Creating music is an organic process, part of which is uncontrollable and mysterious (as a good mental practice, I take good care of reminding myself very often that here I'm the lightning rod, not the lightning, and I pity those musicians who see it otherwise).
I had an example of such uncontrollability this very morning, when in the first mists of awakening I overheard distinctly an instrumental tune. I wasn't feeling particularly musical this morning, in fact I was like "What? Now? Couldn't you come back on Thursday?"
Anyways, at this point I've gone through the process lots of times already, so I knew the drill; I can moan all I want about it, and it's true that it's cold outside the blankets, and that I have to pee, and I'm hungry, but that chunk of music in my head at that moment is like a beautiful figure of ice that is melting quickly as I stroll, and in the crucial period before it leaves, all that matters is if I will get to make a sketch of it accurate enough so that it makes more sense for me later, when I have actual neurons to work with.
My startegy for today, when this happened, was running in my head quickly the beginning of a small PDCA cycle, the Plan part, listing:
1) Tools I will need this time: unplugged guitar + the tablet. Recording software in the tablet: this piece absolutely absolutely required recording 2 tracks (sometimes 1 will do), so the "2 Track Recorder" Android app
2) How to go about rehearsing; this one is always a balancing act depending on the technical difficulty of the music given. On one hand, any time you spend rehearsing before pressing Record is ice melting time, and mistakes and forget creep in (in this particular song, I lost a crucial bass arrangement; I'll never know how it was, so I had to replace it with an alternative that I also like, but which I'm sure is not as good the original).
The other option is start to record from the get go and record all the mistakes in, like "rehearsing live" until you finally get the parts right. The risk here is that in such a long track, when you have to relisten to it after some time has passed, the actual real riffs and structures you wanted to play appear diluted among a lot of "almost there" fragments, and it's sometimes hard to differentiate. Also, when you have two tracks you want to record like in this case, it's not practical to record such long fragments.
I'd estimate this time I got away with a 12% loss of initial material (pretty much my average rate I think, of course a metric like this is going to be obligatorily subjective). The remaining, nevertheless, works, is alive, is something that I "get" when I listen to and want to work on. This time the song was saved.
Under such a regime of providers (but don't get me wrong, it's a wonderful problem to have, in fact I've worked my butt off for years in order to have it :) ), it's not strange that I the True North goal I've set for myself, that goal so over the top that might take several reincarnations to achieve or even never happen, is producing one song a day, from conception to publication, with studio quality.
Let alone my current goal of one song a month meeting my current technical limits, which are still modest. In this goal I'm quite off track right now, in contention mode. March didn't have its song, the song is almost there, it's going to sound great, but I'm flooded right now with "technical difficulties"; it turns out at a certain point my audio distro gets more and more laggy as you add new tracks and effects, and the song has become unstable to the point that it's an agony to work with it. So everything is recorded and half produced right there, trapped in the works of a shaky cage, a system that beeps and farts with everything I attempt, so I've been forced into a technical detour (non value added activity) to study which will be the best replacement.
I'm probing the waters, attending the "hosts" (bandits?) that my process meets along the way, and "one song a month" is perhaps becoming gradually a measurement stick rather than a goal. I really don't know right now; I haven't given it up completely, I'll still stick to trying to deliver this one in April, but it all depends on what happens in the upcoming days or weeks; I'm not going to try to ship at any cost, only to pretend I'm ready when I'm not, just to fulfil a stupid metric that was arbitrary to begin with. Let's hope the technical gods are on my side this time, and that I pick the right battles. Better lucky than good any day.
The good news with this method as I often mention is that whenever I solve a problem, it gets solved for good. Going slow to go fast. I read that whenever Toyota opens a new plant, at first the flow stops every little minutes due to the calls to solve some problem; this can last for months, but the final result is that when the flow smoothens at last, so much problems have been solved that they have the same high standards as the other more experienced plants; they took the time to think deeply and look at the root causes, didn't yield to the temptation of skipping of "fake solving" any problem.