Hi everyone,
My status report for November 2024 is entirely focused on this Moat
issue:
https://gitlab.torproject.org/tpo/applications/tor-browser/-/issues/42605
Although I knew this task would be quite complex before starting, it
proved to be more complex than I thought, as you can see from the
numerous UI versions I made before realizing I’d need to step back to
study things better:
https://www.figma.com/design/4XiUaZPch3ejewlZO56lkR/Tor-Browser-14.5?node-i…
Since Moat can be difficult to understand for non-tech people,
comprehending how it works was the first step. For this, I used a CSD
matrix based on information gathered in a meeting with the devs. Later
on, it was necessary to understand the user flows where Moat calls are
made so we know when we need to ask for the user’s consent. So with all
that, I needed to answer some questions (e.g. what are we trying to tell
users and why?) so I can summarize everything in a way to educate users
on Moat calls and allow them to make an informed decision whether they
consent to it or not - I’m actually in this point of the task.
Besides that, I also worked on very small tasks of support to devs.
Best,
Priscila (aka Felicia) - UX Team
Hi everyone!
Here is my status report for November 2024.
November was a short month for us, as we took a team break to rest a bit
after the big 14.0 release and because of the holidays.
For most of the month, I investigated how to remove Lox's WASM blob and
instead integrate the Lox Rust crate in the browser build system [0].
Also, I checked the various options for exposing this integration crate
to JavaScript, as the front end will remain in that language.
This work will hopefully help us also when switching to Arti in the browser.
We've also discussed the future changes for the Tor daemon integration
in the browser, and I had to review a major refactor made by Henry [1],
which took me some time.
Apart from that, I rebased our 128-based channels (14.0 and 14.5), fixed
the release preparation script to take the needs of the legacy channel
into account [2], and made other similar fixes [3].
Also, I signed a release (13.5.10) for the first time. We decided to
allow more people to sign to be more responsive under some circumstances
and to spread the load on the team.
Then, towards the end of the month, I resumed the work on
anti-fingerprinting. I evaluated Thorin's proposal to abandon
`font.system.whitelist` in favor of font visibility, which would bring
us closer to Firefox [4]. While doing so, I found a problem with the
changes I made to the FontConfig configuration file the previous month [5].
Finally, I proposed a fix for a conflict between the OpenSSL 3.0.x we
ship with the tor daemon, and OpenSSL 3.2.x, required by a transitive
dependency, which prevented the browser from starting in some systems
[6]. If you have the same problem, deleting our libcrypto (form
`TorBrowser/Tor`) should work, but we'll provide a proper fix in our
next update.
Best,
Pier
[0]
https://gitlab.torproject.org/tpo/applications/tor-browser/-/issues/43096
[1]
https://gitlab.torproject.org/tpo/applications/tor-browser/-/merge_requests…
[2]
https://gitlab.torproject.org/tpo/applications/tor-browser-build/-/issues/4…
[3]
https://gitlab.torproject.org/tpo/applications/tor-browser-build/-/merge_re…
[4]
https://gitlab.torproject.org/tpo/applications/tor-browser/-/issues/43322
[5]
https://gitlab.torproject.org/tpo/applications/tor-browser/-/issues/43330
[6]
https://gitlab.torproject.org/tpo/applications/tor-browser/-/issues/43326
