Everybody hates MIMEs - Challenges in keeping p≡p engine development free

From 35C3 Wiki

Description This talk discusses some of the particular challenges we've faced in terms of supporting development of a free software core whose customers (app developers and users) live in the world of seamless, proprietary software. The lessons learned may not be new to anyone who has tried to produce a complex solution using only non-proprietary software libraries, but the specific discussion of the challenges of MIME and encryption in a free software context where the goal is maximum usability, cross-platform compatibility, and extensibility provides interesting use cases for further discussion.
Website(s)
Type
Kids session No
Keyword(s) security
Tags p≡p, email, privacy, free software development, MIME, cryptography, RFC nightmares, software engineering
Processing assembly Assembly:Free Software Foundation Europe
Person organizing
Language en - English
en - English
Other sessions... ... further results

(Click here to refresh this page.)

Starts at 2018/12/29 18:00
Ends at 2018/12/29 19:00
Duration 60 minutes
Location Room:Lecture room M1

p≡p is an end-to-end privacy-focused security architecture which aims to provide a seamless private multimodal communications solution for everyone, focused on email as the first external product. One of our core goals is to maintain the complexity of our client applications in the free software core, isolating both users and application developers from a steep learning curve and ensuring as much of the code which impacts security and privacy remains free, transparent, and universally auditable.

As such, we have to reach users and app developers where they are, and this means engaging with them on the platforms they use. This presents the usual problem - users and app developers are often bound to proprietary platforms which are (sometimes intentionally) difficult for free software developers to target.

Some of the biggest challenges relate to ensuring maximal cross-platform compatibility without licensing nightmares. Worse, not every piece of core functionality we need has a reasonable free software alternative that works for us; developing such a broad, multiplatform core exposes issues with even standard, widely used free software libraries that have, over time, become costly in terms of time, fault detection and isolation, and usability.

This talk discusses some of the particular challenges we've faced in terms of supporting development of a free software core whose customers (app developers and users) live in the world of seamless, proprietary software. The lessons learned may not be new to anyone who has tried to produce a complex solution using only non-proprietary software libraries, but the specific discussion of the challenges of MIME and encryption in a free software context where the goal is maximum usability, cross-platform compatibility, and extensibility provides interesting use cases for further discussion.