Author: Marek Palatinus 2016-08-18 09:15:23
Published on: 2016-08-18T09:15:23+00:00
The proposed idea of building a "standard" on top of the URI protocol is seen as a huge limitation and does not provide any advantage. The issues with the current "bitcoin uri" standard have made it barely functional in most bitcoin apps, due to vague definitions of parameters or inability to send payment requests over it. Complicating things further with an HW API would not be helpful. Instead, the library approach could be used by any Android app, and the library would implement proper transports for various supported vendors, such as USB for Trezor and NFC for something different. This approach would make life easier for app developers without defining artificial "standards". The required data to verify and sign a (standard) bitcoin transaction including P2WSH multi-sig is manageable, making it possible for a signing device to require only a single round of communication. However, multiple rounds of communication might still be possible with a clever design, although >95% of users' transactions would require only a single "shot". Passing a 300kb chunk over a URI scheme requires a couple of milliseconds on standard smartphones or PCs, making it more convenient than the library approach. The URI-based API allows transmitting data of multiple megabytes without dependencies of any form, library support for a particular language, or platform that supports the dependencies of the library (like USBHID, which is not supported by iOS). While some third-party libraries could define high-level APIs and implement hardware-specific protocols and transports as plugins, this can be done already and does not require any standardization process yet.
Updated on: 2023-06-11T19:40:49.623759+00:00