Apache Board of Directors - a retrospective

2019-03-28 19:06 More posts about Apache board director
Last term I had the honour of serving on the ASF board of directors, better explained in context of the ASF governance structure. As quite a few directors (myself included) declined to run for the board this year again, I thought it would be a good idea to think about the past term, write that down and publish those thoughts. As I wanted to give every board member a chance to respond, I shared some guiding questions but left it to board members to choose the channel they deemed most appropriate to share their responses on. So far, two decided to make their answers publicly available: When posting the questions, I intented to share my own answers as well - but refrain from publishing them for as long as possible to not influence others. So here we go: For a bit of context where I'm coming from: When I was first nominated for board I didn't see that nomination coming at all. This was back in 2016. I did accept, I was voted in. Back then I was still in a technical role, even though what I had been doing in various teams and open source projects was much more on the leadership and management side oftentimes. As a matter of fact, after that one year it was time to make a shift in roles in my dayjob as well. I took a one year break in 2017, was elected as board member again in 2018. Seems like twelve months is just about the timeframe that works for me - maybe some of the answers below will explain why that is the case.
Which areas were a lot of fun for you?
What is truely the most fun part is working with excellent people who have a ton of experience and know what they are doing - even though they themselves are sometimes in a mode like "we just need to make this up as we go because there really is nobody who has been faced with these challenges before".

The other part that is really fun is that even besides the fact that there is quite a bit to do, there's always someone who keeps the morale up, who brings their humour to the team. Be it on the board members meeting IRC backchannel or other places.

If you are not subscribed to board@: Every quarter projects are supposed to submit a quarterly board report. To guide their reporting and make it easier for those parts that can be derived from project communication statistics, there's a report generation tool. It will generate a project board report template that includes headings for some sections that can only be filled by a human. Those sections come with default text - text that is entirely non-sensical - but fun to read, should projects forget to fill in the real content. Trying to remember from memory, as I'm offline for the coming weeks, it was something along the lines of

==Project health narrative
Minions took over your project and are having a party.

The last part is the chance to bring people together, work together, trust each other across large distances - to the point of sharing family stories. While face to face meetings are awesome for bonding with people, those I've encountered so far as directors were easy to work with across distances even if I had met them just a couple times for a few days years ago at an ApacheCon. An "assume best intent" kind of attitude is certainly helpful there. On a tangent that also means that typically I'm communicating with people on eye level, only to get scared for a split second weeks and months later when coming across their LinkedIn profile and reading what they do or have done for a living...

Which parts were particularly educational for you?
  • The way the Apache board runs board meetings with several dozen agenda points in very well under two hours by moving as many decisions to asynchronous channels as possible. This makes it possible to participate from many more time zones than I have seen at any other organisation. It also means that people don't have to align on when everyone has time to participate in the meeting and can do most of the discussion and decision work whenever they happen to have time for it spread out over many days. On the flip side this means the entire process feels like it's slowed down a lot - except it wouldn't be possible any other way as those know who have tried to get all nine directors in one room at one date at one time during that day.
  • I had no idea how budgeting is done at the ASF's scale before being part of that discussion. I was very glad to know that there are other directors who have experience with that topic.
  • In every board meeting we go over project board reports: Everytime there are some where I'm like: "You should totally talk about what you're doing there to a wider audience than just your project." - it's great when that advise is actually followed up on - it's even better as so far everytime such learnings were written down behind a stable URL, that URL became very popular as a reference for other going forward.


Are there any parts of being a board member that you could imagine helping with even after stepping down?
Pretty much all tasks the board does can be done by any member. One thing I seriously would like to continue doing is to build more bridges between the board of directors and projects (inviting new PMC chairs to join the board meeting and take away their fear that it might last hours is just one possibility).

If asked I will also continue answering questions on best practices, how to get involved and generally try to pull people in - both into projects as well as into the foundation itself.

With a bit of hind-sight: A few months after leaving the board of directors, what I did continue doing was make time to explain the Apache Way to other people. I'm also interested in keeping the discussion on how to get involved - and how to convince your employer to donate your time to the foundation alive. Another thing I would really love to find best practices and patterns for is how to balance full-time open source contributors with those working on projects occasionally. This isn't much different to the challenge of integrating part-time and full-time employees - something that at least in some German companies has been solved for a long time already. Time to also find a solution for Apache projects taking project neutrality one step further.

Which areas were particularly time costly for you?
We have well over 200 projects. Each is required to submit a board report on a quarterly basis. While most aren't particularly long, the content does add up. While in theory, checking private lists, following a few discussions and listening to project contributors (be they users, contributors, committers or pmc members) would be needed to spot governance issues early, the current number of projects to check for each shepherd means that issues will be discovered too late for precise, scalpel like patches. Somehow that entire process is something that in my opinion could benefit from a better collective understanding of what constitutes a healthy projects, including an understanding of just how much freedom there is in configuring your project governance best practices.

Maybe the Apache maturity model would be a good starting point for spotting best practices. Maybe it would be helpful to also come up with project anti patterns to watch out for to take the guess-work out of the process of writing a board report.

Which areas were energy costly for you - didn't necessarily take a lot of time but were definitely not fun to deal with?
There's just one example that's standing out for this one: Occasions where the board needed to delegate tasks to a new role. The main reason here was that typically the discussions quickly dissolved into mixing role name, role accountabilities, person to fill the role and budget for the role all into one discussion.

While for the final result all four perspectives need to come together, in my experience, it helps to first think about the actual accountabilities that should be delegated. Only after that look for a name that fits, otherwise people will associate accountabilities because of the name, believe that those accountabiities aren't needed and reject the need for the role entirely.

Only after having exact accountabilities start discussing budget - it's much easier to talk about money, once it's clear exactly what the purpose looks like that the money is going to be spent on. Once that is out of the way, start looking for a human to fill the role. Even if there is an obvious choice popping up, make your need for that role to be filled known and go looking for volunteers. You'd be surprised to hear how many new faces we've seen speak up and get active - lowering the load on any usual suspects.

"I wish I had known this before joining the board"
The ASF is really good at hiding where it was incorporated in it's daily operations. Being a director that suddenly moves a lot closer: It's becomes much more likely that your communication becomes of interest in a suppeona (luckily, there's an archive for pretty much everything, so the relevant communication can be handed over fairly quickly). For me this lesson moved the ASF a whole lot closer to a regular company with a BYOD policy though.

In terms of how to run a US Delaware law incorporated 501.c3 foundation I was lucky insofar as there were still several long term directors as well as former directors around that knew about and were happy to talk about the constraints of what could be done and what couldn't be done - and why so.

In your opinion - what are the strengths of the ASF board?
  • They have developed an awesome model of collaboration with an asynchronous decision making process that does scale.
  • The people I had the honour to serve with remained human beings, including the humor and including their private family background.
  • The ones I have seen serve in their individual capacity leaving corporate interests at the door - or announcing when discussions are started because of what they observed at dayjob.
  • They bring a ton of experience from all sorts of backgrounds - both, culturally but also in terms of which companies they've seen from the inside - trying to learn from all of that experience to build a better world that is much more flexible and light weight than your average large scale enterprise.


In your opinion - what are areas for potential improvement for the ASF board?
A lot of the things that I think of reading that question really are things that could be improved even without being a director: Documentation on how the foundation works at the project, and at the operational level; documentation of expectations of projects, making purpose and actual content of previous decsions easier to discover...

The only thing that comes to my mind that the board could watch out for is to have discussions in such a way as to keep all relevant people in the loop, facilitating a dialogue with people instead of about them. For contentious discussions it might also help to have a set of moderation techniques that have proven helpful in asynchronous communcation handy.

In your opinion - what changes should be made to the way the ASF board operates, interacts with communities, interacts with the wider ASF ecosystem, interacts with the public?
While at the ASF we are very good at thinking about inward facing communication, I believe we can become much better at communicating to the non-ASF world. As we grow, I believe that perspective becomes ever more important - we cannot rely on people to just "get" how we work from simply following our projects and becoming ever more involved.

Being community over code to me implies counting the user/customer of our projects as part of that community. They are the ones who have the fresh eyes to spot flaws in our argumentation of why we are doing things the way we do - and spot advantages. However they can only do that by understanding what goal and purpose we want to work towards.

Any advice for new board members - where to look first, what legal implications to keep in mind, what PR implications to keep in mind etc.?
While being a director can be done even with little time available, there are others on the board that dedicate a lot more time to the task than you will have available. Don't let that stress you out. For some discussions it helps to wait a day before writing an answer - either someone else will have writting what you wanted to say after that - or you'll be in a much better state of mind to make your statement without it being hidden behind a cloud of emotions that stop you from thinking clearly.

While being an ASF director may mean little to the people you work with - even if your dayjob builds their entire business on top of an Apache project - it does bear a lot of weight to many other people worldwide: ASF internal, or ASF external (e.g. press representatives). With that power comes a call for using it responsibly: The statements you make, even if they were only meant as a basis for discussions, will have the power to cause fear if not marked as such.

People will seek you out with the issues they are having in their community. Instead of jumping in yourself which does not scale long term, try to figure out how to help them in a sustainable way - e.g. by helping to find the right more public space to talk about the issues they have encountered and discuss a solution in public.

Remember that the board at the ASF is designed to move slowly. As a result do not expect to make ground breaking changes within just one board term. However do expect that the little changes you do make will make a big difference in the long run. Essentially that's a side effect of a foundation that's build on the goal of longevity.

One thing that is different at the director level (as well as the operational level): While in a lot of positions we have setup process that allow for people to take a break without announcement for any amount of time, director and operational positions do mean that there is a need for dependability and reliability - you will need to make time for that position regularly for at least one year.

What are the tasks and time commitment?
That totally depends on how much time and energy you want to commit. At the bare minimum if it's only about oversight a couple hours a week and maybe two days running up to board meetings could be sufficient (that's where projects submitting reports early make life easier for directors).

Tell us about a moment from your time on the ASF board that is most precious to you.
I tried to think of one - except there wasn't. While it does take time and energy, it's also a great source of positive inspiration, including tons of chances to learn from others.