MongoDB outlines what today’s developers need to succeed

MongoDB outlines what today’s developers need to succeed

CTO of the leading NoSQL database provider discusses relational snobbery, the triumph of JSON, the importance of trust, how companies mismanage developers, and how the third tier needs to evolve.

Mark Porter (MongoDB)

Mark Porter (MongoDB)

Credit: MongoDB

With regard to the role of patents in general, I focus on two aspects — engineering and commercial. There is a certain purity in bringing an engineering idea to such clarity that you can express it in a set of claims that form an elegant onion, building on the idea layer by layer. And engineers should be proud of that — after all, reducing chaos to order is literally what we do.

In addition, from a commercial point of view, it’s important for companies to have portfolios they can use defensively to protect against the trolls out there, the ones trying to make money without adding any actual benefit to the universe. 

I’m proud of my patents, and we also have an opt-in patent program here at MongoDB that helps engineers be proud of their innovations — and there are a lot of them in progress.

Tyson: Larry Ellison is such an iconic figure, what was it like to work with him?

Porter: Haha, now the gloves are off, is that it? Larry is indeed an iconic person. I’ve found that leaders like him, or Andy Jassy at AWS, or even my current boss, Dev [Ittycheria], here at MongoDB, set the culture for the company — all the way down to every person typing furiously to build or support customers at the company.

Larry has a mixed reputation, no doubt about it. My interactions with him were technical — around building database and video server technology — and his passion was always to build the right elegant product, the one that would save customers money and help them move faster. I learned a lot from him during the years I worked both indirectly and eventually directly for him.

For example, we had a meeting culture where all exec staff meetings were Monday, then the next level was Tuesday, then down through the company during the week. By doing this, every single employee at the company had the opportunity to hear about new ideas or directions from Larry’s staff meeting, in person, with the ability to comment and ask questions within a single week.

I think where Larry struggled and continues to struggle is that he lets the senior executives around him build a culture of not treating customers well, and he doesn’t jump in and course correct that. All in all, I’m a Larry fan and deeply value the 13 years I had the privilege of working with him and Oracle. 

That said, I think the culture of engineering empowerment, intellectual honesty, and good intent that Dev has built here is pretty fantastic — and I’m still pretty much a student of that particular game.

Tyson: I read that you did some coding on the Apple II in Pascal, and I have to tell you, that brings back memories. (When you were creating software to help Alaskans learn trades, I was writing Ultima clones :)

In the same article you say that “every management level should have an equivalent individual contributor leadership level and the pay should be equivalent.” That really struck me. How can we convince companies to make it so? Especially given the prevalence of the belief that one has to stop coding and start managing at a certain point?

Porter: First off — Ultima. What a wonderful world that was. It was amazing what we could do with 64K of memory, a processor running just over a million eight-bit instructions per second, and 140K on a floppy drive, right? Crazy.

Back to your question about having to go into management to succeed. This is a real hot button with me. For the last decade, at News Corp, AWS, Grab, and now at MongoDB, I’ve worked to have equivalent individual contributor ladders and management ladders. And not only equivalent in pay — but equivalent in responsibility and influence. 

For example, at MongoDB now, Distinguished Engineers are at the same level as Vice Presidents and involved in appropriate levels of decision making and planning. But, like you say, there is this prevalent belief that you have to go into management to make the most money and have the highest title and the most influence. Total hogwash. 

At each of those companies, I’ve written a document about the differences between being a senior individual contributor and a senior people leader. Both roles care deeply about the company, of course, and about the people. 

But the people leader takes a deep visceral interest and holds responsibility for every one of their people’s struggles, growth, compensation, and career. Whereas the senior individual contributor mentors people but also focuses just as viscerally on the quality of the code, the processes, and architecture.

Tyson: I read somewhere that you keep up your coding chops by teaching your kids programming (Scala, Java, and others). Do you have any insights on how to maintain that elusive work/life balance?

Porter: Frankly, given that my kids or wife might read this, I better be honest and admit that we’re all still struggling with my work/life “harmony,” as I like to call it. I’ve always loved working and still do, and it’s sometimes hard for my family to see me making some of my choices. 

That said, I also love my dear wife and soulmate and my five kiddos and I love spending time with them. I’ve made progress. I have even learned to disconnect occasionally in the last couple years.

You might wonder what I mean by “harmony” rather than “balance.” In Agile, we talk about a “prioritised backlog,” which is just a fancy way of saying “a list of things in strictly prioritised order.” But people think they need to divide work and home and build two backlogs. 

Where I’ve seen people fail, and where I am trying to do better, is by having a single prioritised backlog rather than two separate ones which both have a “priority one.” When you have two backlogs, you have to have some meta-prioritisation process to determine which priority one to look at. 

People think this works because “balance” implies some kind of time budget— and so they rigidly agree on boundaries of time and boundaries on clocks and calendars. But with a wife, five kids, and an executive job, that just doesn’t work. The needs of my life surge back and forth between the two and can’t easily be predicted. So the system has to account for that.

Sometimes, for days or weeks or even months at a time, the items on the top of that merged list are mostly work — and sometimes they are mostly family. I’ve found workplaces amazingly supportive during times of family stress, and my family has been unbelievably supportive — pretty much all the time. 

The key, if I had to name a single thing, is not to let the pressure of both — that meta issue of having two priority one items fighting with each other —  get to you. Don’t be in “the sandwich.” Instead discuss what’s going on with all your stakeholders and come to an agreement.

On a side note, I owe the birth of this philosophy to the recently deceased Clay Christensen and his piece “How will you measure your life.” I strongly suggest that every parent, executive, and partner read it.

Tyson: What are you most hopeful or excited about in terms of where software and technology might head in the near future?

Porter: What I’m most excited about is what I call the “new third tier.” We used to think about three tiers as “client/server/database,” or “presentation/application/data,” and the developers for the first two tiers had to provision, wire up, and maintain everything in that third tier. Not only that, but the third tier was just data, not the manipulation, movement, or governance of that data. That’s all changing.

The third tier that developers of today need has to have so much more in it. Look at the cloud providers with their hundreds of services, all trying to provide these foundational services. 

But the problem is that they gave people a toolbox and parts, not the well-designed integrations they need today so that they can focus on their business logic and building apps. I believe in a completely different approach to that third tier. 

With new best-of-breed software platforms like MongoDB, we can re-invent the third tier to have all of those mission-critical capabilities that developers need. Starting with the database, but adding built-in search, mobile sync, visualisation, streaming, analytics, and everything else developers need—all in an open-standards, integrated, composable way.

My personal goal is to make the developer of today feel like they are working with an elegant and delightful system, one that handles all their data fluidly while not locking them into anything they don’t want. 

And, pretty shamelessly, I’ll close up by saying that if you want to learn more about all this, you can see me at MongoDB World in New York, our first big in-person event since the pandemic started, June 7 through June 9. I’ll be doing a keynote and “meet the CTO” sessions as well. I look forward to seeing you and your readers there.

Tyson: This is a tantalizing vision of things to come. Thanks again, Mark. I’m excited to watch where both you and MongoDB go in the future.

Porter: Thanks so much for the chat. It’s been fun and I look forward to readers’ comments. I can be reached at @MarkLovesTech on Twitter or on LinkedIn at MarkLovesTech. And I truly do love tech.

Tags MongoDB

Show Comments