Sixty-three percent of organizations that have open source program offices said those programs are very or extremely business-critical to the success of their engineering and product teams, according to a new survey from InApps, Linux Foundation Research, and the TODO Group.

That figure is up from 54% in 2020, indicating a growing awareness of the benefits of open source program offices (OSPOs) among participants in the annual survey.

The funding outlook for such programs also appears ready for an upswing: Fifty-one percent of respondents said an increase in financial support for open source initiatives at their organizations is very or somewhat likely to increase this fiscal year in light of macroeconomic conditions. Only 38% said the same in last year’s report.

Telecommunications and technology companies were most likely to have or to be planning OSPOs in 2021, according to the survey results, and healthcare organizations the least likely. However, an analysis of the survey’s historical data by Lawrence Hecht, InApps’s research director, showed that interest in government and educational institutions (specifically, universities) has become stronger as compared to other industries.

Here’s how the survey results breakdown by industry for organizations that have an OSPO, or are planning one:

Chart showing OSPO adoption by industry

The 2021 Open Source Program Office Survey was fielded by Linux Foundation Research and InApps from June 10 through June 29, 2021. The goal was to gauge the creation of OSPOs and better understand open source program office strategies for success.

This is the fourth consecutive year the study has been conducted. While the respondents represented fewer technology companies than in years past, the trend of participation by a broader cross-section of industries continued.

Respondents were solicited via social media and by direct emails from the Linux Foundation, the TODO Group, and InApps subscriber lists. Overall, 1,141 survey participants made up the final data set.

Read More:   The Benefits of AI for Service and Operations Management – InApps 2022

Professionalization and Opportunity

This year’s survey results reveal a steadily growing professionalism among OSPOs, as well as opportunities to educate organizations about their value.

Fifty-eight percent of survey participants — up from 54% in 2020 — said their open source program is formally structured, with dedicated staff hours, job titles, and the like.

But of those surveyed who either don’t have, or are not planning an OSPO, 49% said they believe their organization would benefit from a formal open source program, indicating an openness to the idea. And nearly as many — 42% — said they don’t know if their organization would benefit from an OSPO.

Other findings about survey participants without open source program offices, or plans for them, indicate room for further education:

  • Thirty-five percent of participants at organizations that don’t have an OSPO said it’s because they haven’t considered one.
  • Twenty-eight percent said they don’t see the business value.
  • Nineteen percent said they have never heard of OSPOs.

“What I’m seeing is a growing interest for OSPOS far beyond traditional tech companies,” Ana Jiménez Santamaría, OSPO program manager at the TODO Group, said in her interview with InApps. “This new OSPO era comes with its biggest challenges. Organizations need more education and guidance about how to set up OSPOs across different industries.”

The Linux Foundation and other open source groups are meeting the need to fill the knowledge gap. She added, “Foundations can take an important role to help organizations understand and be involved in the open source community in an efficient, transparent and healthy way.”

A Decline in Open Source Activity

The survey also presented some less rosy findings. The share of organizations that regularly use or contribute to open source software, or engage in activities that help build the open source community, declined in 2021 compared to the previous year,

Eighty-one percent of survey participants said their organizations frequently or sometimes use open source software for internal purposes, and 67% said they use it for commercial purposes; both of those figures were down compared to 2020, at 89% and 74%, respectively.

Furthermore, 42% of respondents said their organization’s engineers frequently or sometimes contribute upstream to open source projects, down from 48% in 2020. Thirty-six percent said they train engineers to contribute to open source projects, down from 42% the previous year.

The percentage of organizations that reported to either have a formal or informal open source program, or are planning to establish one, also declined compared to 2020: Thirty-four percent said they have an open source program, down from 42% last year, and 13% are planning one, down from 17%.

Read More:   6 TIPS TO OPTIMIZE THE SEARCH BAR DESIGN

The decline in organizations with open source program offices (OSPOs) was smallest among the largest companies. Sixty percent of organizations with more than 10,000 employees reported having open source programs, down from 64% in 2020.

So, what explains this decline? According to Hecht, who analyzed the survey data and produced the key findings, the drop in reported open source activities may be partly explained by the fact that the survey, now in its fourth year, includes a broader cross-section of industries than in the past.

In fact, the greater variety of industries participating in the survey could be a sign of the movement gaining fresh momentum, Jiménez said.

“Maybe this is telling us that OSPOs [are] being considered a more critical asset even outside the ‘traditional tech sector, and now other industries are starting to learn about the value of open source, willing to invest and implement new initiatives to culture the relationships with the organization and the open-source ecosystem,” she said in an email interview.

“Since these emerging OSPOs are just starting, they might be at an early-stage level in their open source journey,” Jiménez said, adding that the relatively new status that these companies have in the community might also explain the overall decline in OS activities, she added.

Chart showing decline in open source activity by year

Benefits of an OSPO

Even as the survey shows the share of organizations with OSPOs and engaging in open source activities declined this year, it also demonstrates that practitioners continue to overwhelmingly see the business value generated

Fifty-seven percent of survey respondents said they use their OSPO to further strategic relationships and build partnerships. That share was even higher in specific cases:

  • 77% of respondents whose OSPOs report to the CTO.
  • 77% of those who frequently contribute upstream.
  • 75% of survey participants who work for small tech companies.

Code quality, including documentation, has improved, wrote one survey respondent: “A number of projects end up creating synergies between emerging companies, which join efforts to solve a common problem and take the project to a higher quality level.”

The benefits of OSPOs differ according to the size of the organizations that sponsor them. The largest enterprises (those with more than 10,000 employees) were more likely to report benefits for license compliance and faster time to market for their products.

Read More:   Update Dissecting the Stack for Place, Reddit’s Collaborative Pixel Art Project

In contrast, the smallest companies in the survey (those with two to 50 employees) were most likely to report that their OSPO helped them achieve better security testing and increased market adoption for their products or services.

chart showing benefits of having an OSPO

Jiménez noted the survey’s large portion of participants who said their organization didn’t yet have plans for creating an OSPO. “I’ve spent years working for companies that were aware of the OSPO benefits, but they simply were not ready yet,” she said. “Most of the time, the main issue was implementing an open source culture internally.”

In those cases, she said, she believed that unless that cultural shift happens first, progress toward creating an open source program is stalled.

To open source advocates whose companies lack a formal program, her advice is to “start simple. Maybe start opening certain projects between internal teams, and promote open source culture inside the organization. Once that’s mastered, the organization will advance in its open source game and might be ready to open an OSPO initiative.”

In comments offered by survey participants, several presented detailed ideas for creating an OSPO. Several acknowledged that buy-in from upper management was key to getting resources and budget for such an undertaking.

“The main issue within my company is selling it to management types,” wrote one survey respondent. “I would establish the case in terms of possible efficiency improvements in the processes, automation potential, risk reduction in research/implementation, increased innovation, standardization, and better development recruitment prospects. It would help to have figures.”

What OSPOs Do

The study found that the chief responsibilities of OSPOs have shifted over the past four years. Just under 69% of survey participants with open source programs said that fostering an open source culture is their program’s primary responsibility; in 2018, that figure was 75%.

The second most commonly cited responsibility, maintaining OS license compliance reviews and oversight, dipped from 75% four years ago to just under 60% in 2021,

Engaging with developer communities rose from just under 51% in 2018 to 56% in the current survey. These efforts bear fruit. When this is a primary responsibility, the percentage of OSPOs that recognize the following benefits changes:

  • Increased contributions to in-house open source projects from external or third-party contributors (38% to 47%)
  • Increased market adoption of open source projects (34% to 43%).
  • Increased participation in external open source projects (43% to 57%).
  • More influence in open communities source (42% to 58%).

The full dataset on which this analysis was based can be found here.

Featured image by Finn Hackshaw via Unsplash.