It turns out customers care about code contributions. At least, that’s one finding in the latest State of Enterprise Open Source report that Red Hat just released. The report’s author, Gordon Haff, said he “didn’t really expect” that so many (83%) would be “somewhat” or “much more” likely to select a vendor who contributes “upstream” to open source projects.
However, if we follow the money, it’s not obvious that customers actually are more likely to select vendors based on code contributions. Or, rather, it would appear that customers are much more nuanced in how they perceive vendor contributions than we may normally credit.
SEE: How to build a successful developer career (free PDF) (TechRepublic)
Open source matters
Open is fundamental to how software gets built today, whether you’re a software vendor or a software consumer (bank, retailer, etc.). According to the Red Hat survey, companies depend on open source for a variety of reasons, including IT infrastructure modernization (64% of respondents cited this), application development (54%) and digital transformation (53%). And why? Well, among other factors, IT leaders credit open source with delivering higher quality software (35%), access to the latest innovations (33%) and better security (30%). Years ago, cost was the primary driver, and now it barely ranks sixth.
SEE: Research: Digital transformation plans shift due to COVID-19 (TechRepublic Premium)
Given the importance open source software plays in enterprise IT strategy, it’s perhaps not surprising that customers want to see their vendors participating. This helps them offer better support for the software customers care deeply about. The survey asked, “When deciding on a software vendor, what impact does knowing that they contribute to the open source community have on your decision?” Eighty-three percent of IT leaders said they are “somewhat more” (45%) or “much more” (38%) likely to select a vendor that contributes to the relevant open source community.
This sounds great, and it comports with my personal preferences. As a longtime open source advocate, I want customers to feel this way. The unfortunate reality is that they don’t. We don’t need to survey customers to discover their true feelings: We just need to see where they’re spending money.
But perhaps even that’s not a fair representation of customer concern. Maybe, just maybe, when we say “contribute to the open source community,” customers think beyond simplistic leaderboards.
You keep saying that word…
I’ve worked for open source companies for a long time, and I have listened as customers said they wanted open source databases (even as they kept shoveling cash into Oracle), open source content management systems (they kept buying SharePoint), etc. I’m not suggesting that customers didn’t actually want open source alternatives: Look at the success of open source startups over the past decade or two, as well as the rising adoption of open source infrastructure, and it’s clear that enterprises want open source, just like they said in the Red Hat survey.
But it’s also clear that customers don’t only care about open source. And even when they do care about open source, they don’t necessarily think about contributions exclusively (or even primarily) through a lines-of-code measurement. My team recently ran a blind survey of developers and other IT professionals. When we asked customers to define open source leadership for cloud providers, Figure A shows how they responded.
The most popular “contribution?” Making it easy to use that open source software without needing to get a Ph.D. in the particular open source project. In fact, when Red Hat asked about barriers to embracing open source software, the number one barrier today (just as it has been for as long as I can remember) is perceived lack of support, cited by 42% of respondents. The more vendors can minimize the manual labor associated with running open source, the better off customers will be.
We’ve definitely seen this play out in the market. MongoDB, for example, has long offered a great database, but it’s the company’s cloud service, Atlas, that is supercharging the company’s growth. Ditto pretty much any open source (or cloud) company you can name: Customers are flocking to their managed cloud services.
Which is a significant contribution, in its way. Years ago Tim Bray wrote that a company can “develop stuff [and/or] operate stuff. I think the second is more important.” I’m not sure I’d say it’s more important, but it’s certainly of equal importance. It’s no good writing software that enterprises can’t actually use due to complexity or other problems.
Bringing this back to open source, yes, contributions matter, but it seems customers have a more holistic view of the nature of contributions that matter most to their productivity.
Disclosure: I work for AWS, but the views expressed herein are mine alone.