Re: [OT] Vivi, the Virtual Violinist, plays LilyPond music

[Graham Percival]

On 3/18/11, Kieren MacMillan <address@hidden> wrote:
> Graham,
>
>> Has that happened with books?  Have stories become
>> total crap over the past 10/50/200 years?
>
> Actually, yes: no author made a million dollars writing a Harlequin Romance
> novel in the 1500s.  :)

Hmm.  I'll admit that penny dreadfuls were in the 1800s, not 1500s...
but I'm certain that the 1500s still had raunchy, "low-class"
theatrical plays and songs.  I don't believe that everybody sat around
in their castles writing Nobel-quality poetry to each other.

> What I *am* saying is that just
> because my neighbour can now "write and perform a symphony" (quoted for a
> reason) in his garage does not make it "good music".

Of course not!  But regardless of quality, it *is* music.  It's a
human being active, instead of watching American Idol.  It's a human
being creative.

I don't care about the profession of music, be it performers or
composers.  I don't care about Music, with a capital 'M', being the
history and academic study of "good" music.  I consider jazz music to
be the most important musical invention in the 20th century; far
outweighing 12-tone music, Cage, minimalism, or any innovation in
"academic" music.  (the second-most important would be rock/pop music,
even though I don't know what the difference between the two -- and
note that I don't even like Jazz music, and can't stand most rock and
pop music)

I care about human creativity.

A bunch of teenagers in a grungy basement in Seattle in the 1980s
writing songs about how emo they are, using nothing but power chords,
is more creative than somebody sitting in their living room listening
to a CD or Mozart string quartets.  A middle-aged housewife writing
homoerotic star trek fan fiction is more creative than somebody
listening to a CD of Debussy piano music.

I'm not saying that we need to be creative all the time -- sometimes
it's good to relax, and of course it's good to listen/read/view a lot
of art to get ideas to use in your own works.  But I think that
creating new art (of any quality) is more creative than looking at
existing works.

Classical music is no guarantee of high art.  I used to play cello in
quartets for weddings and dinner banquets.  When we played Pachelbel's
Canon, I spent most of my time glancing at the neck-lines of women's
dresses.  Ditto for Mozart divertimento 136.  They're both great
crowd-favourites, they both have easy cello parts (I memorized them
without trying to), and they require virtually no creativity from the
cello player.  At least, not for the venue of "providing background
music while people mingle and drink wine".


> I *do* think so -- and recent studies on youth support my belief with
> evidence. On the music side, consider the fact that recent studies have
> shown a majority of young people prefer the sound of compressed audio (e.g.,
> low- to medium-bitrate MP3s) to uncompressed audio. [Pause here to fully
> appreciate the horror of that statement.]

What am I supposed to be horrified by?

Listening to music produces a subjective feeling in humans.  Suppose I
receive the most aural pleasure by listening to Shostakovich music,
passed through a low-pass filter at 50 Hz.  (for non-engineers: this
means I can hear some muffled "boom" noises, and no chance at melody
or anything like that).  So what?

Tastes change, trends change.  Am I supposed to be horrified by the
clothing fashion in the 1960s and 1970s?  They look ridiculous now,
but (presumably) back then people thought they were trendy.

Maybe 30 years from now, "real audio" (i.e. not compressed, not lossy)
recordings will be all the rage.  Maybe not.  I don't see either one
as a problem.

> A lower barrier of entry by definition allows people to "get into the field"
> with less experience, less training, less discipline, less persistence, and
> so on. Are there some benefits to this? Sure. Does it increase the amount of
> crap we have to wade through. Absolutely.

Of course!  That's why reviewers -- be they humans, or computer
recommendation systems (which is a big area of research) -- are
becoming more important.  The most famous computer recommendation
system is google, of course.  Given 1234 trillion websites (or
whatever), you ask it "ubuntu pulseaudio not working", and it
recommends a list of 10 websites it thinks you want to see.  It's not
perfect, of course... but given the number of websites out there, and
how certain people try to 'game' the ranking... I think that google is
pretty fantastic at this particular recommendation task.

Other people are working on music recommendation.  If you like music
A, B, and C, then which tracks out of all 297,814 tracks on Jamendo
(free and legal downloads) will appeal to you?  In the grand scheme of
things, 300,000 pieces of music is only a drop in the bucket of all
music recordings... but it's a useful place to work on such
recommendation systems.  Other people do this with youtube, doing the
machine learning on audio and video signals.
(some of these systems work on tags, but that's a weak form of the
problem -- it relies on humans tagging the music.  The "strong" form
of the problem considers purely the audio signals, and makes
recommendations based on that alone)

Cheers,
- Graham