[cc-community] 10k great Free songs (was Re: Rcrd Lbl to offer CC and Free as in Beer Music)

drew Roberts zotz at 100jamz.com
Tue Nov 20 08:14:52 EST 2007


Greg,

fine, so it is different. And so we need experiments. Searching for solutions. 
So it is different, but I think you may be overstating some of your points 
below.

No real extensive data here, but from experience, I think we may be able to 
get a "good enough" solution to the music problem.

Point one: back when Napster was on top of things, one of the best ways to 
find new stuff you might like wsa to look at what people had that you did not 
where those people had a goodly overlap with you. Copyright issues killed 
this solution.

iRate radio has done a good/passable job finding me music I liked for me in 
the past if I ignored the license issues. But it does not work for me for 
what I want as it does not allow me to limit its suggestions to only tracks 
with specific licenses.

I think iRate could use improvement along these lines:

allow license preference settings.

have modes, one for listening, one for listening and downloading the best 
guesses, one for rating.

they could improve the play system by making it partially seperate from the 
rating system. one idea i have had is to build in a mood rating system - 
there are times where i really do not want to hear certain songs that i like 
very much.

overall adding metadata to files: genre, mood, content rating?, license 
meta-type, license, more?

I also think I read something recently, perhaps a post about improving the 
recommendation system as it relates to the netflix contest is it? memory of 
it is too dim... that might help in this area.

I think we do not have to get anywhere near an optimal solution to make one 
good enough that people will be happy with it.

This whole discussion does point up something i have been making a point of 
recently... the opinions and recommendations of trend setters, people of good 
taste (widely shared tastes actually), people willing to do the searching are 
valuable (economically) with the situation we have today. Value that often 
goes unrewarded. (economically)

So, to try for some conclusion (due to time constraints) I agree that the 
problem re code and art is / could very well be different. I think we can get 
a workable solution none the less. I still think we are not really going to 
get near that solution for the likes of me until the pool of works under the 
Free licenses is larger.

Make or help make, or fund the making of, music you like and put them under 
Free licenses if you care to have such music...



On Tuesday 20 November 2007 06:22:04 Greg London wrote:
> OK, so *this* is funny.
>
> >> That doesn't scale well at all.
> >
> > But it *has* worked well for me.
>
> I'm talking about group dynamics and
> you focus on you. I might as well have
> said that most people lose when they
> play the lottery, only to have you inform
> me that you won the jackpot once.
>
> Great. I'm glad for you. It could very
> well be that you've got a number of
> sources of music to the point that you're
> satisfied with your arrangement. Does that
> change the group dynamics at all?
>
> Did you hear the other people on the list
> who mentioned their issue with slugging
> through 10,000 songs and most of it was
> stuff they didn't like? I spend a couple
> months going through Magnatune and come up
> with maybe a dozen or so artists that I like?
>
> Do you get that *you* winning the lottery
> doesn't prove that *everyone* should win
> the lottery?
>
> All I've been saying is that music doesn't
> work the same way as software. That doesn't
> mean it won't work at all. But telling me
> it's working for you isn't telling anything
> at all about whether or not it's working
> the *same* or not.
>
> There are issues in finding the "good" music
> that just isn't a problem with software.
> You only need to find one good compiler, one good
> editor, one good word processor, one good spreadsheet.
> Forking is generally discouraged. If you have a
> modification, the developers generally prefer to
> roll it into what they have, rather than forking it
> off into two different applications.
>
> Music is all about variety. One good song doesn't
> do much for your playlist. People need to find
> gigabytes worth of music. And once they've listened
> to that playlist a few times, it'll get old and they'll
> want more, different.
>
> The iPod shuffle can hold about a hundred songs.
> I own one. It drives me nuts. It was too much work
> trying to keep circulating the music on my ishuffle
> from the library of music I had on my PC.
> I stopped using it. How long do you think 100 "good"
> songs will last if that's all people are playing on
> their iPod? A couple months? Maybe?
>
> Little factoid for you: The number one reason that
> most people buy a novel? Because it was written
> by an author they read before and they liked the
> previous novel. This was something I heard from a
> book publisher about a year ago. In the era of
> the internet and the possibility of being able to
> order any book you want and all the reviews that could
> be available, and everything else, it still comes
> down to the problem of the search. Of finding material
> you like. And according to this publisher, the number
> one reason people buy a book is because they already
> knew they liked something else the author wrote.
>
> That to me says the issue of the search, of finding
> what's good, of separating the wheat from the chaff,
> is, on a group level, not prone to technical solutions.
>
> I realize it isn't quite the same as music.
> There are larger investments of time in reading
> a book than in listening to music. But in both
> cases it comes down to a subjective experience,
> of finding the stuff you like, of finding the "good".
> And that isn't a problem that is solvable by any
> whizbang gizmo gadget just yet.
>
> Even with all the amazon stuff about "people who
> bought this book also bought these other books".
> Even with all the online reviews you want.
> Even with all the different people trying to
> offer help and point people in the direction of
> books they might like. Even with all that, the sales
> figures tell a simple truth. *MOST* people buy authors
> they've already read. None of the current existing
> technical solutions have changed the way most people
> choose what books they buy.
>
> Now, if you come back and tell me you always try
> to buy some author you've never heard of, maybe
> some author that some third party recommended,
> then once again you are missing the point.
> You can't change the sales data. It is a fact.
> It comes straight from a publisher. And it was
> something they said only a year ago. For *most* people,
> all the existing methods to try and help or sway them
> to buy one book or another does *not* change their
> buying habits. They buy authors they already know.
>
> I don't have sales data for music. But looking at
> the issues, it seems to me that music is a lot
> closer to novels than it is to software. THe problem
> is sifting through and finding the "good". THis
> goes for artists who are looking for existing stuff
> to derive from, and it goes for listeners who simply
> want to fill up their iPod with a few gigabytes.
>
> I think folks are probably more open to gamble
> three minutes of their life listening to something
> they never heard of before. But again, I point out
> the people on this list who posted their experience
> with slugging through thousands of songs all completely
> new to them. Radio stations might help, but it doesn't
> change the fundamental problem. And I think that the
> sales data for novels show that at some point, you
> just can't solve some problems with third party
> aggregators, taste masters, play lists, and stuff like
> that.
>
> And maybe it can get to the point that it is
> "good enough". But that wasn't my point, was it?
> I keep saying that it isn't neccessarily insolvable.
> I keep saying that it's a different problem than
> free software presented.
>
> Greg
>

all the best,

drew




More information about the cc-community mailing list