[tor-reports] What Nick did in December
nickm at torproject.org
Wed Dec 31 18:44:55 UTC 2014
In December, I did some things that are programming, or related to it:
- I merged a bunch of patches in Tor, and reviewed a whole bunch of
code. Notable things (see changelog for more info)
- Thanks to patches from "teor", chutney networks now bootstrap
in seconds instead of minutes. This will make our scripted
integration testing run much, much faster.
- I merged a bunch of new test cases, including tests from
rl1987 for "resolvemyaddr" and for and
wrote a bunch more.
- Statistical analysis and port scanning for hidden
services. (Thanks to everybody who worked on these.)
- Unit test coverage is now up to 36.56% (up from 30.84% in 0.2.5).
Integration test coverage is at 45.96%, though integration tests
need more automation to actually do the "testing" part. Combined
test coverage is at 59.75%. This is far ahead of track for my
milestone of 60% by some time next year, and 80% by some time the
year after. It's a bit deceptive, though, since the integration
tests don't have enough "test" in them. Nonetheless, I think
this is decent progress.
- I released Tor 0.2.6.2-alpha. The change log is here:
- With help from Rob and Yawning, and idea from Bryan Ford, and
relying on Andrea's circuitmux refactor work, I hacked together a
quick Linux implementation of the KIST write scheduling
implementation for performance testing:
- I helped brainstorming for designs for improved transports,
hidden service improvements, and stuff like that.
And some things that are not programming:
- I met with Aaron Johnson and Rob Johnson, who were in town to
visit. We talked about designs for improved network performance,
improved bandwidth detection,
- I started the month with a multi-day retreat with Mike, Roger,
Wendy, and Andrew to try to figure out our organizational needs
and how to meet them. We came up with a fairly extensive list of
stuff to do, and shared it with the company. This is starting by
making sure that everybody has an assigned advocate to make sure
that they have a contact who can figure out beaurocracy issues
- I met with some HR people to figure out how to make our hiring
process more streamlined.
- I helped Karen a little with some fundraising stuff and Kate with
some communications stuff.
- I wrote a few blog posts, and helped draft a few more. Topics
included why people should be kind to each other, and why
interpreting science is hard.
What I hope to do this month:
- I want to spend more time programming.
- I want to finish building the initial design for the integration
testing framework, and specify more about module decoupling for
- I want to come up with code metrics to see where our biggest
- I want to freeze 0.2.6.x, or get it close to freeze.
- I want to finish ed25519 id keys:
- I want to merge improved work queues:
and consensus diffs:
and as many other pending patches as we can.
More information about the tor-reports