August 27, 2021 was the last day on Amazon Web Services (AWS). I spent two years there, most of them running the company’s open source marketing and strategy teams. On the surface, it helps the world better understand the open source work that AWS does, but in reality it spends most of its time within AWS and the related open that the product team may depend on. Helped me understand why and how to contribute to the upstream of the source. The rest was spent outside the company, working with open source companies such as Confluent and Databricks to improve AWS partnerships with these companies.
Oh, and on the way I helped Extinguish the dumplings This happened when AWS was perceived to be doing “bad things” to open source companies and communities.
In my experience, much of the anger directed at AWS via open source is irrelevant. No, AWS isn’t perfect, but AWS is one of the world’s largest contributors to open source projects. (Whether you’re measuring by the number of active contributors or by code, you can see your data in the following ways: Run this code.. Rather, it’s almost always an error to think of AWS as a monolithic entity with a general approach to open source. This is one of the main misleading myths about AWS, but there are many more to try to tackle here. What I write here is nothing secret, but it’s as if Amazon hides it all at a glance.
Two pizza teams
Amazon founder and former CEO Jeff Bezos Established “2 Pizza Rules” Early in the company’s history, “We’re trying to create a team that’s no bigger than two pizzas can eat.” This is a bit exaggerated, but the principles are adhered to throughout AWS.Teams tend to be relatively small and, just as important, most teams Fully autonomous..
What does this mean? Well, it may be correct that Service Team X is not currently contributing to open source projects, but the same is not true for other service teams (more than 200). For example, the ElastiCache team employs one of Redis’ five maintainers.Other teams are making important contributions to rust, Apache Lucene, Kubernetes, Open Telemetry, etcd, Apache Iceberg, OpenJDK, GraphQL, etc.
Is there a service team that hasn’t used open source upstream yet? Of course, just as it is in Microsoft, Google, Alibaba, etc. But while I was working on AWS, I saw this change. This is a slow process as Amazon rarely works in top-down flats. If you want Amazon to contribute more, you need to focus on individual teams. Equally important, we need to speak Amazonian.
Yes, those LPs are real
“Speaking Amazon” refers to the language and thinking process embedded in Company 16 Leadership Principles (LP). Before joining AWS, I thought LP had to be a cheesy slogan at best, and in the worst case Jingoism, but a common framework for more than a million Amazon employees to talk to and collaborate with each other. I found that it was provided. They are the famous “6 pagerIs used to raise ideas and determine operational plans.
When I joined AWS, I initially avoided using LP in discussions. It didn’t work. “Project X is the right thing to do, so we need to contribute,” I say. A blank gaze. “We need to give engineers more time to contribute to Project Y so that we can influence the direction of the project.” I didn’t go anywhere.
But then I tried to put together my argument in LP, and things got much better. I began to say, “It’s difficult to achieve innovation in a’simple’without being obsessed with customers and building trust with the communities they depend on by providing code.” This also allows you to “claim the highest standards”. This is because our contributions will help us “provide results” and better support our customers. “
Suddenly people understood what I meant. It wasn’t crowded before.Rather, I had to speak a language that called the governing principles All This is done on AWS (and in fact all Amazon). If you want to change the behavior on AWS, you need to use LP to frame the desired result. My team is increasingly skilled in doing this and is rewarded with the involvement of a larger service team than ever in the projects they depend on. This does not mean that there is no room for improvement.
The spirit is pleased
While at AWS, no one despised the importance of open source. The exact opposite. It’s fun to caricature AWS as a bunch of evil Henchmans enthusiastic about strip mining open source, but I’ve never met anyone who fits that description. Rather, when I disagree with the general manager of the service team and others about the relative importance of open source to their business, our disagreement is always the heaviest in a particular situation. I came back to the LP that I did.
For example, “customer attachment” comes first for all Amazons, but it can be read in a variety of ways. You may think that open source contributions are important for customer attachment in the medium to long term, but the general manager of the service team should also consider them in the short term. This means that you need to create a private branch of your code so that your enterprise can fix bugs quickly. Or provide the functionality that the customer requested. Also, while all customers seem to like and want open source, many are even more focused on getting their code up and running ( Tim Bray pointed out years ago).
This means that ensuring a seamless customer experience in the short term can consume engineers who may be contributing to the long-term success of a particular project. We’ve seen this change for the better on AWS, but again, a company full of two pizza teams doesn’t have a one-size-fits-all approach to open source.
Also, LPs such as “ownership,” “invention and simplification,” “claim the highest standards,” and “provide results” contradict the desire to work well with commercial stewards in open source projects. It also means that it looks like. If a customer wants to make Apache Kafka easier, the immediate response is to build a service that can be managed on their behalf, with as few opportunities for moving parts and failures as possible. Another response is to partner to ensure a seamless customer experience. AWS probably thought it would be easier to offer the first option historically, but was encouraged by all the successes we’ve seen with Confluent (in the case of Kafka) and other open source companies.
AWS still has a way to go in this and other areas, but we all do. One of the things I’ve loved most about open source for over 20 years is how much we have to learn as an industry (and as individuals and businesses) despite years of trial and error. That is to say. No one has cracked the code in the perfect way to build and run an open source project or business. We are still learning.
So let’s be patient with each other and try to understand why individuals and companies operate that way, as I did here for AWS.
Copyright © 2021 IDG Communications, Inc.
What you don’t know about working on AWS
Source link What you don’t know about working on AWS