[r-t] Compositions of the Decade: Part 1 - Preamble

[Richard Smith]

Philip Earis wrote:

> a new decade will soon be starting.

Well, in 13 months it will :-)

> Given the vast, vast search spaces, though, computer power 
> is usually just a tool that needs a clever mind to produce 
> a great result. Intelligent design, one could say, is what 
> differentiates composer from monkey.

I whole-heartedly agree with this.  Unfortunately, tools 
like BYROC have given computer-assisted composing a bad 
name.  Any monkey can crank BYROC's handle and wait until it 
excretes a composition.  (That isn't to say it can't produce 
decent compositions too, if used in a thoughtful manner.) 
As a result, it seems to me that composers are quite cagey 
about admitting to having used a computer, other than to 
prove it afterwards.

> It is also of concern that many of the compositions I will 
> include are hard to find, and in quite a few cases do not 
> appear on the web, even on a fleeting medium like a 
> personal website.

I agree with this too, although on this point I'm as guilty 
as anyone.  I'm sure some of this seeming secrecy is 
intentional -- if you've produced a composition that you 
feel is particularly special, it's nice to have a say in who 
rings it first.  But I'm equally sure a lot of the time it's 
just down to bad organisation, or failure to appreciate that 
it's worthy of publication.

> It is hoped that efforts at producing a stable, central 
> online repository for compositions will yield tangible 
> results soon.

I would love to see a single central compositions database, 
and this was something that GACJ and others were discussing 
back in the spring.  It seems to have fizzled out, not, I 
suspect, due to lack of enthusiasm, but more due to everyone 
involved having lots of other commitments on their time, and 
a number of difficult issues that need resolving (and, 
importantly, resolving in the right way).  I'll see if I can 
find time to return to it this winter.

> Coming tomorrow - Part 2: Doubles

I shall particularly look forward to this part.  I have 
tended to neglect doubles thinking that everything 
interesting has already been discovered.  But given that we 
don't even know to within a few orders of magnitude how many 
extents there are, that's something of a sweeping statement. 
And, of course, you can do further things using multi-extent 
blocks.

RAS