Amarok/Archives/PlutoJournalInterview

This is an interview for Pluto Journal, an Italian online magazine. Max, Muesli, feel free to participate.

Simply describe who are you (age etc.), where do you live and what do you do in your life

Max: I'm 23, from the UK and finished my Masters in Chemistry last July, since then I have been living on the cheap working on amaroK full-time. Soon I'll have to get a part-time job to make ends-meet, but I've pretty much decided that working on free-software is the absolutely best thing I can do with myself, at least at this time. Working with complete creative freedom is fabulous, I would find it hard to work on commercial projects where I have less input, and goals other than making the best possible product dictate the development direction.

Muesli: I'm 24, mostly known as Muesli, even though my real name is Chris actually. I grew up and I'm currently still living in Germany, but I tend to move to another country sooner or later.

Mark: I'm 29, living in western germany. I'm currently spending my time hacking on open source projects.

Why do you work in a free software team?

Max: Your only limitation is time.

Muesli: It's one of the rare situations in life, where people actually really try to work together and help each other out. Combined with the constant flow of user-feedback, this keeps you motivated all the time.

What do you do in the Amarok project? Speak about other team members..

Mark: I'm a developer, and these days I'm also doing lots of organizational stuff, like mailing list administration, coordinating releases and so on. As a rule of thumb, the bigger a project gets, the more time you will spend with administration, instead of doing actual programming. We try to solve this by getting our users involved in such tasks as well, like maintaining our bug database.

Max: I develop what I can. But as Mark says, you end up spending more and more time on administration. I'm not really very keen on managing people, but my desire to release a perfect product is overpowering, and you end up trying to persuade people to fix things, or asking for favors. So I'm torn between the twin desires of writing cool code and producing the world's best media-player.

Why another multimedia player?

Max: I merged my reply to this question with the next question.

What do you like of Amarok

Mark: Now that we've released amaroK 1.2, I can say I'm really happy with it. It was a lot of work, as the three of us had basically been working fulltime on it, but it was worth it. But the best thing about amaroK is our team, really. There's this holy trinity of amaroK core developers: Max, Muesli, Mark. The mighty MMM ;) We just got along very well from the first moment, it's great working with these guys. And we have an ever growing community of users, some of them really involved in the project, which is really cool.

Max: I've used a lot of media-players and never found one I liked.. until I came across amaroK. I used it for a few minutes and realised that markey had developed a player that put playlist management first, and not looks or some other gimmick. amaroK also is good because it doesn't force you to only use a database, it puts equal priority to other access methods, eg playlists and your filesystem.

Also I have to agree with markey, we do have such a good development team, all passionate about the software, all workaholics, all prepared to listen to each other and our users. Also they calm me down when I get stressed out ;)

Finally markey's attitude to contribution has really helped make amaroK what it is. He has always said "Just commit it", and this certainly wouldn't have been my attitude if I was in charge (back when I started I mean). But it is a fabulous attitude with open-source. It makes our betas buggy, really buggy sometimes, but we have embraced some real innovation from random contributors, and by the time the stable point-O release comes round we have sorted out the bugs, streamlined the interfaces and made amaroK much better all round. More open-source projects would benefit from a more open contribution-policy in my opinion.

What do you dislike of Amarok?

Mark: I'm not so happy with the current handling of non-local media. With non-local I mean stuff like playing music directly over ftp or http, which is possible with amaroK. Problem is, KIO, which we use for data transport, is not very well suited for multimedia needs. Let's for example look at digital Audio CD playing, you can do this just fine if you have the AudioCD-Slave installed. But you can't seek in the track, because the KIO authors never thought this was useful.

I guess everyone would agree that listening to CDs without being able to seek in the track just sucks. So we can just hope that KIO will improve one day, or use a more suitable VFS subsystem, if such a thing exists at all.

Max:The fact we are flexible in which multimedia-framework and database you use is excellent, I wouldn't have it any other way with the mm-frameworks at least, but it also makes bug-hunting very hard. Also we have very little control of the frameworks, and when they break amaroK get's the blame, and there's often not much help we can offer the user.

Next things to do in Amarok?

Mark: For 1.3 we already got some funky stuff on our TODO, like Replaygain support and an extended playlist. Stay tuned.

Max: I want to see what can be done if you make the playlist component maybe a plugin, or maybe just ship a few types of playlist. For instance a playlist that you drop albums into, not tracks. Or a playlist that is dynamic and built based on suggestions from audioscrobbler and some smart-playlist logic. I think this could be quite exciting, but at the very least it means I can refactor the code for the current playlist and add some much needed features to the "default playlist-view" (like differentiation for http-streams and normal files).

muesli has some good plans for adding removable sources to the amaroK database, which should be very popular I hope :)

What is Radio anoraK?

Mark: Radio anoraK was a bit of a fun idea. At some point we thought it would be nice to start our own amaroK radio stream. You know, we're all pretty much into music, playing stuff all day while programming. As amaroK already offers good streaming support, the next logical step would be to make a stream with the music we like. We could take turns in DJing, and also accept users' playlists and stuff.

The problem is currently that organizations like the RIAA demand quite a lot of money for setting up a public stream, and of course you need plenty of bandwidth if the stream becomes more popular. Perhaps we'll do it one day, who knows.

What do you think of KDE project?

Mark: KDE just rocks, I've been a fan and user since the beginning. First KDE version I used was something like 1.0-beta1. I remember back then reading an article in a german computer magazine, and as I was running Linux already, I just had to try this out. From that moment I got hooked and followed KDE development closely, reading the mailing lists and watching new releases. I was always hoping to contribute one day, well, and with amaroK it finally became true.

Max: KDE has got a lot of stick of late because elements of its usability suck. Which is true, but it is improving, probably driven by the excellent work GNOME have done. But KDE is a really powerful framework. I always enjoy watching my Dad updating his website: copy a file to local disk, edit in notepad (eek!), and then copy back to the server. I just open Kate, point it at the the ftp server and edit all the files directly. It's just wonderfully easy and convenient :)

I think the future is in KDElibs, KDEbase and extragear projects. amaroK is an extragear project, we are an official KDE product, but we have our own release cycle external to that of KDE, which means we release more frequently, and have a lot more flexability. K3B, amaroK, Digikam, Gwenview, Konversation, Filelight, these are all some of the most popular products in KDE, and all of them are extragear. I don't think this is a coincidence. It is important that KDElibs and base continue to release annually or biannually, but everything else should be split out and developed on their own cycles.

What do you think of GNOME project?

Mark: First of all, I'm happy that the GNOME projects exists. With Open Source Software it's not much different from commercial software: Competition usually improves quality. I'm pretty sure both KDE and GNOME wouldn't be what they are now without a worthy competitor that drives innovation.

On the technical level, I dislike GNOME's approach of using C and GLIB for development of large applications and frameworks. I believe this is a very outdated and clumsy technology. It's really much more elegant to use a real Object Oriented language than trying to emulate those features in C.

Max:The competition is absolutely vital for all the desktop environments on *nix. Otherwise I don't know very much about GNOME. From my use of GIMP 2.2.x I know I'm impressed by some of their new UI pieces. I'm sure it won't be long until I try out an Ubuntu live-cd or something similar.