Journal

This journal is going to be a daily account of my creative process. It's intended to be accurate, not interesting. If you decide to read it you will see how a writer and composer actually thinks and works.

In the past artists have typically avoided discussing their working methods in any detail, or concealed them, or even lied about them to create an aura of mystery; and all this because their real methods were and are, as you'll see, actually tedious and uninteresting. It's largely nose- to-the-grindstone stuff. Just working hard day after day to realize your vision.

There are a few reasons I've decided to do this, but the main one is that AI has made it necessary to take the mask off. The public now wants to know whether the art they're consuming and the influencers they're following are real people and not chatbots or AI artists, and being open about actual day-to-day labors will help to clear this up.

Another problem is that the delusional ideas so many entertain about who artists are and how they work have enabled bad actors to foment artist hate, convincing the public that artists and writers in general are "getting away with it" by having fun and easy jobs and are therefore underserving of pay, basic legal protections, or even respect. I've provided considerable value to the public but I'm completely broke, so this kind of nonsense really pisses me off. Artists are generally driven people who work hard but live a precarious existence dependent on a fickle public; the work isn't easy, and it's often not fun either. They're not getting away with anything. Once in a very rare while they do win fame and fortune, but they deserve their chance at fame and fortune at least as much as anyone else, and certainly more than the people who bash them.

So read on if you will, but don't expect fun and games. I'm not trying to entertain you here.

March 28, 2025

I'm starting this journal in the middle of a project, in two senses. First, I'm planning to remix and remaster the few songs I produced in the last couple years, and partially retrack them as well. These songs are fundamentally good, but they were just in rough mixes, as I was busy with other writing at the time; I didn't publish them beyond my Bandcamp site, and hardly advertised them. In a sense they were a trial run to see if the sound was viable. Well, it was. So, I'm going to rerelease them this year with the proper polish. They are still up on my site but I don't recommend you listen to them now. Wait until I have time to redo them as there will be a significant improvement.

I'm also in the middle of a new song titled "Emergency Call, Ontario." No, you can't hear it yet either. Probably in a month. I've already written this song and done half the orchestration. I'm currently working on the production phase for the beginning of the song, because I want to get a feel for how the final mix will sound, and because this is a test-run for fixing the earlier mixes I just mentioned. I also don't like to separate orchestration and production too much. It just becomes unnecessarily tedious. So I chip away at both as I go.

I've mostly programmed the instruments, and over the last few days I made some fixes to the mixes. My goal here is to sound listenable and poppy but still retain some liveliness from the classical and jazz realm. In other words, I don't want it to sound overcompressed. Mixing isn't my forte--in classical music it's basically not a thing--but I'm working hard to try to improve because it clearly matters a lot for popular songs and I would classify these in a hybrid genre, so some of each mentality is needed.

Today I'm tweaking the vocals. This is a very painstaking process. The vocals are synthesized, but in order to get them to sound musical and expressive it's necessary to do massive surgery on almost every syllable and often the parts of syllables as well. For instance, in the first few words, "Can you hear me," I've drawn curves for five or six different dimensions (e.g. loudness, tension, vibrato, vowel character, etc.) and also controlled the volume of each letter independently. E.g. I've tweaked the c, a, and n in the word "can" so that word sounds exactly the way I want it to.

When they hear the vocals are synthesized, today some people jump to the conclusion that "it's AI." But there's a massive difference between synthetic instruments, which have been around for decades in one form or another, and the AI-generated music that has recently appeared. I program the vocals like an extremely complicated instrument. It's similar to the way I've programmed violin parts for years, except there are more dimensions to the sound and we're more sensitive to nuances as well, so it takes an incredible amount of time. The voice synthesizer just produces a timbre and connects the syllables. It does not make creative decisions any more than the violin maker does. I control all the expression.

Now, if my own voice were working properly I might not do things this way. Basically I think about how I would sing it, sometimes actually sing it out loud, and then I program the voice synthesizer to sing in that way. It actually takes longer than just singing it. Unfortunately my own voice is broken. That's why I've decided to realize my songs by creating an imaginary pop idol.

I've also had in mind J-pop idols, who often just provided physical vocal talent and a pretty face while serious composers and producers did all the musical heavy lifting. In some ways I consider this an ideal arrangement, so why not arrange it "ideally"? The reason it makes so much sense is that we interpret the singer as an actor, and we mix up the personality of the singer with the expression of the song. In other words, songs are more similar to opera numbers than people assume. There is a bleed-over between the character in the opera and the number. The only difference is that the opera character is formally fictional and separate from the singer's real identity. The singers of popular songs are in many ways fictional too, but this is not acknowledged. By creating a wholly artificial idol, we can formalize this, and it may well be to the benefit of the music. So, while I've taken this path by necessity, I consider it an interesting artistic opportunity as well.

I may as well address when the next volume of my Memoirs is coming. Not soon. I spent much of last year trying to promote my work to the public and got nowhere. Actually it was a total waste of time because I was shadowbanned on social media. Basically, I've been blacklisted. This is, to be frank, a very big deal to me, because it undercuts my ability to survive and keep writing. Since I need to survive every day, it's something I think about on a daily basis. Now, due to my public invisibility, my current number of readers is very low, well below what my writing actually merits. That means not only no money, but no influence, no nothing. Where my fiction is concerned, we're talking about two years of hard work per volume. Two years of hard work. It takes tremendous concentration to write in this style. Have you ever done two years of hard work for nothing, not just less than minimum wage, but literally nothing? No, you haven't. Am I enthusiastic about spending two years sweating over a second volume that an even smaller fraction of people will read? Maybe a hundred people? Well, the difference between you and me is that I will probably write it anyway eventually. But, I see no need to give it high priority. For the moment I'd rather work on music. If you want more fiction sooner, then get more people to read it. Preferably a number that would allow me to pay for food while I'm writing. Yes, artists need to eat, strange I know. Don't get me started on landlords...

Okay. That's enough for today. This entry is longer than future entries should be. I'm setting a goal of writing here for fifteen minutes every day. I am not expecting many readers so I don't want to put unnecessary time into this journal. It is just going to be very plain and unedited. If you want enjoyable entertainment, go explore the rest of my site!

March 29, 2025

Saturday is the day I call family and go out for a hike. Last week I overdid it and went hiking too far over hills with a very heavy pack and wore myself out, so this week I decided to compensate with an easy walk in the forest.

When I came home I put in three and a half hours of work on the current piece, mostly programming the vocals as I described yesterday. An interesting twist in this song is that they need to alternate between insistent and cute. So for instance, in the lines "emergency call, won't you help me," which are repeated twice, I put a delicate vibrato on the second "cy" and then an increasingly forward and insistent tone on the second "won't you help me." I think this worked very well.

I'm also working on two mixing elements. This piece shifts between an acoustic bass, which I'm accustomed to using in classical music and jazz, and an electric bass guitar, which I'm not. (You can hear that switching effect on Elton's "Come Down In Time," which is probably his best-recorded track.) I spent maybe an hour experimenting with different ways to mix the electric bass under a string orchestra and eventually settled on turning it down while pumping 5k up a lot and cutting the lows from the bass reverb. I'm undecided on whether the electric bass should have reverb as without it sounds a bit more rock and roll and with it sounds more natural, and I'm not quite sure where I want this piece to sound on that spectrum. Will revisit tomorrow.

Finally I put a little more thought into the level of compression I want. Last night I compared three mixes of the 80s classic "Dress Down." The first is the original 80s mix, the second is a 2021 remaster, and the third is a smashed Youtube mix. Surprisingly I liked the smashed Youtube mix the best, even after equalizing the volume and even though it's distorted. I had a think about this and whether maybe I'd taken the wrong side in the "loudness wars." After comparing them several times I finally decided the main reason the smashed mix sounds best is that the original mix was... quite terrible, in that 80s way. There's far too much drum sound in it and it really weakens the emotional power of the song, especially since the drum part is completely uninteresting. Most probably a concession to commercial trends forced by the whims of the 80s public. The ultra-compressed Youtube mix has the side-effect of reducing the drums relative to everything else, which is an improvement.

Secondly, however, there's a lot of detail in the arrangement, and you can hear it better in the smashed mix. (When I say smashed I mean ultra-compressed). It's nice to hear all that detail. The arrangement is masterful and I do want to hear all the parts. I can't say the smashed mix is actually good, because it's distorted and the 2021 mix sounds warmer. But it does make the arrangement more audible, which is a big plus. Ultimately "Dress Down" should have been remixed from the start with quieter drums.

Anyway, evidence that less compression isn't always better (and "vintage" mixes weren't always better either). It might destroy some information, but it reveals other information and on balance that can be a win. So, with this in mind I made some further tweaks to the compression of my current piece today, trying to find a sweet spot between making everything as audible as possible and avoiding distortion. I'm particularly concerned about avoiding distortion of the cymbals as I use quite a lot of them and don't want them to sound like static.

Links to the three mixes of the song I mentioned below in case anyone is reading.

Looks like I went over my 15 minute limit. Will try harder tomorrow and set a timer to keep the entry short.

Original 80s mix of "Dress Down."

2021 mix

Smashed Youtube mix

March 30, 2025

Today I continued the same projects as yesterday and also had to deal with a slightly more interesting problem.

I'd previously orchestrated one section of the verse with simple electric guitar chords and some background string lines. This sounded fine but I wasn't totally happy with it. I felt there must be some way to make it better, and also that it was a little more rock and roll than I wanted, and not in a good way but in a boring way. So, I wanted to improve it, though I worried I'd come up with something more complicated but less effective that I'd end up having to throw out later.

I tried to imagine different ways of doing it and the first thing that occurred to me was some brass hits to replace the guitar chords. That idea was really not very good. Then I thought that I'd write a traditional orchestral horn and bassoon accompaniment part to replace the guitar. I tried to imagine this and it seemed in my head like it would work.

So, I wrote these parts and they sounded okay but now too lacking in punch. The way I ended up solving this was fairly interesting: a hybrid of the original electric guitar part and the orchestral horn-bassoon accompaniment. It sounds orchestral but with some rock-and-roll punch. I'm fairly pleased with it at the moment but to decide whether it really works I need better perspective. My worry now is that it's too busy and in the context of the whole piece it will give too much busyness and emphasis to a section that wasn't meant to have it. I may not be able to decide that for certain until I've finished the rest of it. I'd say there's a 40% chance I'll come back to it later and revert to the original, simpler accompaniment. But, that's the way it goes sometimes. You gamble that a complex realization is going to be effective but you don't always know. You have to work it out before you can see for sure. Well, I do, at least.

After that there was another problem to deal with. The voice I'm working with tends to get too aggressive in loud sections. This is not really a problem with a male voice but to my ear it's pretty unpleasant with a female one. Think Heart. I don't like Heart at all. So, I need to redo the louder sections of the vocals in some way that doesn't sound aggressive. Possibly by tweaking the volume to be louder instead of making the voice "naturally" sing louder. Or possibly there's some other way I can get volume without an aggressive tone. I need to work on this more tomorrow.

All in all it was a rather tedious day due to the amount of vocal programming I had to do and the horn writing was a bit frustrating as well because my horns have gotten dynamically miscalibrated to my strings (I haven't used them in a long while as I've been focused on string-only music for the last couple years), which forced me to do a lot of fiddling that shouldn't have been necessary. If I write something more truly orchestral in the future I need to spend a day recalibrating all the instruments again. But, for the moment I'm pleased with the new arrangement. I hope I stay pleased tomorrow.

I ended up cutting the composing day short as I had an idea for a mini-story to be included in my "Ailom" essay. It's called "Pop-Op." I wrote it in a couple hours; it's nothing fancy but it communicates a new idea effectively. I immediately posted it on social media, and... it's not a hit. At all. Unsurprisingly. Well, I still think it's worthwhile.

Anyway, I really want to spend April entirely focused on music with no more distractions from extraneous ideas, so I need to stop having them whether they're good or not.

Ok, my fifteen minutes are up and today I'm staying on schedule.