Operations group meeting notes for 2007-03-13

Attendees: Katie, Sheila, Jared, Philippe, Aparna, Ted, Mitch Priscilla and Mimi for the Logo part

Logo

Pris and Mimi introduced the semi final logos. Petroglyph not done yet. Rapid room feedback:

• Philippe: all look really good, none so bad needs to be rejected, thinks the grass treatment is pretty good
• Aparna: feel the grass is imbalanced
• Sheila: like them all, grass is pretty interesting
• Jared: think the starfish is the strongest (prosocial, adaptable)
• Katie: would like to see them in a Windows context (not just Mac), like the pinwheel as a concept, grass appealing but not sure works fine outside desktop
• Mitch: like grass but not sure it registers well, starfish stands out, easy to identify
• Ted: like them all, has most feeling for the grass

Need to have a follow up discussion on the pr list.

IT : mod_mbox work

• Revamped webui for the list archive.
• Budget: $1500 moving forward, $600 already spent.
• Need ops authorization. Ops approves this project.

HR : comp time non-policy

• Informal policy was that a manager can give comp time afterward when the demand is imposed by the work environment
• We do not however count the hours on a per person basis, we expect people behaving as professionals and taking care of that
• Personal comp time are covered by PTO
• For exceptional circumstances, OSAF allows unpaid leave

Strategy

Service level expectations

• Migrations take a lot of extra hours from Cosmo devs. This may include Build and QA.
• Outages take a similar toll on IT (Dave and Jared)
• Will we, should we have more people on pagers? What's the expectation? Right now the expectation is "best efforts".
• Issue is: we do not forecast a routine need for 24/7 support. The problem is how we handle the semi chaotic (with spikes) transition period till we get to the routine part of the cycle.
• As long as we don't see issues outside the migration period, we don't see a reason to change the current practice of "best efforts".
• Managers need to schedule this (migration) correctly and set expectation correctly for the relevant teams
• Managers can compensate time for exceptional work load if any
• We should talk about this at the staff meeting

Amazon hosting

• Amazon is hosting some services (S3, EC2) allowing to use the Amazon servers (e.g. run a Cosmo cluster from there)
• They have metering tools that would allow to run a private Cosmo instance for a small company for instance
• Could be an option for scaling post Preview

Post preview: adoption

• Communicate (conferences, blogs, etc...)
• Try to recruit through connections, influential people if possible
• Listening to user feedback, that group of people might be different to what we imagine our target to be right now
• Need to create some specific content for launch
• Map the traditional marketing plan with actions
• Interop has a special link to adoption, switching (from something else) is important for adoption
• We'll certainly want to have a faster pace of releases
• That means to change our development process so that we allow more projects to be developed in parallel while maintaining the trunk stable
• We also need to reach out to get developers involved

-- PhilippeBossut - 13 Mar 2007