[bind10-dev] some suggested topics for the team call

[JINMEI Tatuya / 神明達哉]

I also want to manage the time slot for each topic better (again) by
giving a specific time period so we won't easily consume all 90
minutes.

- Revisit poorer progress in recent sprints and how we can improve it
  - managing review queue
  - setting clearer priority
  - possibly better use of daily call
  - possibly using historical statistics (needed hours, gap between
    estimation and actual, etc)
- ever lasting homework.  these seem to get stuck, but for the beta
  release I guess these are becoming more important and urgent.
  - making bind10.isc.org top page nicer/more user friendly
  - security procedure status update, set up separate repository,
    practice, etc.
- DB backend usage
  it doesn't make sense to discuss what kind of DB we use and how we
  optimize it without knowing who wants to use it.  Personally I've
  never heard the expected usage; it's always been guess.  We really
  need this info, and if we currently don't have it, need to plan to
  get it.
- PostgreSQL vs MySQL
  People talk about PostgreSQL backend for BIND 10 recently, but from
  my quick experiments, it has some missing feature in C API
  (https://lists.isc.org/pipermail/bind10-dev/2012-October/003868.html)
  and (at least in very limited benchmark scenarios) generally slower
  than MySQL (see, eg.
  https://lists.isc.org/pipermail/bind10-dev/2012-October/003866.html)
  MySQL also supports query cache while PostgreSQL doesn't.  Depending
  on query usage it's another advantage of MySQL.  I have no
  personal reason for pushing MySQL, but experimental facts so far
  seem to indicate MySQL is probably the better choice for our next
  target.
- How to improve query performance with DB backend
  https://lists.isc.org/pipermail/bind10-dev/2012-October/003866.html
  It involves substantial schema changes.  The timing of supporting
  another backend is probably a good opportunity to do this type of
  change.

---
JINMEI, Tatuya