Did you know that you can upload files directly to the Bitcoin network, as permanent, immutable, and monetizable Bitcoin transactions?
These files can be all sorts of things like documents, images, and videos, and can be combined to make other things, like applications or interactive pages.
With files, you can publish things that reference one another through URI schemes.
And because it's on the blockchain, ownership is provable, the content is permanent, and it can be directly monetized forever.
Bottle is a browser that lets you surf the Bitcoin network for these things, and brings them all together through Bitcoin native URI schemes such as B:// or C:// (or any other protocols we add in the future).
Note that the address bar below uses a b:// address:, not HTTP or HTTPS. This HTML file is 100% hosted and served from the Bitcoin blockchain, and has nothing to do with where a "server" is located.
Let's take a deeper look. It's not just about static files like images. You can even publish and surf Bitcoin-native hypermedia documents.
In the following HTML/JavaScript application, note there is no "HTTP". Everything is served from B:// or C://.
HTTP is the protocol for the old world, the server-client based cloud paradigm.
By moving away from HTTP and authoring everything in a Bitcoin native way (using Bitcoin transaction ids and content hashes) we can build a completely self-contained network of documents which can exist forever on the blockchain.
1. Hello B://
Bottle supports B://, a URI scheme for public files, which uses Bitcoin transaction id as the URI.
2. Hello C://
Bottle also supports C://, a content addressed URI scheme that generates URI based on the file contents, instead of transaction id.
3. Bye HTTP://
We start from scratch. With Bottle, the Bitcoin protocols are the first class citizens, not HTTP. Eventually the goal is to phase out all HTTP and HTTPS references except for a handful of trusted hosts (of course, users could easily add hosts to the whitelist manually).
4. Infinite Protocols
Bottle is just getting started. The goal is to come up with the most sensible way to support various Bitcoin-native protocols. We welcome feedback and contribution.
Why a standalone browser instead of building as an extension for existing browsers, or waiting for mainstream browser support?
1. Build for the future
Many things we take for granted in the old "web browsing" experience--including the security model--no longer apply in the new world of Bitcoin.
The thing is, Bitcoin is NOT "the next web". In many ways it's completely opposite of what the WWW is, which is why Bitcoin is so powerful.
That's why it's more beneficial to start from scratch instead of forking an existing full-fledged browser built for the existing WWW, with many legacy features that can constrain future directions. We can create a new user interaction model optimized for the new Bitcoin world order.
2. Bitcoin-Native
Bitcoin has a fundamentally different architecture than the old web in many different ways, with built-in immutability, a self-contained authentication model, and natively monetizable/traceable files.
Instead of thinking from the old WWW mindset, we should think from a Bitcoin-native mindset.
Bottle can discipline us to publish Bitcoin-first documents, build Bitcoin-first apps, each interconnected to one another in Bitcoin-native ways.
Bottle plugs into B:// or C:// endpoints, which can be constructed by crawling the Bitcoin blockchain. Anyone can spin up B or C nodes as a Docker container using Planaria, or build their own implementations.
Bottle is powered by Electron, which means we not only have all the usual Browser features out of the box, but also have access to more potentially powerful features.
Bottle listens to all protocol requests prefixed with B:// or C://, transforms them into their corresponding HTTP/HTTPS requests and hits the B or C endpoints, in realtime.
This approach can be easily extended and even modularized to support future protocols without creating huge technical debt.
Instead of scouring the content to replace all B:// or C:// occurrences with HTTP/HTTPS counterparts, this is a clean solution that doesn't touch the content at all, making it infinitely scalable.
It simply transforms the traffic. This approach even lets you handle dynamically generated URIs, which would be impossible to handle using a regular expression string replacement approach.
Browse CSS
Browse Images
Browse Videos
Browse PDF
Browse (and Use in your app) Executable Script Code
Browse HTML documents that reference all of the above inline