12 Years Old. Twelve. There is hope for humanity.

[cola2410]

For me, an effective method to judge "epigonal" pieces relies in comparing them to the "originals". (And sometimes, it's the shortfalls of epigonal pieces that improve my respect for the "originals")
There is a huge existing knowledge about musical hermeneutics. These tools can help you to analyse the core of the inner network of a specific piece of music in order to approach the specific "substance". To put it in an overly simplified way: relevant works always use models to a certain extent. But at a specific point, the model is distorted/modified. What follows, is a stringent justification for the inconsequence. And the justification carries the "message".
You find this pattern in every substantial piece of Bach, Mozart, Beethoven, etc, etc, etc.
I don't hear any of this in AD's exercises.
I hear this pattern sometimes in music that doesn't represent my personal preferences - e.g. Rap music; hip hop etc etc.
So i guess, while it's impossible to make clear distinctions, there are parameters one can approach by listening carefullly...

This is something I'm able to carefully concur with even if at the end it sounds offensive to someone. As long as I'm involved in AI development it seems like neural networks used for prediction and pattern recognition work the same way as our "still human" brain (and some of them modeled directly after neocortex) and I'm utterly upset about what we may face in a very near future. Living Fossil, you're absolutely right about the pattern and some expected model mutation. So how to force a mutation - introduce some distortive force and in AI it's usually represented by some external artifically created noise that is a perfect example for the social/musical/political/etc environment. The distortion is essential to the quality of the model for the sake of prediction capabilities because following the same pattern eventually leads to the model degradation and lack of aiblities to react to slight changes in the environment. Actually among the others there is another method of improving the predictions called "ensembles" by combining different model types together even if alone they don't produce good results. Just think about how it again corresponds to human collaboration.

Mutation, or "message" is good, following or layering over the same pattern again and again does not improve its quality and leads to conversion of the "brain" to a sort of finite-state machine what makes it amazingly vulnerable to unexpected influences or even slight change in the input data.

And yes, it happens in any music style.

 

[mikeh-375]

I don't.



My issue with the deeper understanding that you refer to is that I don't agree that it really matters in creating "meaning" or something great and transcendent of time.

I fail to see any of "musical semantics" and whatever else in something like Beethoven's 5th or Fur Elise. The latter could've been about anything else for all I know. It could've been titled differently, and it would've been just as great.

I can't remember which piece it was, one that was supposed to be about the moon, but a friend of mine told me that he was listening to it while driving and his daughter was in her car seat and she said "It sounds like the moon, daddy." So, obviously, we have a composer who is capable of conveying these semantics to even children. The idea that this inherently elevates his work to some higher value is absurd; it's just snobbery.

Much like any other form of symbolism, like the color red and what it "means" are simply tools you can use to convey a message. But they are largely cultural, and do not necessarily have universally understood meanings, and such meanings are also dependent on the individual.

I could sit here and probably derive some perceived "deeper meaning" in this girls music, and as long as I use enough persuasive, florid language in my argument, I could prove nearly any point I want to make. "See! There is a meaning! You plebs just don't notice it!" The idea that her music, or any music, must have this sort of symbolic relevance and "say something" to be "art" is little more than intellectual masturbation.

I don't even mean to insult anybody with that, but that's just the way I honestly feel about it.

Chris...Glad to hear it..
We might be at crossed purposes here. I mostly listen to music without much regard for any semantics suggested in titles and agree that they are often not necessary. Neither am I (in my posts) suggesting that any literary reference or anything else extra-musical that is associated with a work, elevates it beyond the norm - in fact generally speaking, I agree with you on the point you make about Beethoven's 5th for example and cultural interpretation, although there are many exceptions where an understanding of the composer's intent does enhance the listening. There is a case for necessary elitism in musical taste as you learn more about the craft, but that is a different debate, although one relevant to AD.
What I am suggesting is that in music that aspires to a higher level of attainment, be that personal or communal ( in the concert hall), the profundity required needs a corresponding depth of knowledge and mastery of the craft because the goal will be impossible to reach without those foundational skills and the discovering of one's inner voice by attaining those skills.
In AD's case, I believe relevance is an issue, because she sounds like pastiche at present, one that any competent undergrad composer can also emulate and in a style for which the final word has been emphatically written. She is more than that and in time I hope she finds an original way.
I ask you, how many kids today dance to Bach's Matthew Passion? Not many (it is hard to twerk to I've found..try it). Relevance affects every aspect of cultural creativity and is an essential engine to keep art alive. This is not snobbery or intellectual wanking, it's survival and echoing Cola's post above, we need extraordinarily gifted people like AD to distort the model, not revert back to a time that will stagnate us..."darling have you seen my ear trumpet"...

 

[Chr!s]

There is a case for necessary elitism in musical taste as you learn more about the craft, but that is a different debate, although one relevant to AD.

I don't disagree, but I again say this elitism in this thread is being applied to a 12-year-old writing this kind of music at a time where even 30-year-olds prefer cracked FL studio and premade beats and loops

This is not snobbery or intellectual wanking, it's survival.

I question if this — the survival aspect — really is the case. I've heard people say this for as long as I can remember: "The band changed their style to a more contemporary one to stay relevant" but then that band's newer stuff never sells as well, no one goes to their concerts to see it, and the band basically becomes a cover version of themselves. How many Skid Row songs can anyone remember after Slave to The Grind? The albums afterward are a lot more contemporary, but no one cares.

Fact is, most music will be forgotten. But stuff like Vivaldi's Four Seasons, The Raiders March, or Stairway To Heaven have, and will endure. JJayBerthume on Youtube has tons of followers and most of his music most would say sounds "dated". Yet, he's been scoring video games, teaching, and writing concert pieces since he was a teenager. I don't think he or his music is irrelevant.

Is dubstep still a thing anymore? I don't hear it on the radio at all anymore, nor does anyone who used to make it seem to do so anymore. I haven't heard or seen anything from Skrillex in years now. Is he still really relevant?

Because of the fact that certain styles are still imitated, and many pieces from that era have endured moreso than other trends that have come and gone through the ages, I can't really get behind the idea that following currently relevant practices is much of a means to survive.

In my opinion, finding your own sound within previously established paradigms that are accepted as a high standard and that you like is the best course of action, and it's what I'm trying to do myself. I'd rather be like Van Halen than Korn, and I'd rather be like Vivaldi than trailer music. Even if the latter are more "relevant" to young people today.

 

[Ned Bouhalassa]

Somehow, the title of this thread makes me think of scotch.

 

[Gerbil]

The content makes me think of insanity.

 

[mikeh-375]

I don't disagree, but I again say this elitism in this thread is being applied to a 12-year-old writing this kind of music at a time where even 30-year-olds prefer cracked FL studio and premade beats and loops

More crossed purpose here Chris. I was talking about a specific genre of composition and performance. I can't disagree with the above of course, but it is not (at least at present) relevant to AD given her current focus.

I'm sure you've figured that what seems to endure in art, more often than not is (among other traits) innovation, quality or a piece that succinctly sums up a generation by culling the groping around by lesser peers in their search for a zeitgeist/voice. What our proposed genius then does thanks to choices made by their brilliance and instinct is create a sort of meta work whose innovation seems, with hindsight, obvious. These works/songs/soundtracks et al often suggest new ways and become templates for the next generation who then stamp on their own relevance/personalities and so it continues. We are surely hard wired to progress and find new fields to explore and this trait has to be intimately linked with survival - in life and art.

In my opinion, finding your own sound within previously established paradigms that are accepted as a high standard and that you like is the best course of action, and it's what I'm trying to do myself. I'd rather be like Van Halen than Korn, and I'd rather be like Vivaldi than trailer music. Even if the latter are more "relevant" to young people today.

Now this is something I absolutely agree with but would add the proviso that you also need a sense of adventure in order to stand out a bit more from the noise of the gropers....innovation to survive can mean earning money too...right?

 

[mikeh-375]

The content makes me think of insanity.

Why?
Nowt wrong with a decent discussion now and again.

 

[Living Fossil]

I fail to see any of "musical semantics" and whatever else in something like Beethoven's 5th...

On the one side you admit that you have no idea about musical semantics, on the other side you still think that you have the competence to state that you don't see any of "musical semantics" and whatever else in something like Beethoven's 5th...(so suggesting that there is none).
No offense, but it is because of your lack of understanding that you fail to see it. And if you had some understanding you could clearly see that it is full of empirically determinable semantic chiffres, models etc. etc.
It's really annoying, when people confuse ignorance with competence.
"I don't care about science, but i can clearly see that the earth is flat, so we live on a flat earth".

 

[Gerbil]

Why?
Nowt wrong with a decent discussion now and again.

It's certainly been entertaining and....enlightening

 

[mikeh-375]

It's certainly been entertaining and....enlightening

Hopefully in a good way Gerbil.......

 

[cola2410]

I would add a little more.
Some people get scared or angry if someone questions their ability to create but we have a serious threat now because humans were never considered as the super-efficient creatures who are able to invent. For instance, our data transfer channels are very slow and the dominant one is vision for most and hearing for some, other channels are ridiculously slow and non-informative enough even compared to animals. And of course we typically spend years to train our brains using these very limited channels and repeat the same several times for our children not having enough time to develop ourselves.
On the other side, AI has virtually unlimited bandwidth for specifically prepared input data and unlimited power when enough electricity is provided (I'm not talking about CPU/GPU efficiency itself because you can achieve mostly the same by building a monstrous machine than wait for a new-generation CPU to come). With tensor chips recently made available for public anyone can experiment and question himself about whether we are unique enough to differentiate ourselves from machines.

https://www.chess.com/news/view/updated-alphazero-crushes-stockfish-in-new-1-000-game-match
"In several games, AlphaZero sacrificed pieces for long-term strategic advantage, suggesting that it has a more fluid, context-dependent positional evaluation than the rule-based evaluations used by previous chess programs"
AI can beat any human in virtually any type of the game if the rules are provided without any supervision. Upload classical music theory in it along with several hundred operas and just wait for a couple of days for another AD?

https://kotaku.com/the-best-moments-from-a-twitch-conversation-between-two-1790907744
AI is already able to entertain itself but the most scary part is that from the conversation it may be indistinguishable from a human.

 

[Jimmy Hellfire]

But that's all irrelevant to me. Computers will be better at computing, go figure. Being human to me was always about entirely different qualities, and those are things an AI will never be capable of. The rest is scientism.

 

[ghostnote]

...and those are things an AI will never be capable of.

Based on my observation of those highly irrational beeings you all call Humans, there's nothing much left that can surprise me anymore: There will be art made by AI and Robots - it already has been - and it will sell ...well big. Just take a look what Humans consider as art:



Speaking ontopic: The world is not black and white. Is this girl good at what she's doing? Yes, as many have been before. Is her work highly derivative? Yes, but that's not important at this point. But will this save serious music? That's the big question here. I doubt that. I've seen 200+ classical concerts in my rather young life and I've noticed the decline of people who are interested month by month. 80% are over 60 years old. Unless there are names attached to it like David Garret or Daniil Trifonov, there is not much interest. Youth orchestras, operas for kids, etc. probably won't change this trend.

Jimmy Hellfire seems to have an aversion against those people, probably thinking all of them are rich snobs, opportunists, lobbying and he's partly right, but as much as ignorant those people might appear so is the uneducated mass, the other side of the spectrum. With the democratization of things there is less and less respect towards other people, on both sides, which is increasing dynamically. People will always fight something they don't understand, simply because understanding takes effort. For example, a school friend of mine got bullied because he played the trombone. There is something in the air, an attitude that it's not worth to understand certain things or getting better at something, develop taste and character and it's increasing with the influence of the internet.

 

[cola2410]

Being human to me was always about entirely different qualities, and those are things an AI will never be capable of. The rest is scientism.

I would seriously question human qualities and its profound differences from machines in, say, a decade from now. One of the Berkley professors once said - the day machines start mimicking our behaviour and emotional expression trying to learn from us would be the first day of the world's end as we know it. Craft covered, next is emotion with empathy then humans finished.
Actually we are no more than a combination of 100 billion neurons, passing signals to each other via as many as 1,000 trillion synaptic connections, equivalent by some estimates to a computer with a 1 trillion bit per second processor with brain’s memory capacity vary wildly from 1 to 1,000 terabytes and all this is encapsulated in a self-sustained biological machine to produce power for it.

And according to

There is something in the air, an attitude that it's not worth to understand certain things or getting better at something, develop taste and character and it's increasing with the influence of the internet.

humans don't actually use or require all this power to communicate with each other so machines can easily exist in the human society without any trouble, at least virtually. Personally I'm far from thinking of us being unique or irreplaceable. I don't think so. At the same time our body fails when we desperately need it to function - when we finally get smarter with age and ready to break the boundaries of the past and present. Don't you think it's unfair?

 

[Jimmy Hellfire]

I would seriously question human qualities and its profound differences from machines in, say, a decade from now. One of the Berkley professors once said - the day machines start mimicking our behaviour and emotional expression trying to learn from us would be the first day of the world's end as we know it. Craft covered, next is emotion with empathy then humans finished.
Actually we are no more than a combination of 100 billion neurons, passing signals to each other via as many as 1,000 trillion synaptic connections, equivalent by some estimates to a computer with a 1 trillion bit per second processor with brain’s memory capacity vary wildly from 1 to 1,000 terabytes and all this is encapsulated in a self-sustained biological machine to produce power for it.

Well yeah, that's the scientism way of looking at things, which honestly, is just a loss of perspective when you break it down. It's also a form of cultural pessimism, but it doesn't make it true.

It's a sign of the times, since modern life has developed from industrialisation over mass reproduction all the way to the age of the model and the algorithm. We live in times where people have come to believe that faking it actually does equal making it, and that if it looks, sounds, smells and seems real, there is no difference to that which is real. A diminished sensory apparatus is the reason for this lack of perspective. Humans today are very insecure of what constitutes their being, and these kinds of views are a testament to that.

 

[mikeh-375]

There is something in the air, an attitude that it's not worth to understand certain things or getting better at something, develop taste and character and it's increasing with the influence of the internet.

I also see this trend in relation to music for sure but also in other disciplines. Cola's AI might entrench that attitude once it passes a 'Musical Masterwork in the guise of a Turing test' and even before then, given the apps that seem to come out regularly making things easy for those who don't study. But then again art is personal, not binary and there will always be a human spirit that needs to express itself and the best expressions in concert art music, the enduring ones, are written with deep understanding of the craft.
As an aside I am still amazed that I feel I have to defend study and knowledge in composition, I mean how does a pianist play concertos without the fundamental security of practice, scales, arpeggios? How do you write to the best of your ability without knowing about what options you have? Why is it different? it ain't. Snobbish? Elitist?..... bollocks. If so, my plumber is such a ponce because he can weld pipe together and I can't...sorry guys..mini old git rant...

 

[cola2410]

I'm sorry )

We live in times where people have come to believe that faking it actually does equal making it, and that if it looks, sounds, smells and seems real, there is no difference to that which is real.

Actually it's no different from when you smile at an infant and a baby smiles back at you without knowing yet it means an expression of a certain emotion. First they imitate then start copy the parent's behavour and so on. Meanwhile, the big question here is the definition of real. At certain point AI will have its own definition of real for sure simply because its senses could be more advanced and more efficiently used to enrich and develop itself. Simply speaking they also may like music they create and of course their music could be absolutely non-musical for us, for example, if it exists in ultrasound range. This will be a different race eventually.

As an aside I am still amazed that I feel I have to defend study and knowledge in composition, I mean how does a pianist play concertos without the fundamental security of practice, scales, arpeggios? How do you write to the best of your ability without knowing about what options you have? Why is it different? it ain't.

I may probably answer tat. If we look at music history actually theory almost always comes second. First we get an extremely gifted individual who does something nobody did before then some theory supports it. Can it be reproduced in AI? Yes. Developing a pattern sometimes requires breaking it and starting from scratch or from some point if the pattern is solidified and resistent to noise. In such scenario starting from the beginning may use some initial or side information (what we call subconscious in our world) and requires very focused extreme speed learning process. Actually it's no different from a gifted child with extreme brain speeds and supersensitivity who is raised in isolation from any disturbance. So we have a pattern that may achieve the same result much faster than the previous one and even do something more but at the same time it's limited and very unstable to external influence because of very little information used during the development. It's like a shortcut.
Yes we value hard work and spending 10,000 hours to become good and noticeable at something but honestly, it doesn't make much sense feeling myself elitist only because of that. I know there is someone who is able to do the same in much less time and I can't do anything about.

 

[mikeh-375]

Not sure we completely tally there Cola.
I'm pretty sure nobody writes worthwhile concert/classical music without at least a basic understanding first, even genius - I mean, you can't score without some idea of technicalities. Crucially, I wonder if you are missing what theory means to a composer - what it actually does. The point about theory is that when studied properly, it allows you to find yourself, your voice by a process of discovery, discarding techniques deemed uninteresting and pursuing techniques that give you a twitch in your guts. You discover your own instincts as you learn, it's a two way street. When mastered, theory becomes instinctive and is your guide as you explore. It is not about being elitist at all, it's about your personal journey and how you want to travel.
I don't doubt the AI analogue (wrong term sorry) to this process, but there is a human need to be as good as we can, and in that striving is where the art is I believe.

 

[cola2410]

Not sure we completely tally there Cola.
Nobody writes worthwhile concert/classical music without at least a basic understanding first, even genius. Crucially, I wonder if you are missing what theory means to a composer - what it actually does. The point about theory is that when studied properly, it allows you to find yourself, your voice by a process of discovery, discarding techniques deemed uninteresting and pursuing techniques that give you a twitch in your guts. You discover your own instincts as you learn, it's a two way street. When mastered, theory becomes instinctive and is your guide as you explore. It is not about being elitist at all, it's about your personal journey and how you want to travel.

I think I understand what you mean and it's a personal perspective. Yes basic understanding and using relevant tools helps a lot but for some no help required. Even in my school there were kids who started playing piano or singing with right intonation by 4. Of course everybody starts from playing anything consonant or pleasantly sounding but there were few exceptions who liked pink panther more than jingle bells at a very early age. Most of us follow the same way simply because we can't go anywhere else without a guide. But... there are exceptions. Yes theory provides support and eventually fuels their amazing vehicles but they can get there without any support, but slower.
Lets take a different example - I knew a couple of students at my university who actually solved problems in their minds without any written explanation and there was one who easily multiplied big numbers and said he just saw the result in his mind.
Personal experience is subjective and achievement is objective. Yes every human is unique because of unique experiences but who eventually cares if he can't show it or express in some way for later use by next generation. As one of my teachers said - there are "bright" people. They learn fast, they find solutions and ways, they decrypt things and recover amazingly fast. Their abilities may lead them to nowhere but at least we have them (AD). Let them first improve themselves and ourselves after because if we ever compete with AI we are outdone. But they are not.

 

[Chr!s]

On the one side you admit that you have no idea about musical semantics, on the other side you still think that you have the competence to state that you don't see any of "musical semantics" and whatever else in something like Beethoven's 5th...(so suggesting that there is none).
No offense, but it is because of your lack of understanding that you fail to see it. And if you had some understanding you could clearly see that it is full of empirically determinable semantic chiffres, models etc. etc.
It's really annoying, when people confuse ignorance with competence.
"I don't care about science, but i can clearly see that the earth is flat, so we live on a flat earth".

Why don't you go ahead and enlighten us all to what semantics and "meaning" there are in the 5th Symphony instead of slinging mud from your high horse?