can add this feature request to the "feature request--MC 12" thread, currently on page 9998.

[bopz]

Why yadb cannot continue to update from freedb is a mystery to me, although I have a hunch it has something to do with open-source issues, etc. 
....
Nothing's preventing you from ripping to a cddb- or freedb-enabled program, and then importing into MC. 

It's not ripping, it's finding tags for a lot of disorganized mp3s.
In this case, viewing multiple options from a database is neccessary.
The best way to present multiple choices for matching selected tracks from a music library to an online database, YADB included, is something like the lined-up side-by-side matching I described above.

As for multiple databases, if we had multiple-choice matching, and the ability to mix choices, exclude tags while matching, then submit the corrected data both to the user's library and the online database, this would basically amount to a free synchronization service for YADB with other databases. There shouldn't be a legal problem for JRiver since it's just people submitting the data as usual, just with more information at their disposal.

I've had mostly dud seeks to YADB, mainly because of the messy tags and album info already in the files. Also, I need to see the original album lined up with my list of tracks side by side. That's why something like the Music Library program I pointed to above has been helping, though it's been slow getting MC to load the updates.

The other problem is the lack of multiple choice. I've had to pick between several releases of an album, or several freedb submissions of the same CD, to get the right data and in the formatting that I like.

[tcman41]

To me the only way to get 100% correct tags to one's satisfaction is to get a good mp3 tagger and go through the collection once, get it correct and then before any additional files are added simply run them through the tagger and this way one's collection will always be correct as far as tags go. I've managed my collection this way over the last 8 months overseeing the addition of 12,000 files and my collection is clean.

The tagger i use is here:

http://www.zortam.com/

TC

[bopz]

I'm fairly new to the forum, at least compared with the fossils, but I can tell you already that the  subect of yadb vs freedb or cddb qualifies as a full-fledged "sore point".  jriver has yadb, and they're stickin' with it.  Originally, yadb was started with some version of freedb, I'm told.  Many of these same fossils spent long hours of their own time cleaning up the tag info, and adding plenty of their own.  The yadb database is extensive, and I've never had an issue with it.  But others occassionally have, mostly with obscure stuff that they later found on freedb.  Why yadb cannot continue to update from freedb is a mystery to me, although I have a hunch it has something to do with open-source issues, etc. 

I need to return to this point.

As I've said, If users are voluntarily updating YADB with metadata which they may, or may not have, downloaded from freedb, then this amounts to a free synchronization service for JRiver.  All JRiver has to do is make this process easier.

There is always going to be various bits of metadata missing from YADB, or any other database. Genre info, spelling errors, year info, artist info, just to name some things I've noticed while ripping and tagging. 

It would be easier to correct all these online databases if information could be easily compared between them. Ultimately someone is going to figure out how to market data such as: the majority opinion of genres, artist bios, critic's reviews, notes, blah blah blah, and then the proprietary database owners are going to have to pay to aquire that in order to keep up with the Joneses.

If some marketing geniuses figure out how to make all this metadata available easily and at low cost to big media companies and utimately to the average people using media software then I'm all for it. But if no-one is actually doing that, then what's the use of keeping YADB behind a wall?

[John Gateley]

Thanks for the ideas, we are looking at improving YADB in the next version...

j

[Alex B]

You could make the YADB submission from existing tracks a low priority backgound task. Currently it reserves MC completely.