Opinionated summary of Tidelift State of Open Source Maintainer report

[rahulporuri]

Get the report from here or here. If you don’t know who/what Tidelift is, Tidelift partners with and pays FOSS maintainers.

I hope the summary makes you interested in reading the entire report because it’s worth the 1-2 hours that it will take to read the 60+ page report.

Sadly, only 8% of the 344 respondents of the survey (25+) are from Asia, and we’re not sure how many of them are Indians. So please keep that in mind when you’re trying to interpret the results of the survey in the Indian context. Developers residing in India (and broadly Asia) are usually underrepresented in most Software Developer surveys.

• After screening for quality and completeness, we analyzed the answers from 437 respondents who maintain at least one open source project.
• The most cited stat from that previous survey was that 60% of maintainers described themselves as unpaid hobbyists. We asked the same question again this year to see if things had changed. As it turns out, they have not changed a bit.
• Meanwhile, the percentage of maintainers saying they earn most or all of their income from maintaining projects is almost identical at 12% this year versus 13% in 2023. And the percentage of semi-professional maintainers was 24% this year and 23% in 2023.
• Paid maintainers are more likely to have co-maintainers, unpaid maintainers are more likely to be flying solo
• Paid maintainers are able to spend more time/resources to find co-maintainers (33%) vs unpaid maintainers (23%)
• Paid maintainers are able to spend more time/resources for feature development (84%) vs unpaid maintainers (55%)
• 78% of unpaid hobbyist maintainers are working ten hours or less per week.
• An overwhelming majority of maintainers prefer to receive predictable monthly income, with 81% choosing that option.
• In this survey only 5% of maintainers report receiving income directly from companies (this answer choice was not an option in previous years). Another 5% report getting direct payments from individuals, which is steady compared to 2023, but much lower than the 10% of maintainers receiving this type of income in 2021. And only 3% of maintainers report that they have received income from open source foundations, which has remained steady across all three surveys (it may be surprising to some that this percentage is not higher)
• Because governments around the world have taken a greater interest in open source software security over the past few years in the wake of prominent security incidents like SolarWinds, Log4Shell, and xz utils, we asked maintainers in our latest survey whether they were receiving income directly from governments or other public entities. But to date, this income source is a non-factor, with only 1% of maintainers reporting receiving direct payments from governments or other public entities
• We plotted out the ages of open source maintainers from the first survey we completed in 2021 through this year’s survey, and what it shows us is that the percentage of maintainers self-reporting that they are 46–55 or 56–65 has doubled since our first survey in 2021 (2021: 11%; 2023: 27%; 2024: 21%). Meanwhile, the percentage of maintainers under 26 has dropped precipitously from 25% in our 2021 survey to 12% last year and 10% today.
• Almost half of respondents (45%) have been open source maintainers for more than 10 years. Meanwhile 24% have been maintainers for 6–10 years and 23% have been maintainers for 2–5 years. Only 7% of respondents reported that they’ve been a maintainer for 1–2 years and 2% reported that they’ve been a maintainer for less than a year, which may be another troubling signal that the current crop of maintainers is aging and not being replaced by a new generation.
• Only 6% of maintainers identify as female, which is slightly down from 8% in 2021 and 9% in 2023, but probably not statistically significant given the sample size.
• European countries represent the largest group of open source maintainers (48%), followed by North American maintainers (38%)

[sudo_navendu]

Thank you for the excellent summary!

I might be too close to this, but adding significant data from India and other Asian countries might skew these results.

These countries have a significant contributor/maintainer base and thriving COSS companies. Even top projects under the Apache Foundation have large maintainers and commercial companies backing them from countries like India and China.

With that said, here are my thoughts after reading the report summary.

An increasing number of maintainers work for for-profit companies with an open-core business model. They might be underrepresented in the survey. From a strictly personal point of view, I see more people and companies taking this approach to building software.

Maintaining hobby open source projects is hard. Finding new maintainers is hard. Developers who have the skills to contribute either don’t have the time to contribute outside their work or aren’t interested. This could also be due to a lack of newcomers in project communities. But I do see a lot of students and new developers, especially from the subcontinent and Africa start contributing to open source through programs like the GSoC and the LFX mentorship programs. FOSS clubs and meetups like the ones run by FOSSUnited have proportionately more young people. It might be that there isn’t a clear path of progression from a new contributor to a maintainer. For example, contributing to open source as a newcomer is often well defined, like picking up good-first-issues or joining community calls. But more often, there isn’t enough documentation or mentorship to progress towards being a maintainer. Again, this might be better understood if there are more Indian and Asian participants in the survey.

The United Nations had a conference recently where they talked about these sustainability problems. Ruth summed up the lack of newcomers really well. Maintainers and users should definitely be wary of this and foster new maintainers. At FOSS United, we can bring maintainer panels or sessions that talks about how contributors can become maintainers (I volunteer to write more articles about this).

When it comes to governments and open source, I don’t expect governments to directly fund open source projects because it is impossible to identify which projects to sponsor and how the money should be distributed among downstream projects. Instead, governments can use and contribute to open source projects or encourage consultants to use open source projects for government-run software. A good recent example is the Kerala Government working on a new Debian-based OS for their schools.

[rahulporuri]

It is good to see you again my friend

@sudo_navendu

These countries have a significant contributor/maintainer base and thriving COSS companies. Even top projects under the Apache Foundation have large maintainers and commercial companies backing them from countries like India and China.

I don’t know if they will skew the results - I left this out of the summary but the report explicitly lists % of maintainers who are getting paid by foundations and/or getting paid to work full-time by companies e.g. COSS.

@sudo_navendu

An increasing number of maintainers work for for-profit companies with an open-core business model.

I agree with you and I don’t think they capture this information - whether the project they work on is an open-core project/COSS project

@sudo_navendu

It might be that there isn’t a clear path of progression from a new contributor to a maintainer. For example, contributing to open source as a newcomer is often well defined, like picking up good-first-issues or joining community calls. But more often, there isn’t enough documentation or mentorship to progress towards being a maintainer.

Astute observation - they do mention in the report that there needs to be a better infrastructure/resources/mentorship to onboarding regular contributors as maintainers.

@sudo_navendu

At FOSS United, we can bring maintainer panels or sessions that talks about how contributors can become maintainers (I volunteer to write more articles about this).

Funny you should say that because @Ram_Iyengar did a panel at IndiaFOSS on why contributors lose steam (and therefore don’t end up reaching a point where they can become maintainers). We definitely need to more of this - like the very first time we interacted in BLR when you moderated the panel discussion

@sudo_navendu

I don’t expect governments to directly fund open source projects because it is impossible to identify which projects to sponsor and how the money should be distributed among downstream projects. Instead, governments can use and contribute to open source projects or encourage consultants to use open source projects for government-run software. A good recent example is the Kerala Government working on a new Debian-based OS for their schools.

I think we fundamentally disagree here (and I agree mostly with @rushabh ) - i don’t think governments have the capacity to create FOSS. It might not be in the best interest of the market or the society if the government even attempts to build capacity to create FOSS.

Instead of attempting to build capacity and create FOSS, the government should create and enforce tenders that mandate the use of existing FOSS or in cases where software has to be made from scratch mandate that the final solution will be FOSS. Why not help (even subsidize) the Indian industry to build FOSS capacity, which they can then use to support orgs across the world.

Regarding funding, I agree that it is impossible for the root/upper parts of the government to identify which projects to sponsor but the government also needs to decentralize FOSS funding to the tier/orgs which have clear and direct visibility. For example, organizations at the leaf/tip of the government should have the flexibility to directly donate/allocate funds to FOSS e.g. FOSSEE, NPCI, UIDAI, significant Public Sector Undertakings (PSUs) like Public Sector Banks, ONGC, NTPC, BHEL, SAIL, etc.

I strongly believe this because of a recent interaction with people from IRISET (Indan Railways Institute for Signal Engineering and Telecommunication) - they are using FOSS projects like Asterisk and benefitting from it so they should have the freedom to also support the FOSS projects directly.

[sudo_navendu]

@rahulporuri

It is good to see you again my friend

Likewise.

@rahulporuri

I think we fundamentally disagree here (and I agree mostly with @rushabh ) - i don’t think governments have the capacity to create FOSS. It might not be in the best interest of the market or the society if the government even attempts to build capacity to create FOSS.

I mostly meant incentivizing or even mandating companies that build software for the government to use and build on top of existing open-source projects. The example I used here might be an exception and is not the right example. The Kerala Government, KITE, initially built a Malayalam Ubuntu to make computers accessible to all and not just English-medium students (back in the early 2000s). Now, Indic language OSes can be easily built on top of Debian or Ubuntu thanks to all the translation efforts by volunteers. And I agree that it is not a good idea for governments to build open source software now.

@rahulporuri

Instead of attempting to build capacity and create FOSS, the government should create and enforce tenders that mandate the use of existing FOSS or in cases where software has to be made from scratch mandate that the final solution will be FOSS. Why not help (even subsidize) the Indian industry to build FOSS capacity, which they can then use to support orgs across the world.

The new Kerala IT policy draft includes some plans to incentivize the use of open source software. We have to wait for the detailed policy document on open source for the complete plans.

@rahulporuri

For example, organizations at the leaf/tip of the government should have the flexibility to directly donate/allocate funds to FOSS e.g. FOSSEE, NPCI, UIDAI, significant Public Sector Undertakings (PSUs) like Public Sector Banks, ONGC, NTPC, BHEL, SAIL, etc.

That’s an interesting idea. But I’m still unsure how effective and fair it would be. Another approach might be to allocate developers to help maintain and contribute back to upstream projects instead of a financial contribution like how large tech companies often have employees contributing to critical dependencies. But it would also depend on the open source project.

@rahulporuri

Funny you should say that because @Ram_Iyengar did a panel at IndiaFOSS on why contributors lose steam (and therefore don’t end up reaching a point where they can become maintainers). We definitely need to more of this - like the very first time we interacted in BLR when you moderated the panel discussion

Still kicking myself because I missed IndiaFOSS. I have to catch up on YouTube.

Related to what governments should do with/for open source, a good first step might be to inner source. i.e., have reusable repositories of code that consultants can use across different projects. Might be very naive because I have no clue how governments manage software now.

[rahulporuri]

@sudo_navendu

Another approach might be to allocate developers to help maintain and contribute back to upstream projects instead of a financial contribution like how large tech companies often have employees contributing to critical dependencies

I was going to respond negatively to this but on second thought, this might be feasible. At an agency-level, there might be capacity to perform self-directed work, without the need for hand-holding or top-down task giving - which are essential to contribute to FOSS meaningfully.

@sudo_navendu

Related to what governments should do with/for open source, a good first step might be to inner source. i.e., have reusable repositories of code that consultants can use across different projects. Might be very naive because I have no clue how governments manage software now.

looking in, https://openforge.gov.in/ might be what you’re talking about - the destination for Govt-created FOSS but also inner-source projects, but from what I can see, it’s a dumpster fire.

The current Director of ICFOSS reached out to have a chat about working together. If you are interested, I will pull you into the conversation.

[sudo_navendu]

@rahulporuri

The current Director of ICFOSS reached out to have a chat about working together. If you are interested, I will pull you into the conversation.

Yes. Please do. I would be happy to help out.