Ask HN: How would you build a 1000 year Bandcamp? | Hacker News Reader

Ask HN: How would you build a 1000 year Bandcamp?

20 pointsrgbrgb9 comments

I’ve been mulling this over again after seeing the online outrage over Bandcamp’s acquisition. I think people were really upset because they see Bandcamp as an indie platform that tries to do The Right Thing by artists by staying small and taking a relatively small cut of sales to run the service.

I feel pretty mixed about it. I’m happy for the founders and early employees getting rewarded for their hard work but recognize that it’s the end of an era for bandcamp since their profit motivation and sense of ownership won’t be there for much longer. As a tech-optimist I’m excited about futuristic new platforms that could flourish in it’s place.

That said, I don’t really want to philosophize about any of that here.

Instead, my question for all the builders on HN… how would you, both technically and organizationally, architect a version of Bandcamp that would last 1000 years instead of 15. And I mean, kudos on 15, that’s ages in tech but clearly a lot of people wanted more.

- What stack would you use and what infra would host it?

- Would it be an open source project?

- How would you organize and incentivize development?

A few thoughts to kick it off…

- We all know software’s gotta change over time as platforms and user expectations change. The computers we’re using in 20 years are going to crush what we’re on now and user expectations will update (though probably a lot slower than we usually think). So we need a mechanism to keep updating it.

- If it’s going to last, it should probably not be possible for a big co to take it over and cover it with ads or something.

Just some Sunday fun :).

Patronage ala Rockefeller. Putting everything at the mercy of the free market will only lead to getting eaten by the Goliaths of that sector. Amazon sold its first book in 1994 but didn't make a cent in profit until 2004.

And that was started by a DE Shaw banker with a $300k loan from mom and dad! What hope does an artistic, not overly commercial venture have in that world?

Bandcamp was never going to survive on its own with its incredibly limited selection of mostly niche artists. Hell, Apple Music couldn't survive on its own when it was called Beats Music.

How could it, when numerous indie distributors before them were eventually swallowed up by the behemoths? Independent Digital stores like eMusic and CDNow came and went in the 00s. Before them it was indie labels and specialty record shops, all eventually swallowed up by a bigger player once the landlord raised the rent.

We have a 1,000 year old band camp actually, written music. It has a higher barrier to entry, is a rather lossy way to encode things, and took generations of copyists to perpetuate but a nonzero amount of the content stuck around as well as the skills to read, write, and perform it.

haha, I should have defined what I think is essential from bandcamp but I don't think you're going to convince many people that written music is a comparable substitute. For now we could punt on a rigorous definition and think of what we're looking for as something you could link to in the bandcamp acquisition discussions (or 2052's "bandcamp v9" acquisition discussion) and people would be like "awesome, this is a great alternative."

The question really becomes how do you make any platform/service that will stay true to it's mission and stand the test of time. The answer seems to be private ownership + the will of the founders to keep it that way and not sell out.

Craigslist is a great example of this; I highly recommend looking up Craig Newmark and his ideals. It's been around for nearly 30 years in more or less the exact same form, doing the exact same thing. It's privately owned and Newmark could sell the whole thing and be instantly propelled to billionaire status, but choses not to. There are no outside investors or corporate shareholders pushing for a bigger and bigger return every quarter. And no bloated teams of six-figure salaried engineers inventing work for the sake of work to justify their jobs and their resumes.

That mindset will at least keep the platform going for the rest of the life of the founder. On the 1000 years part, the next question is probably how would one legally structure a private company to restrict outside investments/purchases and straying too far from its original mission.

Blockchain. Seriously. A few non-updatable smart contracts and IPFS is about as close as we can get to "out of the reach of money" as it cannot be altered once deployed.

How can software that cannot be altered last 1000 years though? The environment is changing... I don't think there's any software from even 30 years ago that could run unchanged on any of my devices. I think there's a path with this stack but I don't see how you get there unless you have some way to do maintenance and upgrades.

Maybe there's a voting system or something for merging changes in?

Or maybe the immutable on-chain piece is just the contract with artists and there could be various players that are developed by proprietary groups (e.g. centrally hosted clients pointing at decentralized data, like how most daaps work). I guess then you still get into the bandcamp scenario where your favorite player gets bought, but data abuot you transactions would still be there... that would get us half way there.

I wonder if artists would agree to store music essentially for free in a public-ish place -- I think that would be a requirement of this architecture. In some ways they already do that with bandcamp (you can stream whatever for free, but not download files unless you buy), but this would be kind of a new level where you could technically download all the files. Something tells me it would still work for artists because I think the bandcamp audience is there largely because they want to buy the music directly.

spitballing!

Engrave it on stone in three different languages, use OCR to import it. If it must be digital media then store as plain ascii text files stored on multiple types of media. Plain text remains the only consistent digital media format over time. Digital archeologists will likely have jpeg rendering tools in their toolkit as well.

hm, my bad for a confusing prompt. the idea is not to come up with a way to preserve data for 1000 years, but to build an artist-friendly indie music platform that could last 1000 years.

Reply on news.ycombinator.com