Well, hopefully it does well in the UK and they start selling them over here! Yamaha is on a roll lately.

TENORI-ON [Yamaha]

So it's Japanese language only, and only available in Japan so far. But everyone needs to listen to this...

VOCALOID 2 [Gamersweb]

I recently had a discussion with Kyle Johnson on the dilemma of whether to refine a song, or abandon it and start over. We agreed that all his unfinished songs were little sources of guilt--that rather than start a new song, the old ones should be finished. I've felt this way about video games as well--why buy a new game when there's a stack of in-progress ones?

Before I started making music professionally, I had tons of little stubs and stems of songs littering my hard drive, never to be finished. The purpose of these songs wasn't to create a new complete song for others to listen to, but rather it was to explore. They led nowhere but that makes them just as valuable as the songs that were fully completed. The problem is, you don't know when you set out what type of song it's going to be. And you get attached to the song as you write it.

It's difficult to abandon this thing you've poured hours of work into, especially when it sounds almost good enough. You've become fully invested in it and you don't want to say all those hours were wasted. Here is where you tell yourself, it's not always about the end result but more about the experience.

The trouble is, the longer you spend working on a song, the more you want it to work. Before long, the main reason it is still being worked on is that it's been worked on so long. You hold out the hope that it develops. And with each hour it becomes harder to let go.

It's comforting to visit gfx artists' websites and find pages of unfinished sketches. Why isn't this more acceptable in the music world?

I don't have a solution yet, but I am going to try evaluating songs as they are made to determine, "is this a song, or is this a sketch?" Recognizing that sketches have their own value as part of the process, I might be more willing to give up on them.

The same is true for parts of a song: the segment you wrote that sounded really cool on its own, but doesn't fit in with the rest of the song at all. It's so hard to let go of it and exclude it just because it doesn't fit in.

To combat the attachment syndrome, I'm going to try making some songs that are designed to be throw-away from the start. As a side benefit, it's easier to start on a song when you know your first attempt will be thrown out, and there isn't so much pressure to create something brilliant in the first stroke.

To summarize, here are some theories I've come up with that I'm going to explore over the next few months.

• All songs start out as sketches. Only some become "releases."
• Be willing to break free from a song and start over, no matter the emotional attachment.
• It's faster to write three sketches and pick the best one than to try to refine a perfect song in the first go.

How do you guys deal with this attachment syndrome? What are your techniques and how well do they work?