tl;dr — Every successful social startup should bend, but not completely violate, one or more social taboos, e.g. one of the 7 deadly sins.
Edit: When designing www.fwdeveryone.com, we purposely incorporated Pav's theory by building the platform around the taboo of getting involved in other people's relationships, in this case by letting people read (and reply to) other people's email conversations.
As a side note, it amuses me to see this repeated so often, given that it's not true. If you do the math, Spotify pays about the same as a radio play. You just have to remember that each radio play goes out to a few hundred thousand to a few million people each time.
The reason artists don't like Spotify is not because they pay poorly, it's because they pay fairly. When you get a radio play, it's assumed that all the people the station reaches listened to the song. Spotify knows exactly how many people listened to a song, and pay accordingly. Unpopular artists don't like this because they get accurate counts.
From what I understand, they actually pay based on how many times a song is played, which is not the same as how many people listened to it. If I only listen to an unpopular artist, but I don't listen as often as other people listen to popular artists (lower play count, even though it's still my favorite artist), nearly all of my subscription fee goes to the popular artists.
In the old days of purchasing music, it didn't matter how many times a person played the disc/track - it cost the same. You might argue that each play represents the marginal utility to the listener (and should thus be the basis for compensation), but I'm not sure that's entirely true.
Yes, I should have been more precise. Just like with a radio play, the second play in theory goes to the same mass of people, but is counted again. So I should have said "they know exactly how many people and how many times it was played, and pay accordingly". I believe they discount multiple plays from the same person, to prevent abuse.
> In the old days of purchasing music, it didn't matter how many times a person played the disc/track - it cost the same. You might argue that each play represents the marginal utility to the listener (and should thus be the basis for compensation), but I'm not sure that's entirely true.
You might argue that, but you might also argue that when you bought a CD, you bought 10-15 tracks, even if you only wanted one or two. So again, for a consumer, Spotify is more fair, because you only pay for what you consume, and artists liked the old model where you paid for 15 tracks even if you only wanted two.
My same argument applies to a single track purchased via iTunes - it doesn't matter how many times you play it; it costs the same. The bundling of tracks on an album is something all artists did/do, so it doesn't affect the fairness question in my opinion (at least, not fairness between artists).
I was similarly never willing to pay for HBO before on-demand streaming, but am willing to pay now.
That said, it's an easy statement for me to make given that I was buying music on vinyl, then CD, then digital downloads before there was much in the way of streaming. So, even leaving aside the vinyl, I had a large collection of owned music before modern streaming hit in a big way. I still buy things every now and then but I have a large core of older music I care about--which is the bulk of what I'm interested in.
This is not true IMHO. You need approximately 4 million plays per month to get the equivalent of Californian minimum wage! [1]
[1] https://mashable.com/2018/01/31/copyright-court-rules-stream...
Edit: You might get might more if you own all the rights yourself but not many artists do.
Just last week I was belittled, because I didn't want to come to office two times a week at 9am.
Of course, it came at the expense of artists, who aren't making as much off of Spotify as they did off of album sales. Well, not the top X artists anyway, I like to think that the artists that used to be in the long tail are now doing pretty well for themselves. Relatively.
...oh wait, that's a real thing that you did. You just didn't put the http:// in your comment to avoid self-promotion.
After reading your examples (which are a really helpful way of showing what your service does, BTW - thanks for not burying that in a video or forcing me to sign up), I can see the appeal for your service. It's Instagram for email: In exactly the way I keep every photo my phone takes backed up and organized on Google Photos, but selectively post to Instagram, I could selectively send mail on a "social email" account and keep normal transactions private. Fascinating! I suppose I have a small portfolio of emails helping some people solve problems (running/cycling tips, local routes, etc and some tech support) that I have meant to post to a blog but just never got around to it. Making that blogging automatic would be neat!
Nice idea and implementation, I apologize for mocking it! I'll definitely keep the principle in mind.
We currently use read-only OAuth, so basically the same in terms of friction but easier to secure.
> And users wouldn't need to inform people who email them that their email is public
We actually currently require double opt-in permission for each message contributor in a thread to share stuff publicly. That might change in the future, we're still playing around with the model.
The new best practice with social apps is basically to launch while erring on the side of having too much friction for the app to be successful, and then keep successively reducing the friction for people contributing positively to the community. Otherwise you risk gaining a bunch of immediate traction but going into the death spiral of negativity, which is the main way that social startups die if they already have initial traction.
This wouldn't have worked ten years ago, but it makes more sense in a world where things take off more slowly anyway due to increased network saturation.
Interesting concept, but that one is a bit of a stretch...