Last.FM iTunes Informer (My Software)

in Mac Software edited January 2014
Hello everybody,

a few days ago I started a new project that aims to enrich the tag information of your iTunes songs.

An older attempt was the iTagProperly that adds style and mood information in the comments and grouping fields of a track. This information was taken from but it wasn't that much useful because: allmusic does not have an API to obtain its database infromation so I had to scan the HTML code to extract the styles and moods and this system would break every time allmusic gets an interface redesign. Because of that the user had to tag tracks one by one. On top of that information could be obtained only per artist and not per track.

Now focus is on Last.FM. Last.FM for those who don't know it is the largest music community that tracks what people listen to using several plug-ins for the most popular players and starts building a neighborhood around the user and his/her's music taste. It also allows the user to tag whatever track with any keyword they please.

This information is available through web services in XML format. It provides the most popular tags for every song known to its database. This information is being used in Last.FM iTunes informant in order to enrich an entire iTunes library per track with the tags specified by thousands of people. Massively. Needing only a couple of clicks.

Imagine the playlists you could produce after your music collection has been tagged this way.

The very first premature version is available for download.

What I need from you is:

1. To find a fancier name.

2. To check Last.FM's web services just in case you notice another web service that could be useful apart fro the "Top Tags" for "tracks". that I am currently using.

3. To try the application is several of yous songs and report back here any issues you might be having.

4. To find someone who can make nice OS X eye candy such as icons and logos.

5. Not to ask for a Windows port. I don't want to do it but if someone is willing to I am glad to provide some guidance.

Version 0.1 Universal Binary

This version does not communicate with iTunes and it does not support unicode characters (I think so). Its for premature testing purposes only.

Some features and options.

- Sleep Interval in Options: ?? Last.FM does not allow the same IP to contact theis servers more than once per second (it bans it). Using this option the app waits for the seconds defined before every request.

- Maximum Tags: Here you can define the maximum numbers you want per track. The most popular ones have priority. If you get less tags than you defined then the song actually has less tags on it.

- So, just type Artist and Track press the Last.FM button and check the tags in the Tokens Field and in the Text Field. In the first one you can change the tags sequence and the text field one has the tags comma seperated just as they will be written in the comments or grouping field in iTunes.

- Tags in Last.FM are all lowercase. For aesthetic reasons I capitalise the first letter of each word.

I haven't found a way to make the Tokens Field multiline (as it is in the addresses field in mail) It must be pretty hard as I use Java in xcode to develop the app. I have only found ways to do this in Objective-C but I don't know any Objective-C. If anyone can help please let me know. I will soon implement the iTunes connectivity.

Please give it a try and come back with bug reports and suggestions!



  • Reply 1 of 2
    It's an interesting idea, but without iTunes connectivity it's hard to tell how useful it would be. As far as the interface, it's awkward to have the " Tracks" button above the text input fields, and the text box at the bottom seems redundant if you're able to see all the tags (ideally) in the tokens field.

    Unfortunately, as a standalone app I don't think I would use this much, but it would be a nice feature to have in the official app; it could tag each song in the background as you play it, the same way it reports your recently-played tracks to the server.
  • Reply 2 of 2
    i like the idea of it, but i find's tags to be too vague or inaccurate for my taste. i'd really love something that works with allmusic. i'm surprised they haven't licensed their database to someone for this purpose.
Sign In or Register to comment.