Date and Time: 6:00am PDT / UTC 13:00, Thursday October 08, 2015
Convener: Bin Hu (AT&T)
Uli (Huawei) summarized that in this proposal, Chris (Ericsson) outlined 3 cases.
The first case is an easy one. Committer steps down voluntarily. No argument.
The more difficult case is when those committers are inactive or unresponsive. And also the definition of "sufficient evidence" is vague and unclear. Michael B (CenturyLink) suggested that if a committer misses 2 consecutive votes in a row, the committer can be seen as "unresponsive". Uli (Huawei) asked what if someone has an extended vacation, such as 6 weeks in summer in Europe, which is quite common. Then it will be too easy for removing a committer. Bin (AT&T) indicated that some company may have policy that doesn't allow employee to work during vacation. Then the discussion eventually reached the following condition as "inactive" and "unresponsive":
- not present in the project meeting for more than 6 months; AND
- not contributing any patch for the project for more than 6 months; AND
- not participating in the email discussion of the project for more than 6 month;
The third case is "disruptive". It is clarified that different opinion is not "disruptive". And in fact, different opinion is encouraged and valuable, and healthy for the project to make it better. The "disruptive" is pretty much like if a patch gets "-1", but the committer still ignores the "-1" and merges the code, which may break the code and/or delay the progress. So it is pretty much some type of wrongdoing. Of course, it needs more clarification such as severity of the outcome etc.
Michael B (CenturyLink) suggests that we may use the word "retire" instead of "removal".
Uli (Huawei) has the action item to revise the proposal based on above discussion, and try to find a better wording than "removal"
Bin (AT&T) indicated that this item was listed for further discussion, because it was on TSC agenda for vote this week. But it wasn't voted on Tuesday, and couldn't find any record of why the voting didn't happen.
Uli (Huawei) said that he wasn't sure why it wasn't voted. His understanding of the issues are:
- ATC is unclear;
- The timeline of election process
- In some cases, upstream projects doesn't want to have a different governance model from OPNFV. For example, ONOSFW. They want to have the same list of committers as that in ONOS upstream project. But Ashlee is not present, otherwise we may hear more from ONOSFW.
Parviz (Juniper) asked if Board needs to approve such procedure. Parviz further asked how to guarantee platinum members will have 2 seats in TSC. Uli (Huawei) clarifies that platinum members are guaranteed to have a seat. The other seat may be through election. Parviz suggested that each platinum member should have 2 seats, one designated, and the other is the project lead from the platinum members. Of course, only one project lead to limit the total number 2 of one platinum member.
Parviz (Juniper) will drive the change of TSC Charter to allow 2 voting members in TSC for each Platinum member per OPNFV By-laws:
- A representative designated by each of the Platinum members shall be TSC member (per section 5.5 of By-laws) and
- One project lead from a Platinum member shall be TSC member. If there are 2 or more project leads from a Platinum member, only one project lead is allowed, who is chosen by that Platinum member.
We ran out of time. Postponed to next week.