As another separate point that I'd like here but isn't part of this comment.

I think that all transactions _inevitably_ break down when (at least) one party has no skin in the game or when they misestimate their or the counter-party's stake. (Call it the busterarm theory of economics).

The only possible skin in the game that Apple has here is their brand. Their offering here is basically a value statement that Apple's brand and what it brings is worth more than artists will be losing in streaming dollars.

I think Apple is significantly misestimating both their risk and their counter-party risk here. _And_ the risk to their brand is effectively zero. It'll be interesting to see what happens. My impression of their music offerings was that their real value was the trust they had with both artists and consumes; they probably don't see it being that way.

One cannot operate a paid media streaming music while simultaneously proposing that the value of that media is zero. There's that old business maximum that if you want to know what a company cares about, look what they spend their money on.

> 4 albums a year

And that would be an incredibly productive artist. Most artists release something like 1 album every 1-2 years, at best.

Maybe in genres of music that you're accustomed to. There are a few where high output is the norm. The output of a lot of producers and (especially) ghost-writers is pretty hard to fathom. I have a great story about Japanese record labels that's pretty long for here though.

Many well-known artists have released multiple hit records in the same year before though and many more are releasing their own albums and producing other peoples. In the early 60s popular artists sometimes put out two hit records a year (Bob Dylan, The Beatles '63-'65)...Led Zeppelin did that once in '68. David Bowie put out two hit records AND co-wrote and produced two Iggy Pop records in '77. CCR put out three LPs one year...

It happens a lot more than you realize.

Another thing to note is that artists often (can) produce far more music than ends up on their albums; they, or their producers / publishers, realize that not everything they produce is good music. Just like code / software. So while you can produce thousands of lines of code a day if you churn out enough, only ten or so will actually be worth keeping.

Agreed...but I hope my (or anyone's) useful output as a programmer is significantly more than 1%.

I really want to drive home the "you're going to work just as hard, but for less money" argument to get people thinking about it the right way.

It's no different than music piracy, which actually I don't care about, just don't be dishonest about doing it as a business.

    > Exposure isn't necessarily good for an app. Users 
    > frequently _cost_ money. Ask anyone who has
    > significantly scaled freemium services.

Then it's not a business, it's a charity.

Not necessarily. Cost of users isn't linear with count of users and while it might go down as your userbase grows, you might not have the finances yet to handle that growth.

You can't just "be ready to host a 100 million users, tomorrow" in the early stages of your business.

If you don't have a contingency plan for how you will handle - operationally, financially, infrastructure-perspective, monetization - user scaling at speed, then not only is your business a charity, it's a hobby, too.

If there are credible signs you will acquire 100m users tomorrow, and you can't walk in to a VC and walk out with a big chunk of money for a small amount of equity, it means you don't have a real business plan.

Additionally, I would be interested in any examples you have of companies acquiring 100m users overnight, or anything within two orders of magnitude of that.

Accidents happen. Partners cheat. Here is a simple example from Shift Jelly about Amazon App store and their free app of the day.

http://blog.shiftyjelly.com/2011/08/02/amazon-app-store-rott...

> All we got was about 300 emails a day to answer over the space of a few weeks, that left us tired and burnt out. For all we know most of the people who wanted our application, now have it. To add insult to injury Pocket Casts relies on a server to parse podcast feeds (allowing instant updates on your phone), and all these new users forced us to buy more hardware just to meet demand. Hardware that we are going to have to support indefinitely at our own cost.

Can you fault Shift Jelly for not being able to see through this?

Amazon offered them a deal. Amazon made the terms of that deal super clear by putting it in bold type. Shifty Jelly accepted that deal. Calling that cheating is beyond comprehension.

That's the thing about "exposure" though. You have no control over it. You can't plan for it. You are missing the point. 100 million was just a number. The corpses of businesses that scaled past their ability to serve their customers are real.

But part of your argument has merit: it's the same reason that "exposure" doesn't create album sales. There's no real chance (really, there isn't) that you have a product that will reach 100m people tomorrow, so you shouldn't be producing music as if you would.

    > You can't plan for it.

Sounds like a good way to get sued by your investors.

    > The corpses of businesses that scaled past their ability 
    > to serve their customers are real.

Turns out that a Rails App + someone calling themselves a "Serial Startup CEO" on LinkedIn don't necessarily equate to a business.

> Sounds like a good way to get sued by your investors. You can try and anticipate it and you can model the risk and try to mitigate downside risk. You cannot "plan" for it though - it's a factor outside of your direct control. It's not a target you can hit or not hit, you just are prepared and lucky either way.

> Turns out that a Rails App + someone calling themselves a "Serial Startup CEO" on LinkedIn don't necessarily equate to a business.

Rails, like any other tool/framework, works perfectly well for certain kinds of problems. I would say this is very true for the kinds of businesses that most "serial startup CEOs" are solving. Twitter gave Rails a bad reputation because Twitter's problem was exceptionally bad for Rails at Twitter's scale. In most other cases it's just poor engineering discipline, whether it's in choice of tool or how it is used. It turns out that ActiveRecord isn't really a great tool for dealing with 450M+ write transactions per day. Wasn't designed to be.

Rails is powerful and ubiquitous for very good reasons. The tools that often get the most hate end up quietly doing their jobs or, worse, growing up and becoming better and everywhere. JavaScript, PHP, Lisp, C, Java? And that's just a few programming langauges...There's a holy war over Text Editors that aren't ever going away.

It's not a "web 2.0 startup" specific problem we're talking about. Rapid growth is notoriously one of _the most common_ reasons that companies fail (regardless of if that failure puts them out of business or not).

It's a marketing channel. You pay real costs to provision and support free users so that some of them will become paid users.

It only works if there is a high enough proportion of conversions to paid.

    > It only works if there is a high enough proportion of
    > conversions to paid.

That is additionally what makes it a viable business. Loss-leaders are nothing new, in any industry, despite the cute name of "freemium".