Hacker News new | past | comments | ask | show | jobs | submit login
Hacking Soundcloud: Creating an Interactive Track (haywirez.com)
65 points by haywirez on May 31, 2017 | hide | past | favorite | 17 comments

Cool idea -- other people already do something similar though with WIP, if a rudimentary version of a track gets interest, they come up with a better arrangement, mix, mastering etc and just upload the revised tracks as replacements.

That said, if I was to do something "interactive" I'd probably set up some kind of site where producers could audition hooks and then make the most upvoted ones into an actual track. Or even further, let the users build hooks that then get arranged into a full track (by a producer)...

The base idea of updating an interactive track is good. But changing distortion according to view count seems bad.

Maybe try with integrating comments that people make over sound (on timecode) to create a kind of "discussion" between users (use synthetizer to recreate human voice from text of course) ? The same could apply with the video stream but that would be more complicated and costy

Great idea with the speech synthesis! Of course there are many more creative uses to explore. The tracking engine I made for this[0] is flexible, it can essentially track and use every measurable interaction across all connected networks and assign different weights to them. For example, I can also connect the weather, and if it's a nice day in my location, every metric can count for more.

[0] https://songsling.io

Subscribed ! There's a lot of fun things we could do

So did you create many distorted tracks ahead of time or are you automatically adjusting the distortion down? If the latter I wish you talked about how you did that but maybe I missed that?

Automatically making music seeks a lot tougher.

I created 10 distorted versions. But I'm also working on another follow-up track which will have over 200 auto-generated versions :D

A small nitpick, but I'd like to point out that you don't need that () around `text` in your code:

  .then( (text) => {
changes to

  .then( text => {
Not only is it much cleaner that way, but for single-argument arrow functions, it saves you the trouble of typing the parentheses.

Programmers are lazy. :)

You don't need braces around single-statement if/else/for blocks either, but you put them anyway because you're a good boy. Same could be argued for single-argument arrow functions. My point being, this is a debatable point.

And here I thought that was language specific like how Python from what I experienced did not use semi-colons to end statements.

What about an if statement checking if something is true, is it enough to just say:

if (value) { // true }

I have seen some shortened code that I did not write and wasn't sure it was valid but it works.

Reminds me of lambdas in Java 8. It's technically allowed to omit parenthesis for single-parameter lambdas (and tools like SonarQube encourage it), but I never do that, for the sake of consistency and readability.

P.S.: If enough artists put a "price" on social metrics within their works, it would generate market dynamics between the platforms. This would force them to behave in a way that is beneficial to the creators.

Are you concerned with "fake"/bot generated (pay for metrics) social metrics?

Also, what are your thoughts about the use of blockchain technology to mediate the price and payment for artists' creations?

Some Services Exploring the Concept:

- http://myceliaformusic.org/

- https://ujomusic.com/

Well, you as an artist using bots would be shooting yourself in the foot. If others do it for you (competition?), they're wasting money and also making you look better in comparison...

I've been closely following both Mycelia and UJO. Also helped out bootstrapping Resonate[0] a bit. I think most of these initiatives have the same problem; they are putting a horse in front of a car - using new technology to address yesterday's model.

The reality of today's content consumption is that there are other useful, unaccounted tokens of exchange alongside money. Going forward most people will simply not pay for recordings[1], and especially not in meaningful amounts for niche content (the driver of human culture). But they can contribute a lot by spreading the word or otherwise caring. Building audience size and "metric capital" is more important than the small number of direct sales we can generate otherwise.

I think free content and patronage with money or word-of-mouth is a better model. It is working really well for podcasts, blogs and similar. It is not typical for music as we are still stuck in the cultural anomaly of the 20th century.

My idea was to try and reintroduce a gentle value connection between the maker and the audience. Something resembling a gift economy[3], not an exact accounting of debt. There are lots of good ideas to be found via anthropology.

[0] https://resonate.is

[1] Licensing is a separate issue, a much better use case for blockchains.

[2] http://thayernick.tumblr.com/post/80563891112/do-you-wanna-k... (note this is 2014, before streaming)

[3] https://en.wikipedia.org/wiki/Debt:_The_First_5000_Years

Thanks for the reply.

1 - For patronage, how do you view the Patreon model?

2 - A gift economy seems good, but how do you eventually "gift up" to the necessities (food, water, shelter)?

Patreon is great, the only thing is that we shouldn't need a middleman that takes a cut. But insofar as teaching people a new behavior, it's wonderful. It will take a decentralized identity management solution to move past that. Something like uPort[0].

To the second point - we are already probably at a level where basic needs could be met with relative ease, given the will[1].

[0] https://www.uport.me/

[1] http://mashable.com/2014/07/07/google-founders-interview-kho... etc.

Change to Show HN?

great idea!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact