It’s not a surprise that I love Amarok. Take a look at the related posts below. I think nothing beats it as a music program. If the KDE4 for Windows libraries were more robust I’d recommend it for everyone. As is, I think it’s the best on Linux. But, until the most recent updates there was something lacking that really bothered me. On a stream like NPR’s All Music Considered it wouldn’t update the Now Playing track to the song that was on. And that made it really hard to use that stream in the best possible way - to discover new music! It’s fixed now and that’s awesome! Woohoo!

A few months ago Lifehacker had people post images of their Smart Playlists. So I decided to share some of my Dynamic Playlists from Amarok so that others could see the syntax needed for various types of playlists.

Throughout the month of October I alternated between listening to newly acquired music (of which I had quite a bit) and a dynamic playlist that was weighted so that most of the music had auto-ratings above 90, less music above 75, and even less above 50. Any song that I listened to only once would fit in the third category. After that, if I skipped it just once it would fall out of the group. So it gave me a nice mix of music including forgotten favorites. Now that I’ve listened to a good chunk of my library since the last time the Amarok database was corrupted I may modify the playlist to include the caveat that it couldn’t have been played in the last x days. I’m not sure if I want to make the date large enough that I only hear songs once per quarter or my standard two week waiting period. Given how little time I have to listen to music nowadays I might go for the 90 days. I also spent some chunks of time listening to my music on random on Google Music. That’s completely random so I end up hearing some songs I didn’t even remember ever acquiring. A lot of my new music came from Jamendo when I got an email with the month’s most popular songs and I decided to do a search for ska. Although I really enjoy ska, I got into it rather late so I don’t have very much ska music. (Mostly just Five Iron Frenzy, Save Ferris, No Doubt, and the Orange County Supertones - who are an OK ska band - I mostly like their least ska songs)

Recently I was looking at this old post and the screenshots of Amarok 1.4 reminded me of something I loved about that version of Amarok - the stats that would display when you were playing a song - like if you were playing a Five Iron Frenzy song it would tell you your three favorite (or most played) Five Iron Frenzy songs.  That functionality never made it back into the Amarok 2 series.  So when I was looking through the Amarok scripts in the script installer I came across Rating Statistics.  It basically gave me everything I wanted other than being linked to the currently playing song.  It allows you to search your favorite songs based on all kinds of criteria like number of plays, user-assigned rating, or auto-score.  You can also assign some weights to it and get your best songs based on that weighting.  Here are the stats about my library:

I’ve spoken about Amarok’s Dynamic Playlists before.  Despite my best attempts, I couldn’t quite get it to work correctly.  However, Amarok 2.4.3 has revamped the dynamic playlists engine and interface, so I wanted to check it out and see if it worked better as well as testing out some of the new features.  For one thing, they changed the interface to more closely match the interface for static playlists.  See:

I started using KDE in November of last yea r so I figured that I’d give an update on how things are working for me four months in.  First off, KDE 4.6.x has not yet hit the official Fedora repositories.  Since I like to yum upgrade or preupgrade from release to release, I try to stay with the official repos and RPMFusion.  So no KDE 4.6 for me.  At this rate, it doesn’t seem that it’s going to make it until around Fedora 15.  But, if that means they iron out any extra bugs, that’s fine with me.  So, with that said, let’s get to the info.

One awesome thing that is easy to notice in free/libre software is how international it is.  While proprietary software is mainly based out of the US - Windows/OSX - free/libre software comes from all over the place. Mandriva is based out of Brazil and France.  SUSE was originally developed in Germany.  Miguel de Icaza, one of the founders of Gnome, was born in Mexico.  Choqok, the best KDE-native microblogging software is created by an Iranian.  So something that Linux has always done better than Windows is support more languages. Microsoft has to pay to create language translations so they have to make a market analysis about which languages to support (and it still doesn’t cover non-Microsoft programs)  With Linux, it’s all volunteer work (or paid by companies that care about localization) and if the programs are written correctly for KDE and Gnome, they will all be able to take advantage of the translation work for their program.  “Save” should probably translate well across all well-written programs.  I think this is one of the reason why all the regions of Spain have their own Linux distros.  I don’t know this for a fact, but I would guess that Windows probably only comes out in Castillian (official or regular Spanish) and not in Catalan, Andalusian, Basque, etc

So, as I mentioned before, I wanted to try and make sure to get mostly unheard music on my random playlist so I could go through all my music.  So I put in a bias to make sure that there was an 80% that the next song picked was unplayed.  I started getting even more played music showing up than before!  I was baffled!  Then I realized it was my piss-poor understanding of statistics at fault.  Telling it that there should be an 80% of the music selected having never been played is equivalent to saying there should be a 20% change that the music selected should have been played before.  What the heck is going on?  So, if you look at my collection - when I moved to KDE 4, it lost the previous stats.  So all of my music was unplayed.  Let’s say I have 10,000 songs.  So when I tell it to randomly play music (with no bias), there’s, at first a 0/10,000 chance of a played song coming up.  After that, there’s a 1/10,000 chance of a previously played song showing up.  After a day, there’s a 26/10,000 chance of a played song coming up.  So because of randomness I could end up with previously played songs.  Stats only tell us what will happen in the long run.  In 10 flips of a coin you could get 10 heads.  But in infinite flips you should get half of them heads and half tails.  26/10,000 = 0.26% chance of a previously played song coming up.  So, when I set the bias so that theres a 20% of the songs that come up will have been played, I’m greatly increasing my changes of hearing a song again vs just leaving it on random!  At least for now.  Once I’d listened to most of my collection, it should flip.  There will be so many previously played songs that those will be random enough to satisfy me.  Right now not enough of them have been played relative to my collection’s size so the same ones keep coming up.

So I’ve used KDE for about a work week.  During that time I’ve pretty much gone to using the KDE versions of all my programs except Konqueror.  I’m not sure if the Fedora 14 version of Konqueror is the one with Webkit, but last time I used Konqueror with KHTML it was mucking up a bunch of web pages including my blog.  So I stuck with Google Chrome, which is what i use on Gnome, LXDE (Lubuntu on my laptop), and on my Windows 7 install.  (Also, I stuck with gPodder for podcasts because that’s working perfectly)  So how did it go?  First of all, I love the stock screenshot tool in KDE, KSnapshot.  I love that lets me choose full screen, region, window under cursor, and section of Window.  With Gnome I hit print screen and then I have to edit the png in the GIMP.  So it gives me less work for my Linux-related blogging.

There was a time when I thought Amarok was the best music player on Linux.  I even used to run it in Gnome as you can see from this 2005 screenshot.  In that first link you can read me gushing over Amarok 1.4.  I loved all the integrated technologies, especially the metadata juggling Amarok did.  The first few Amarok 2.x releases with the KDE 4 libraries were complete crap.  They were ugly and were missing nearly all of Amarok’s features.  (Mirroring the complaints people were having about KDE 4 at the time)  When I took a look at Amarok and KDE 4.4 in October I said I would take another look at Amarok.