Hi all !
Time flies and you guys are due for another update from us here at MaidSafe. Our company operations are no longer local. They have followed the worldwide growth of the community. The dev team in Troon of eight is now reinforced by no less than four remote developers spread over the world. They are found in Ireland, Denmark, Slovakia and the US. Last week Paige from San Francisco visited our offices and so did Francis from Quebec, Canada. Both the remote developers and the independent SAFEpods serve as a proof of strengthening ties for the MaidSafe project worldwide. We are very keen on growing even faster, so feel free to reach out and contribute!
This global expansion called for reorganising our entire workflow. As a result we can now better coordinate our own work with the work of our remote developers. The new workflow puts everyone on an equal footing, and that includes you. All you have to do to get started is fork your personal copy of the MaidSafe code. Once you’ve done that you can follow the contributor workflow.
Great, so you want to dive into the code? Talk to us on the dev mailing list and we’ll help you register at maidsafe.atlassian.net. Here you can follow the tasks we are currently working on. Always make sure you are working on a defined task, or propose a new task to work on on the dev mailing list. A general rule of thumb, put all your work of one task in a new branch that starts with the task ID (eg MAID-179). If you use the same branch name across different submodules for the same task, we can track it as a whole.
What is even better, when you’ve completed a task, you’ve built it and ran the tests, you can submit a pull request to the next branch of the MaidSafe code. We will test your code cross platform and review it. You’ll have followed the exact same procedure as any MaidSafe developer on the project! We only need to figure out how you can buy us yum-yums, when your code breaks all platforms
This improved workflow will also allow us to build a more detailed roadmap for everyone to follow up on. Finding better ways to keep you closely updated is really important to all of us here! To wrap up, let me briefly highlight some work from running projects.
When we closed last weeks’ sprint we drew a line in the sand. Non-crucial failing tests have been temporarily disabled, important tests have been fixed. This has reduced the coverage of tests on our code base, but in return we know where we stand. All new code can only be accepted if an accompanying test increases our code coverage.
On rUDP the design of a brand new rUDP 2.0 is being worked out and has been proposed to the boost ASIO community. This overhaul will promise a great increase in data transfer speeds. However, our focus stays on - call it - rUDP 1.0 for the launch of the SAFE network and the introduction of congestion control remains primary.
Important work in routing has been completed. A design change now implements ‘send acknowledgement’. This beefs up the fault-tolerance and security of the close group at routing level even further. Traffic will flow around a bad node. To test this code we are working on a new tests. With a focus on routing initially we are also introducing a new category of tests: performance tests. With performance tests we will continuously track the performance of our code changes in a real network.
See you soon
your MaidSafe team