Feature-based crowdfunding

[ghost]

Back in mid 2013, I had an idea that Hive would actually be crowdfunded on a micro scale, i.e. on a feature-by-feature basis. Looking back at those mockups now (and wow, 50 BTC was worth a lot less then), I actually think this was not such a bad idea. It would allow users to literally vote with their money on the kinds of things that they want to see going forward. This in turn would steer the direction only as much as we don't parameterize it (which is to say, we want only to 'offer' things we're committed to doing if funded anyway).

Thoughts? CC: @mikehearn who has been doing a lot of thinking about this kind of stuff, but maybe not this level of granular detail in particular.

[mikehearn]

That's exactly what I'm implementing with Lighthouse, right?

I talk about crowdfunding of individual features here: http://blog.vinumeris.com/

So yes, I am hoping this model makes a lot of sense and I'd love for Hive to be using Lighthouse/on vinumeris.com when they're ready.

[ghost]

How does it work exactly to be using it on vinumeris.com? There will be a web interface to this, correct? And I assume this will be Bitcoin-only?

[mikehearn]

I think the blog posts discuss this, but yes Bitcoin only (the crowdfunds are managed on the block chain). There is a desktop app for making the pledges and claiming them, although it vaguely resembles a web app visually. Usage of a website simplifies some things but is optional. I plan to run such a website at vinumeris.com for Bitcoin and decentralisation related projects. Websites are for things like project discovery, discussion, social integration, collection of pledges from the desktop app and "pre-pledging" i.e. promising an amount denominated in dollars/euros but without actually interacting with any financial system (just to reduce the impact of volatility).

[ghost]

So this will require pledgers to download something, correct? Just want to make sure I 100% understand what "optional" means. :)

[mikehearn]

Yes, pledgers always have to download the app and that will not be changing. The only other way to do it is to have what is effectively a temporary bitbank/hot wallet that stores peoples pledges and allows the pledgor to get their money back on demand, or when the project expires. That introduces a lot of security and possibly regulatory headaches I could do without.

[sull]

i'm into this approach. have written on it and its part of active project plans. micro-granular-feature-reason-based tipping.

[ghost]

Hey @w-hive!

I was going to build this in rails but rails has a steep learning curve (atleast in my experience) so I have decided to build a special version for Hive built in PHP + Laravel. I should be done in around 2 weeks!

Ill message you on skype asking for more specifics!

[ghost]

Hi Nahtnam, I don't want to discourage you, but at this point (~8 months after we talked about it), I'm not 100% sure that this is the direction we're going to go with it. We might go with the project by @mikehearn now. I do still want to see this happen, but I don't want to waste your time either...

[ghost]

@w-hive Kk. Ill create it either way... If you dont want it, then it will be a great learning experience.

[ghost]

Good attitude. 👍

[ghost]

@mikehearn Rather than having a separate download (which I think is understandable, but inconvenient), can we integrate the Lighthouse functionality into for example hive-android and hive-osx? I asked about this in private some time ago; what is your take now?

[mikehearn]

Yes, that would absolutely be possible. What I'd suggest is a two-step process. In the first step you'd add a simple Hive App that just lets people "pre-pledge", which is effectively a promise in dollars/local currency to make a real pledge in bitcoins in future. You literally just give an amount and email address which goes into a database. The goal is to avoid long term fundraising which would be exposed to BTC volatility. Then people still have to download the app when enough pre-pledges were gathered and make it real, but they did promise so hopefully that overrides peoples natural lazyness :-)

In the second step you could reuse/integrate some of the Lighthouse code and give it a Cocoa/Android UI. You'd probably only do the most minimal UI possible though; Lighthouse is quite complicated because it lets you create and manage projects too rather than just pledge. You wouldn't need that.

Reusing the code should not be hard as you're already bitcoinj based. I would need to resolve the Java 8/Java 6 difference however, either by downgrading my code or finding enough neat hacks to let the newer code run, like RetroLambda. Anyway, technically it means:

• Upgrading to bitcoinj 0.12 + HD wallets, as Lighthouse is also HDW based. On Android you can just wait for Andreas to do this for you and then rebase.
• Switching out the Wallet class for PledgingWallet, which is an extended version of the bitcoinj wallet which knows how to create and reserve pledged outputs, store pledges to disk, do the special signing etc.
• Integrating the LighthouseBackend class (or parts of it), which knows how to parse project files, query the remote server for the project status, upload new pledges and expose user-relevant data model changes on the UI thread.
• Wiring up the data (e.g. list of pledges) coming from the backend to your custom Cocoa/Android UI, whatever that may look like.
• Having a project marketplace that lets the user browse and access projects (files) that you're trying to raise money for. When selected the project file should be downloaded, stored on disk and the path given to the backend class.
• Finally, making sure your user community actually finds out about this awesome opportunity to give you money :-)

Lighthouse isn't open source yet, so the above plan isn't doable until it is.

OTOH pre-pledging, building an enthusiastic community and possibly even the project market is something you can start on right now.