"This week has been amazing. Such a diversity of things that happened and we achieved. Things are speeding up and getting more concrete and clear."
If one of your colleagues starts sharing these kind of messages you know that the week was pretty exciting!
- Iteration 3 of the use-case slides
- iOS memri browser
- Pod prototype
- Incorporation research
- Hosting architecture plan
Iteration 3 of the use-case slides
Continuing with where left of last week with the use-case slide deck. This week it was time to test it out and start using it for real. After a couple of interviews some first conclusions might be that most people understood what we're trying to build. And when they do, they get pretty excited! For those who didn't understand our idea yet, the challenge we have will be to get our ideas even better formulated!
To explore reactions further, we're looking into making a video where we share our product and ideas so that we can get more feedback! To be continued...
ps: here's another sneak preview of one of the functionalities: Viewing your data in a timeline view!
This week we focused on building docker images and setting it up in such a way that we can easily and efficiently host it and connect it to the front-end.
iOS memri Browser
Setting up a Apple developer account for memri proved to take a bit more time than we hoped. So we're pretty happy it is now confirmed and completed.
We spent this week refactoring the codebase and writing tests and documentation. The goal of this milestone is to be able to get the first (alpha) users to use memri! This refactor aims to create an easy to understand architecture based on the learnings of the past few months developing memri's iOS app.
Our main focus has been on the what we call Views, which are similar to Pages in classic Web Browsers. In the memri Browser a View is described in a declarative format similar to how HTML is used to define web pages. We're experimenting with something that looks a bit like CSS and HTML combined, and that allows one to very easily create a graphical user interface to view and manipulate memris. What we call memri is a data element and their inter-relationships. After this refactor we will have a solid base, documentation and tests to invite the world to participate in taking back control over each of our own data!
And, last but definitely not least; The rich-text-editor is working and it's behaving like we want it to behave. Which is always a good thing. See for yourself down below!
The Hosting Plan
This week we investigated various configurations for managing the pods with a focus on the future security improvements we have in mind, while keeping the hosting costs as low as possible. Our aim is to democratize our own information and that means that hosting needs to be as affordable as possible. Because memri is architected with privacy-by-design, the cost to run the computers that host the memri software is for the account of the user benefitting from it. And that means the cost needs to be as low as possible while the security of the data is as high as possible.
Our hosting and security plan will go in much detail and will be available on our wiki once we have completed the first draft.
We're also investigating a way for customers to subscribe and automatically set up a pod. We found some solutions that are worth having a closer look at, but no clear winner that connects billing and account management with Docker/Kubernetes provisioning.
Lastly we investigated a solution for backups that are incremental and allow for fast handover during failover (e.g. when a region goes down). Something we'll look into more next week.
Kick-off with Decentranet
Thursday we had a great meeting with Decentranet, that was also the kick-off for our collaboration with them! They will help and advise us with reaching out to the world. The first meeting already brought forth some great ideas! We're excited to get things rolling and to soon make memri available to world!