Show HN: Songftware – Music Written Like Software | Hacker News Reader

Show HN: Songftware – Music Written Like Software

songftware.com

15 pointscrablar21 comments

I really, really hope this is satire. But I think it may be serious.

The author is pushing the same ideas in other places (sorry for the signup-walled Quora link):

http://jeffmeyerson.quora.com/Musicians-and-Engineers?srid=p...

As the OP appears to be the original author, I will ask the following: Do you really believe the "engineer mindset" is the best way of thinking in all situations? If so, why are engineers not the top musicians in the world? Many engineers are/have been in bands. But the majority of successful musicians follow at least some of the "bad traits" you list. Wouldn't the basic laws of consumer choice and capitalism bring "engineer musicians" to the top of the charts every time if those habits were really bad?

This is the author.

It isn't satire. Have you ever tried to write electronic music with a group of people?

Yes. I will hurriedly agree that the process is not perfect. But your followups to that first axiom are a little insane, and will alienate almost every musician.

You basically argue musicians should act more like you. Do you really want a world of yous? Because I don't, I believe there is room for every mindset. If you really want to improve electronic music production, try starting with the users. Find out what their needs are. I would hesitate to separate you from your users, but you appear to do that yourself in this article.

As in many industries, we settle for the best music we have. I posit that our music would be better if it were written with some engineering principles.

Yes there are many good engineer+musicians. But I've never seen a group of musicians use a KANBAN board.

I have successfully done so. One huge question: why would it be advantageous to scale groups of composers? Many people can do more than enough with no collaborators at all.

Example: you have a minimum-sized team that can work on that Trello board. One person for each column. After putting 20 songs through the assembly line, you find that things are getting bottlenecked at the percussion engineering part, and you are missing deadlines. So you hire another percussion engineer.

It's like a service-oriented architecture. If there is a lot of demand on a service, you spin up another copy of it.

Whats needed is better instruments, not better software. The problems of writing electronic music with a group of people is best solved by getting everyone sync'ed up over MIDI with their own local gear, and having at it .. you know, just like any other band over the years.

have you put in the time and achieved the musical results that warrant the kind of bold statement you are making?

I've written an album that's on Spotify, and several hundred other songs. I've probably put 3000-5000 hours into writing and thinking about music.

I'm continually proud and disappointed with my own results. I know I could do more if I was working with a team.

Every art form has processes and techniques. Not every process is named or enumerated. Many artists treat their original processes like trade secrets, or only make any effort to share them with their immediate circle of friends.

The "Why" described is that when the instrument is a single computer running a DAW, it has poor concurrency. You could change the instrument model instead to one that re-enables concurrency. You are doing the inverse presumably because you haven't had a good jam session before and only feel comfortable sitting behind the DAW. To get around this problem, you're rationalizing a process which is actually less efficient than jamming, as it straightjackets the creativity into a "waterfall" process instead of an "agile" one.

Actually, I think waterfall might make more sense for working on music.

The Postal Service used waterfall when making their album.

I haven't really found a true best way to collaborate on writing a song. Here are some of the ways I've written my songs:

Sit down at the piano, press record, and make up a piano piece on the spot. This is a fun creative exercise - the results often suck but sometimes something good comes out of it. You can't be super-original harmonically or you'll lose the thread and stop - but it's good for practicing discipline. [1]

Sit down with a couple of people you get along with, press record, and jam. Same thing as above - you can't get too crazy harmonically, but you can definitely mess around with layering and intensity. [2]

Write some lyrics and then try and write some music by yourself to fit around it. I'm not sure, but I think this sort of music leads to simpler song forms. [3]

Write some music to the point that they suggest syllables and vowel sounds, and then pound your head against the wall trying to find words that fit those musical moments you want. This is really, really hard but eventually rewarding. [4]

Sit in a room with your bandmates, jam together to come up with material, and then work/argue together to craft it into a song form. I haven't done this but I think it ends up with stuff that can be cohesive (good) or homogenous (bad). This can be done over a distance through dropbox etc also.

Come up with some electronica beats and then hire someone to come in and yell out phrases she reads from magazine ads, punctuated by dirty talk and sexual grunts. I haven't done this either, but it's responsible for a good percentage of the top 40.

(Links just for examples from my own history, listen or don't - I don't intend this as self-promotion.)

[1] https://soundcloud.com/curtsiffert/slowrain [2] http://thesalvagery.bandcamp.com [3] https://itunes.apple.com/us/album/damn-my-eyes/id899061469?i... [4] https://itunes.apple.com/us/album/she-believes/id899061469?i...

Very cool. A couple questions: is the song really the right vehicle to pass around? By analogy isn't that like moving the binary instead of the code? I understand the lack of DAW cross-compatibility but it seems like you might need all of the components that make up a song along the way in order to work on a subsequent task.

It looks like this is still in the experimental/conception stage. Right now it seems more like a mental model for music than actual tools. I'd be interested in seeing what analogies besides code would be applicable or might have lessons learned. For example, I know that collaboration and sharing of mechanical engineering design software is relatively less useable/mature than sharing a git repository. For artists that collaboratively work on 3D models or images I'd imagine that the workflow is passing around .psds or whatever, rather than final images because everyone is on the same platform.

I'll be paying attention as this develops.

Passing an entire song between phases is only one model. Within that model, the spec I prefer is: whoever is receiving the song gets to choose what format he receives it in. Lots of small files, one big wav, whatever.

I have heard good things about Splice, which allows git-like collaboration.

What I want to underscore: there are great tools for individual and small-group creativity. I'm interested in processes for a large team that allows asynchronicity and scalable development.

I think this is 20% genius and 80% confused. However, that 20% may completely outweigh the 80. (as always)

I have worked in all these environments and yes, there are aspects of process ... and principles, techniques, and mindset which can be taken from engineering and applied to creative pursuits. Good point.

I could have said a lot more, but I didn't want to deploy the wall of text. I would like to pique enough interest to get people to send me an email and talk about working together.

One of the fundamental flaws of this idea is that it seems to treat the process of making music as a means to an end that should be changed for the sake of vague improvements in terms of "scalability" and other inapplicable buzzwords.

For any musician who reads this and finds it interesting, please send an email to songftware@gmail.com.

I have a strong bias to action and want to experiment with the ideas I laid out.

I'm a little confused. Are you proposing a new framework for composing songs? Or are you hiring musicians interested in your cause?

Both.

If anything, I think it should be the other way around, write software more like one does music.

Reply on news.ycombinator.com