[tor-project] minutes from the sysadmin meeting (OKRs edition)


At our first meeting of the year, we didn't have time to complete the
web team OKRs and prioritization for the first quarter, so we scheduled
another meeting to do this. Here are the minutes.

We might have more of those emails in the weeks to come, as we have a
bunch of brainstorms and planning sessions coming up. Let me know if
this is too noisy...

# Roll call: who's there and emergencies

anarcat, kez, lavamind, linus joined us.

# 2022 Q1/Q2 web OKRs

gaba and anarcat previously established a proposal for a set of OKRs for
the web team, which were presented during the meeting, and copied below:

## Proposal

* OKR: make the donate page easier to maintain and have it support
   .onion donations ([milestone 22][])
   * [new lektor frontend][]
   * [vanilla JS rewrite][]
   * [we can donate through the .onion][]
* OKR: make it easier for translators to contribute (almost done! not
   ambitious enough?)
   * translators can find their own copy of the website without help
   * bring build time to 15 minutes to accelerate feedback to
   * allow the web team to trigger manual builds for reviews
* OKR: improve documentation across the organization
   * [launch dev.tpo][] (Q2)
   * "Remove outdated documentation from the header", stop pointing to
     dead docs
   * we have a plan to fix the wiki situation so that people can find
     and search documentation easily

[milestone 22]: Improve Tor's donate page · The Tor Project · GitLab
[new lektor frontend]: Integrate new front-end with lektor (#37) · Issues · The Tor Project / Web / Donate Static · GitLab
[we can donate through the .onion]: Get stripe donations working via onion (#36) · Issues · The Tor Project / Web / Donate Static · GitLab
[vanilla JS rewrite]: donate-static vanilla js rewrite (#45) · Issues · The Tor Project / Web / Donate Static · GitLab
[launch dev.tpo]: dev.torproject.org work (#6) · Issues · The Tor Project / Web / dev · GitLab

## Progress update

The translation CI work is already going steadily and could be finished
in early Q1.

We are probably going to keep prioritizing the donate page changes
because if we postpone, it will be more work as updates are still
happening on the current site, which means more rebasing to keep things
in sync.

## Things that need to happen regardless of the OKRs

We have identified some things that need to happen, regardless of the

This key result, for example, was part of the "documentation" OKR, but
seemed relevant to all teams anyways:

* teams have less than 20 tickets across the three lists (backlog,
   next, doing), almost zero open (untriaged) tickets

We also need to support those people as part of sponsored work:

* s9 usability - Q1/Q2

    * support web maintaince based on the UX feedback

    * Work on torproject.org usabilty issues based on user feedback

    * Work on community.torproject.org usabilty issues based on user

    * Work on dev.torproject.org usabilty issues based on user feedback

    * phase 6 may bring more TPA work but we need to make the schedule
      for it with TPA

* s30 - anti-censorship - Q1

    * bridges.torproject.org - Q1

* s61 network performance - whole year

    * support the work on network simulation

* s96 - china censorship - whole year

    * support snowflake scaling

    * support rdsys deployment

    * support moat distribution

    * support HTTP PT creation

    * support monitoring bridge health

    * support creation and publication of documentation

    * support localization

* s123 - USAGM sites - Q1/Q2

    * support the project on onion sites deployments

    * most of the work will be from February to April/May

    * new onion SRE and community person starting in February

Non-web stuff:

* resurrect bridge port scan
   * do not scan private IP blocks: kez talked with cohosh/meskio to get
     it fixed, they're okay if kez takes maintainership
   * make it pretty: done

Some things were postponed altogether:

* decide if we switch to Weblate is postponed to Q3/Q4, as we have
   funding then

We observed that some of those tasks are already done, so we may need to
think more on the longer term. On the other hand, we have a lot of work
to be done on the TPA side of things, so no human cycles will be wasted.

# Prioritise the two set of OKRs

Next we looked at the above set of OKRs and the [2022 TPA OKRs][] to see
if it was feasible to do *both*.

[2022 TPA OKRs]: 2022 · Wiki · The Tor Project / TPA / TPA team · GitLab

Clearly, there was too much work, so we're considering ditching an OKR
or two on TPA's side. Most web OKRs seem attainable, although some are
for Q2 (identified above).

For TPA's OKRs, anarcat's favorites are mail services and retire old
services, at least come up with proposals in Q1. lavamind suggested we
also prioritize the bullseye upgrades, and noted that we might not want
to focus directly on RT as we're unsure of its fate.

We're going to prioritise mail, retirements and upgrades. New cluster
and cleanup can still happen, but we're at least pushing that to
Q2. We're going to schedule work sessions to work on the mail and
upgrades plans, specifically, and we're hoping to have an "upgrade work
party" where we jointly work on upgrading a bunch machines at once.

# Other discussions

No other discussion took place.

# Next meeting

TPA mail plan brainstorm, 2022-01-31 15:00 UTC, 16:00 Europe/Stockholm,
10:00 Canada/Eastern


Antoine Beaupré
torproject.org system administration

1 Like