The April resignation of Eliot Horowitz, MongoDB's CTO and co-founder created a temporary but significant dip in the MongoDB share price.
It's not hard to see why. Since the very inception of MongoDB, Eliot has been a consistent and articulate owner of the MongoDB technical vision and the apparent creative force behind MongoDB's architecture and feature set.
In my opinion, Eliot's vision and connection with developers more than any other factor led to MongoDB's success. MongoDB emerged during the explosion of NoSQL solutions around 2008/2009. Some of these solutions—Cassandra and HBase, for instance—initially offered superior architectures or competitive features sets compared to Mongo. However, what set MongoDB apart from the very beginning was its laser focus on making developers' lives easier. Developers could easily find drivers, utilities, example code, and free courseware at www.mongodb.com while competing products often attempted an enterprise play as they left developers to figure it out for themselves.
Eliot's vision was never to create the most sophisticated technological solutions—he consistently focused on pragmatically solving real-world developer problems. He also was completely willing to take things that had worked from relational technology and replicate them in MongoDB.
Eliot left MongoDB at an interesting time. In terms of market opportunity, MongoDB seems perfectly positioned to ride an increasing momentum toward cloud databases and continues to be the default choice for new web-based development. However, I do get a bit of a "what next" vibe from the company—the technology appears to have hit a plateau of usability and it's not clear what the next few major releases will contain.
In appointing replacement CTO Mark Porter, MongoDB has chosen someone with a high degree of "geek cred" that will hopefully satisfy developers who want to hear from someone who can speak their language. However, he also brings experience from the enterprise database and cloud world that will undoubtedly be useful.
Mark Porter has a long history in database companies, having worked in Oracle server engineering for more than 3 years as the head of the Amazon RDS business. In his youth, he also worked at NASA and Caltech.
Similar to many of us who experienced the rise and rise of relational databases through the 1980s, 1990s, and beyond, Mark Porter seems to have a genuine enthusiasm for the greater simplicity and productivity provided by systems such as MongoDB. In his introductory blog post, he discussed how the mismatch between the world of developers and the database led to ORM (object relational mapping) frameworks that created absurdly convoluted database schemas and application code.
He says—and I agree—that while the relational model is probably the most "correct" representation of data, it's often not the best tool for getting the job done.
It's too early to see what if any impact he will have on the direction of MongoDB. Arguably, MongoDB only needs to continue with a "steady as she goes" strategy to maintain growth. I would argue that it's sufficient for MongoDB to keep developers happy and continue to perfect their cloud-based Atlas database service.
However, it's also true that if MongoDB aspires to displace the relational database heavyweights—Oracle and SQL Server—it will need to bridge a considerable technology gap. There's also a sense that a new breed of SQL-based technologies—CockroachDB, for instance—are rising stars in the database landscape and could threaten MongoDB’s position. Crafting a technology strategy to navigate these challenges will depend largely upon Mark Porter's leadership.