Feedback wanted: what do you want from GUADEC?

Community
, ,
1

The GNOME Foundation Board of Directors is planning to revisit our future events strategy, particularly the question of where GUADEC should be held. We would like to hear input from the community before we discuss it ourselves.

Historically GUADEC was always held in Europe as a single global event. In 2022 we held our first GUADEC outside of Europe, in Mexico, and began a pattern of alternating between events held in Europe and elsewhere, with this year’s Denver event being the latest.

When GUADEC has happened outside of Europe, it has also been accompanied by the Mini GUADECs in Berlin, and an increasing hybrid event approach. We received a lot of valuable feedback on this topic after the most recent GUADEC.

The goal of making GUADEC a global event was to increase diversity and reach different audiences. However, there are other goals that need to be considered in our events strategy, including environmental impact, cost, logistics, and allowing the existing community to meet face-to-face.

Therefore, we would like to hear from everyone:

  1. What do you consider to be the main goals of GUADEC?
  2. Where would you like GUADEC to be located in the future?

We would also love to hear any personal factors which influence your preference, such as where you are located, or your experiences from previous GUADECs.

This is your opportunity to have an influence on the future of GUADEC, so please let us know your thoughts.

Many thanks!

10 Likes
2

Thanks for opening this thread Allan! I’m going to jump in and share my experience, with the strong bias that Berlin mini-GUADEC 2024 was my first in-person GUADEC.

I don’t think any metrics was set up to measure whether we reached different audiences or not. I also think “increasing diversity” is too loosely defined to be meaningful.

My experience of GUADEC is that in practice it gathers contributors to GNOME, and not that many users who are not already contributors. The event seemed to be very inward facing to me both in terms of physical attendance but also in terms of communication before and after.

I deeply care about leaving the world a better place for current and future generations. Since in the field of digital technologies, that’s where I can have an impact. To me that means leaving making sure people use technology they can trust (not just creating or contributing to it): tech that doesn’t enable surveillance, tech they can modify for their needs, and tech that doesn’t steal their attention.

GNOME could be the unifying platform for that technology. But for this to happen, GNOME needs to be

  1. A vibrant project that covers general computing needs, and that is appealing to use. There is always room for improvement, but our community has been doing a phenomenal job at it, and I would feel confident handing a GNOME computer to relatives. For it to keep happening we need
    1. Volunteers to stay engaged in the project because it matters personally to them.
    2. Vendors & consultancies to keep their expertise and find GNOME(-adjacent) business
    3. Donors (individuals and organizational) to hear about the impact GNOME is having on the world, and to partner up with people who make it.
    4. Press coverage, both from traditional outlets and independent press (such as YouTubers). I wouldn’t focus exclusively on tech, but I would also include press that cares about social and ecological issues (and how GNOME can help address them).
  2. Distribution channels to reach people. In other words: we need non experts to be able to put their hands on a GNOME machine without having to cast arcane spells, or without having to risk putting their only machine in an unusable state for them. For that to happen:
    1. We need device manufacturers to ship GNOME-based systems on their devices
    2. We need corporate orgs to use GNOME-based systems, and understand their needs (e.g. they need to be able to get paid support somehow, likely by relying on integrators). This is not for the sake of giving free labor to orgs, but for the sake of creating a system that is not just for a niche. Work takes a significant part of life. It can be an important vector to spread (and financially sustain) GNOME. This also helps gain critical mass to incentivize vendors to distribute their software on GNOME.
    3. Because of linux (let alone GNOME)'s market share, we need local communities to help each other set GNOME up and spread the word.

I see GUADEC as an event where:

  • Existing contributors can gather and discuss issues in person
  • Existing contributors can showcase their work.
  • Vendors, consultancies and integrators can talk about their needs and how they would like to scratch their itches upstream
  • New contributors can be onboarded
  • Press can come and cover the event to get the latest news about GNOME and a glimpse of the future
  • The Foundation can produce marketing material to promote GNOME and make it more appealing to all of the people and organizations mentioned above.

I don’t have a specific location in mind, but wherever minimizes CO2 emissions and visa headaches, which probably includes making remote attendance easier :slight_smile:

8 Likes
3
  1. Online hackathons. Participating to the culture where the event.
    Maybe more local things/groups - oriented.
  2. IDK, travelling cost money
4

Thanks to the Board for opening this discussion up here. I have a lot of thoughts but will try to be succinct (lol yeah right…).

Getting Face-to-Face

I’d been around FreeDesktop and GNOME-adjacent spaces for quite a while with elementary before getting much more involved; I started working on elementary around 2010, when we mostly existed as a sort of siloed project doing our own thing. We were young and hacking things together; we had thought about and tried contributing to some GNOME stuff, but were mostly unsuccessful due to our inexperience (and our desire to do things differently from GNOME).

We attended one Ubuntu Developer Summit in 2011, but I don’t recall mingling a whole lot with GNOME folks there (I was a bit overwhelmed by all of the people and simultaneous tracks, but still had a good time!). So, we just kind of kept doing our own thing—until @sri invited us to the GNOME West Coast Summit in 2015, which was a sort of GNOME/GTK/xdg-app (what would become Flatpak) hackfest.

There we finally met and hung out with people working on GNOME face-to-face, and it completely changed the trajectory of elementary—and honestly, my life—for the better. In particular, I had the opportunity to meet @sri, @matthiasc, @chergert, @cosimoc, @alexl, and a handful of others who were so incredibly welcoming to us relative outsiders to GNOME. We broke through several blockers to making elementary OS better, but more importantly, we developed lasting relationships with the people and project of GNOME.

A few years later in 2018 when I was able to attend GUADEC for the first time, I flew halfway across the world and walked into the venue, nervous and sort of uncomfortable—until @matthiasc greeted me and invited my over to a group of people chatting. We talked about elementary and GTK, and I started to get a whole lot more comfortable. I ended up having a great time that week, and felt like I really became a part of the community; I started to contribute to GNOME design discussions, attend more hackfests, regularly attend GUADEC, and stay in closer contact with GNOME folks afterwards.

When I left elementary, it was such a natural transition into the GNOME community for me, because I was already here and a part of it. So for me, these face-to-face experiences are monumentally important—not just for the increased bandwidth of communication and ability to work through things so much better in person (though that’s a huge benefit!), but for the humanization of the people behind usernames—and the deep, lasting connections that are made.

I genuinely don’t know if I’d be involved with the GNOME community today had I not been able to attend the in-person events I did over a span of years.

Local Community

I’ve been very impressed with the efforts of a handful of folks (hi, Berlin!) to build and nurture a local community. I’m honestly a bit jealous, because I’ve not had the time to do that nearer to me. Maybe if I had that sort of local community supporting me, I’d feel similarly connected and not feel the need for larger in-person events—but that’s not the case for me where I am today, and definitely wasn’t the case for me in 2010 or 2015 or 2018. So the global event with the opportunity to meet familiar and new faces from all across the GNOME community still seems critical to me personally.

Remote/Hybrid Events

When I was at elementary, I organized and helped host a two-day all-remote developer conference. I learned a lot there, and it really made me realize a well-done remote conference can work as an event—but it’s not a complete replacement for at least occasional centralized, in-person events. There was good sharing of information and some nice follow-up energy from people, but there was none of the high-bandwidth communication and none of the humanizing social bits.

When helping organize and run this year’s GUADEC in Denver, I felt like we did a decent job of enabling remote participation, but there was some friction with having two “separate” events between Denver and Berlin—and I feel like I really missed out collaborating and just hanging out with people who I’d been looking forward to see because they were halfway around the world.

Climate

The climate impact of traveling has come up a lot in the discussions about how to do GUADEC, but I continue to be reminded of @pwithnall’s analysis of GNOME’s climate impact where conference travel has less carbon impact than the usage of our software. That’s not to dismiss the carbon impact of travel, but to contextualize it; I do think it is necessary as humans from around the world building things together to get together in person every now and then to do it effectively—including to bring new people into the community, and to actively work on reducing the power consumption of that software that goes out to millions of people’s power-hungry computers.

My Proposal

I don’t think we necessarily need to have GUADEC as an annual thing where everyone gets together all in one place—it’s pretty clear that’s not going to happen anyway, as we would probably end up having a large European event every other year where everyone in Europe and whoever can make it from the Americas will attend, and then an event outside of Europe the other half of the time where the Europeans all meet up in Berlin. :wink:

If that’s the reality, I feel like we could take a multi-pronged approach:

  1. A less-frequent but centralized/in-person “flagship” event. Call it GNOME Conference or something. I don’t really care where this is held personally, if we are able to have some amount of travel sponsorship to ensure people who would not otherwise be able to attend can also join. This could be every two years, maybe, ensuring a regular cadence but literally halving the climate impact. There should also be an explicit focus, as a policy of the event organization itself, to encourage the least climate-intense options for things like travel, local transit, lodging, and food whenever possible.

    Remote participation for talks should be possible and encouraged for those who cannot otherwise make it in person, but the focus should be on getting people together face-to-face and optimizing for that (similar to GUADEC in Denver this year).

  2. More frequent local events, with more documentation and support from people who have organized these successfully! Basically, let’s work to develop something like a Local Community Kit where we lay out tips and tricks, provide resources like editable designs for signage, etc. that local organizers can take and use to make a successful event. These would not need to be centrally organized or even regularly scheduled, and as we’ve seen with events in Berlin (and I’m sure elsewhere), these don’t even have to explicitly be only GNOME-focused; pairing with another open source community or a slightly wider scope of interest can encourage cross-pollination of ideas and more local community development.

  3. Regular completely remote events, let’s say an annual GNOME Online Summit more like edw where it’s designed to ensure contributors are sharing their work and collaborating around shared topics, but it’s accessible to anyone with an Internet connection. This can coincide with local events as well, of course, as another way to encourage local community building.

I have no idea how practical that all is, but I think it could be a bit of a restructuring that balances climate impact, travel costs, remote participation, and local community development. And dropping the GUADEC name could signify that it’s not just “changing how we do GUADEC,” but that we’re reworking how the GNOME community does events more broadly for the better.

And lastly, we shouldn’t be afraid to iterate! While it’s nice to try to get into a rhythm, we should also be open to trying things and gathering feedback from the community, then tweaking as needed. We don’t need to have a single “correct” solution for the rest of eternity on our first try. :slight_smile:

7 Likes
5

Hello, interesting question, let me try and be even more succinct :slight_smile:

What do you consider to be the main goals of GUADEC?

People who work together online can spend time together in-person.

People who have similar interests and goals can meet each other and form new partnerships

Celebrate the GNOME project’s successes and contributors.

Where would you like GUADEC to be located in the future?

I’m happy for it to move around, but it’s nice if it could be in Europe at least once every 2 years.

We would also love to hear any personal factors which influence your preference.

I don’t have a local GNOME community on my doorstep, the nearest active contributors that I know live 3-4 hours drive away.

This is your opportunity to have an influence on the future of GUADEC.

After GUADEC 2024 I feel that “hybrid” events don’t achieve what we need from an “all hands” community event. Timezone differences exist, online social events are very limited compared to face-to-face ones, and effectively the community splits according to geographical location. As long as we can have an in-person meetup I think it’s valuable to do so.

Hybrid events (i.e. multiple meetups happening on the same date in different parts of the world) are super interesting and scale much better than an in-person meetup. So I would invest more effort in building our capability to organise hybrid events, where we can experiment and try different approaches. Some ideas:

  • a 2-3 day event where we organise many hackfests in different places, where each local group agrees their own goals for their event (with a thought to “marketing” what each group achieved after the event, as well)
  • a new “online-first” conference, where all presentations and discussions happen online, but local “viewing” parties take place where possible

The key being to explore these ideas in parallel to our existing in-person event. (So we might do them in november/february/march, away from when GUADEC currently happens/

a side note on the climate crisis… i’m acutely aware of how damaging the airline industry is to our future; at the same time, what percentage of annual airline ticket sales are for travel to GNOME events, and will anyone notice outside GNOME if we reduce this small number to a smaller number? positive actions have more resonance than negative actions – creating new “local-first” events is a positive action, while reducing the scale of our existing event is a negative action; so to “lead by example” we will do better to focus on building new events, not adapting existing ones

6 Likes
6

Thank you for these questions! :slight_smile:

  1. What do you consider to be the main goals of GUADEC?
  • Grow GNOME’s community of contributors, for example by helping people to feel welcome in the community.
  • Develop consensus around the GNOME project: plans for future technical and non-technical work. (It is very important to me that these have some overlap).
  • Build strong cross-functional relationships within the project, separate from existing technical and geographic circles.
  1. Where would you like GUADEC to be located in the future?

In a perfect world where resources exist to do it, I think the calculus could include “where do we have the most activity from new / likely potential contributors, and where (geographically) would we like to see our community grow to improve its diversity?”

With that in mind, I think the recent pattern of alternating between Europe and North America has been a good start, and I think beyond “where” we need to be asking “when”. I think it would be worth trying a large scale event like GUADEC less frequently, such as every two years. This would free up resources to run other, less expensive events with different goals, and allow that larger event to focus more on its own goals.

6 Likes
7

Typically, I have considered Guadec to be a catalyst. After I come home from one, I have run out of questions to make, and my head is stuffed with new ideas, better ideas, better plans to tackle old ideas, and generally a much sharper view on many areas of the project. I’m sure many others find it as energizing as I do.

But I said typically, since I have not found the same amount of enjoyment from full-remote or split Guadecs than I get from “proper” ones. I find the high bandwidth to come mainly from being able to spend time together with other people for an amount of days, talking with the people you typically work with, corridor mingling, … This may sort of work for some with split guadecs, but the isolated groups (and timezones) don’t allow for much inter-group interaction.

And Guadec talks are nice, but it’s not where this high bandwidth comes from.

I think it is great that local communities gather and enjoy Guadec remotely, and we’ve shown that we can decently do remote talks to fuel remote events, but we can create brand new events that don’t necessarily overlap with Guadec.

I think it is nice to attain ourselves to a low carbon footprint. Perhaps at the expense of less represented or most remote countries, a deciding factor to attend to could be contributor demographics, i.e. choosing places that allow most contributors to travel the least.

But while having some established goals about (low) carbon footprints is amazing, I think there are possibilities for creating a net benefit (e.g. performance hackfests, improving the churning of our CI, or just the general betterment of the project as it happens at Guadec) that we might be exchanging for a nominal carbon footprint that will turn to smoke anyways. We now know how low can we get, and the impact it has globally and to the project, there’s perhaps some wiggle room to set up some expectations/goals that we feel comfortable with for a Guadec-sized event.

5 Likes
8

Audience:

I think we should have some clarity on who are the target audiences of GUADEC. It’s obviously open to all, but still we should have some clarity here, as I find this to be not clear. For example, here are 2 versions in our pages.

  1. GUADEC is GNOME’s main conference, which has been held annually since 2000. It is a contributor-focused event where members of the project get together to share what they have been working on and make plans for the future. (Source: https://handbook.gnome.org/events/guadec.html)

  2. GUADEC is the GNOME community’s largest conference, bringing together hundreds of users, contributors, community members, and enthusiastic supporters for a week of talks and workshops. (Source: https://events.gnome.org/event/209/)

Currently, I find it more of a “I’m a contributor and I’d like to meet other contributors / core devs in person” for high bandwidth information flow to be the main theme here, which sounds like [1], and is perfectly fine, though I think this limits the scope of GUADEC.

If [2] is more accurate, then we should have events in GUADEC, which should cater to all sections of the audience. (e.g. puzzles, quiz, find-the-bug-in-this-code bounty and gifts / goodies for winners / contributors etc).

Also, a gentle reminder that we should also look into the data from GUADEC '24 feedback form which was submitted a while back.

Location:

I think ideally we should switch between new locations & locations which are known to attract maximum participation every alternate year. But, this is also tied to our budget constraints, so might not be as simple as it sounds.

Environment:

I think if we’re restricting air travel due to budget constraints, it’s perfectly reasonable. But, we shouldn’t be discussing emissions and environment impact from air travel for the current scale and frequency of GUADEC. It probably makes more sense for an event like FOSDEM which attracts thousands of participants.

That’s not to say we should not be concerned about the environment. We should focus on things which impact the environment on a daily basis (than annual events) like:

  • Providing better and advanced power management features and support in GNOME Desktop.
  • Track and address performance issues in GNOME as priority.

which are well within our realm of possibility, and which will also present GNOME as an environmental friendly / efficient desktop. This would also promote GNOME in better light.

Thanks!

3 Likes
9

I consider GUADEC and its ilk to be an incubator of ideas and projects. Every time I’ve gone to GUADEC - I kicked off some kind of project and they continue today.

But what I’m thinking is though is that we need to generally figure out how to deal with north and south america. North America tends to be very expensive and difficult to find venues and never mind the costs of travel to get there.

Europe has been mostly efficient in this way because we have the highest concentration of core developers in Europe.

My proposal is that we keep it in Europe for the time being because it moves the project forward, but we should look to figuring out how to diversify our core contributors.

As, Cassidy’s experience dictates, reaching out to the other GTK based desktops and make it welcoming for them will also be good. We’ve seen the good that happens with Linux App Summit and its predecessor West Coast Hackfest. They do change the trajectory of other projects. Better partnerships with downstream desktop projects will only help.

More regional conferences as we have been doing in Africa and India but also Brazil and Argentina. Flathub data shows that there is a lot of consumption of Linux apps. We should start building outreach to govt, private institutions and businesses. The nature of the conferences could be more oriented towards third party developers and businesses. We’ll be even make better arguments when Flathub allows monetary exchanges.

For non-GUADEC conferences, I wonder if we can find good ways to pass feedback to core developers. These conferences should host UX designer feedbacks, or designer workshops complete with certificates. (do not underestimate the power of certificates, I have the receipts)

As for carbon savings and so forth. I’m not sure that this works in practice. To gain the kind of carbon savings you want to do on travel requires govt policy. If 6 developers decide not to travel, there are 6 people who are willing to travel and that plane is going regardless. Instead as a foundation we could pushes for trains for regional travel and stop short flights. It seems dumb to have a short trip from Portland to Seattle unless you’re going on a multi-leg journey when a train will be fine for most things.

We could figure out carbon savings for CI/CD pipelines, that’s where we should be targeting - how many CI/CD runs equals one plane flight in carbon costs?

My two rupees.

5 Likes
10

These are all great thoughts. Thanks to everyone who has taken the time to respond so far.

My personal perspective is that there are a number of important “soft” aspects of having a single global face-to-face event, including:

  • integrating and motivating new contributors
  • reaffirming our community, through shared a shared sense of belonging, values, and aspirations
  • personal relationship building (which in turn improves working relationships)
  • sharing of ideas and developing common goals

For me these are some of the most important functions of the event, and are what I’d personally prioritise. I don’t think they can be done nearly as effectively remotely, compared to in-person.

I do feel that we have missed something important as a result of not having a global face-to-face GUADEC in recent years. It’s easy to forget but, between 2000 and 2019, the GNOME project had a single global conference every single year. In comparison, in the past 5 years it has happened just once. My sense is that this has had significant negative impacts for the project.

But I also know that there are other sides to the story, and there are no easy answers to the location question. I’m interested to know:

  • How many people who attended the events in Denver and Guadalajara would not have gone to a GUADEC in Europe.
  • What GUADEC’s CO2 impact has looked like for the past 4 events (me looks at gnome-environmental-analysis).
  • What other open source projects have tried for their events strategy, and what the results have been.
5 Likes
11

As someone who went to conferences 10 years ago but mostly doesn’t anymore I have a slightly different perspective than many others who have replied.

Inclusion and “Diversity”

The notion that central and / or in-person events are somehow able to include everyone or at least be open to all contributors is still very popular. However, this is simply not that case. The truth is, an in-person event, decentralized or not is better for some people in the community, while it is worse / impossible for others:

  • People who can’t afford travel / accommodations (this is partly solved by sponsorships, although never completely, as you also potentially need to take time off, pay for things that are not covered etc.)
  • Many disabled people who can’t (easily) travel, or for whose the effort is too much, or the accommodations not enough
  • People with health issues, especially immunocompromised people for whom exposure to many / unknown people can be very risky
  • People who can’t easily travel to the country in question because of difficult / impossible visa requirements
  • People who can’t travel at all due to the risk of never being let in again into their country at all
  • People who live far away from the places where the event is mostly held
  • LGBTQIA people, especially those who live in places where their mere existence is in danger at the place where the event is taking place (such as the US for that matter)
  • Single parents and caregivers

This list is not exhaustive, but I think it brings the point across: The people that are mostly affected by these issues are also often generally more marginalized people, which leads me to be of the opinion that relying mostly on central in-person events for social exchange has a net negative impact on “diversity”, and might even cause these people to feel left out and / or even feel alienated by the expectations that the only way to form meaningful connections with people is via these in-person events.

What to do about it?

In my opinion, in-person events can and probably should still happen, but they should not be called GUADEC, and should not have the scope of GUADEC. GUADEC should be the word reserved to an event where as many people as possible are invited, to broaden everyone’s horizon, and make as many people as possible be included in the “main community event”.

I think there should be three things happening:

Forming more online social spaces

There are already channels on Matrix that at least have a social component, but there is probably more that can be done to make more people feel included in a social structure.

Local / Hybrid events

Smaller “Local” / hybrid events throughout the year. This is already happening, but maybe there could be things to improve here to make it easier for anyone to do these kind of events in their local area. Maybe also have some sort of centralized calendar, where these events can be found :slight_smile:. We should also be collecting experiences from the people who organized such events in the past in a central place as to make it easier to organize them.

GUADEC

This has always been considered the “main” event. Being a main event means it has the responsibility of having to cater to the needs of as many people as possible. As elaborated above this is simply impossible for a centralized in-person venue based event.

There is a need to also meet in-person for some people. So we probably need some sort of satellite events too. Those in-person events should however not compromise the decentralized / online experience. Often hybrid events have the issue that following online is a much worse experience than if these events were online-only. This is especially because of the audio situation being sometimes so bad that it’s impossible / very draining to follow. FWIW the Berlin event this time was much better at this than Denver. Maybe there could be some exchange in that matter :slight_smile:

Also accommodations like subtitles should be something to have again - we had those once but don’t anymore.

However, these in-person events should in my opinion strictly be labelled “satellite” events, and one event should not be prioritized more than the other or labelled as the “real” event. Having a “main” in-person event just creates expectations (“everyone” is there) that are impossible to follow through with, and makes many people unhappy if the event does not live up to those.

8 Likes
12

+1

I wanted to attend an in-person event for years and it was always impossible

  • in 2017 I had no way to get a passport
  • in 2018 it was very dangerous for me to travel at all
  • in 2019 I was waiting for the passport
  • in 2020 I had the passport but lockdowns
  • in 2021 ditto
  • in 2022 I had just moved to another country and was exhausted from that
  • in 2023 I would be denied visa due to the actions of the country my passport is from
  • in 2024 - I don’t have energy for travel like this anymore

In all of these cases it would be very difficult (before 2022) and likely impossible (2022 and later) for me to get a visa, and even when it is possible, it takes a very long time and the process is very exhausting (and expensive).

So there’s no chance I would attend any physical event at all. Remote events would be possible at least, tho online guadecs were pretty disappointing. Almost everyone I know who attends in-person events says they don’t care that much about talks and it’s mostly about the social stuff - meanwhile online events did not include basically any social spaces (and those that did exist were widely ignored) and instead focused on talks - and ofc remote talks are basically just blog posts read out loud with crappy audio and video quality since nobody has decent mics or camera (me included if I were to do a talk), and generally would work a lot better as, well, blog posts. There were “social events”, but they boiled down to non-technical talks, more or less.

I agree with Fina it feels rather exclusionary. I don’t know what the solution is, but as is I don’t really think it works at all.

5 Likes
13

Super valuable comments, @alicem and @fina . Thank you.

If we were to run an annual online-only event, would that be something that would interest you? There are interesting event formats we could try. For example, we could do an online unconference (no talks, self-organised breakout sessions that people sign up to on the day, and so on).

3 Likes
14

Thanks for bringing this up Allan, and sorry for taking so long to reply. It’s interesting to see what other people think, and I find myself agreeing with a lot of what others have said. Sorry for the long reply, I have a lot of thoughts about this. Hopefully they’re at least slightly coherent.

What do you consider to be the main goals of GUADEC? I think that having some annual largeish (bigger than a hackfest) conference is useful to, in no particular order:

  • Showcase what the GNOME community has been doing recently, to an external audience
  • Showcase what people are currently working on, to other developers, to create excitement and get input and contributions from others
  • Provide an immersive and social way to introduce new-ish members of the community to GNOME development and let them get ideas about what they might like to contribute to
  • Provide a synchronisation point in the release cycle — a soft deadline which people can aim for (“I need to finish this feature so I can show it off at GUADEC”) and build on (“we discussed foobar at GUADEC and now have 3 months to implement it before freezes”)
  • Let people have general hallway conversations about project direction, features, etc. which then result in hackfests where the actual work gets done
  • Allow people to meet those they collaborate with and socialise
  • A holiday! Going to GUADEC each year has made me visit places I wouldn’t have thought of going to, tack a holiday on the end of the conference, and experience that culture. It’s also enforced time away from my laptop — even at the conference, focused hacking time is limited and more time is spent on your feet. That kind of break is good (for me, at least).

Where would you like GUADEC to be located in the future? Firstly, I don’t think my answer is specific to ‘GUADEC’. We happen to call our big annual conference ‘GUADEC’ at the moment, but this answer applies to any big annual conference. We could rename things if we significantly change how they work, and that would be fine.

I think the strategy of alternating between continents and providing increasingly good support for remote attendance and presenting, coupled with official satellite events, has worked fairly well in recent years. I’d be happy to see that continue. I’d also be happy to try out other approaches, such as a fully decentralised event where there are multiple satellite meetups but no single ‘main’ location. Or perhaps a shift in focus away from one big annual event to perhaps having more regular, local hackfests and one big bi-annual event. As others have said, we can experiment and iterate with different approaches.

Analysis. From the analyses I’ve done of GUADECs since 2019, a few things pop out to me:

  • Travel emissions are the biggest emissions from hosting a conference, by about 2 orders of magnitude (over the server emissions, food, merchandise, etc.).
  • Having a satellite event in Europe when the conference has been in the Americas has measurably reduced emissions.
  • People consistently voice support for satellite events and improved remote participation in survey feedback.
  • There’s been an order of magnitude more people attending GUADEC remotely (via YouTube live streams or BigBlueButton) in the last couple of years than have attended in person (but our in-person attendance figures haven’t really gone down). i.e. Remote attendance brings a much larger audience. As others have said above, it also importantly removes barriers to attendance.
  • That audience is largely read-only though. People have consistently asked for better ways to participate remotely in survey responses, but have very few suggestions.
  • Anecdotally, people ask for the ‘main’ conference to be hosted in places which are more easily accessible by low-carbon transport in survey responses, but I haven’t done a sensitivity study to see how that would impact overall travel emissions. I don’t think we have enough data resolution on people’s travel to be able to do that.

There have been some suggestions in this thread that we shouldn’t be focusing on emissions from GUADEC, because the emissions from people using GNOME on their computers are much higher, so we should focus on that instead. There is some truth in that, but I think it’s a false dichotomy — we can reduce both. If we say that we don’t care about emissions from GUADEC without coming up with a resourced, measurable plan for how to reduce emissions from people’s computers, we’ve achieved no reductions at all.

So one outcome of all this could be for the Foundation to say “we can’t see a great way to improve the status quo of GUADEC emissions, but we are going to put some money into contracting people to monitor and reduce desktop power usage for each of the next few GNOME releases”. That would result in a good environmental impact.

Looking at it from another angle, the average per-person emissions from someone attending GUADEC 2023 (Riga) in person was roughly 1.2tCO2e. A rough current personal carbon budget is 3–4tCO2e (basically the total world emissions left in order to limit global warming to 1.5C, divided by the global population and fitted to an emissions reduction path to 2050). Are we important enough as a project that we require people to use on average 30–40% of their annual carbon budget to attend in person? So I think it’s right we should be looking at ways to make the conference happen in some form while lowering this number.

As others have said, leading by example and building new events might be a more positive way to go about this (with lower perceived costs of failure as well).

Socialising. I find I actually get more and better socialising done with people at hackfests than at GUADEC, although both are good, and this is very personal to how I interact with people. Others will have their own way of getting the most out of GUADEC.

If there was less of a focus on a big annual conference, I might have more of an incentive/need to try and make something happen locally. It’s worked well in Berlin. It might not work well in the north west of England, but might be interesting to try, and we might get more contributors out of a more vibrant local scene.

1 Like
15

I actually have data on this, and CI/CD pipelines are not the target you’re looking for (sorry).

From the analysis I’ve done, the CI pipelines for GLib use about 30kWh/month. At 200gCO2e/kWh, that’s emissions of roughly 70kgCO2e/year.

A single short-haul flight (e.g. from London to Berlin, return) is about 340kgCO2e. GLib is one of the bigger CI users in GNOME, so my guess is that one person’s short-haul flight to GUADEC emits as much as all of our CI pipelines across GNOME for the year.

Reducing CI pipeline emissions is good (and my blog post has suggestions for how to do that, if anyone reading this is interested), but mostly from the point of view of freeing up shared runner resources and reducing development round trip time. It’s far off a golden bullet for reducing project emissions.

1 Like
16

Thanks for this, @pwithnall . I really appreciate the effort put into data analysis and have found your work extremely useful.

A couple of points that I think it’s worth clarifying:

In the post-conference survey of this year’s mini GUADEC attendees, only 3 out of 24 respondents (12%) indicated that they would have gone to Denver if the Berlin event hadn’t happened. That survey also showed that 6 out of 24 respondents flew to Berlin.

In my mind, this makes it a bit debatable how much the Berlin event reduced CO2 emissions.

This isn’t to say that the Berlin event isn’t beneficial of course. You certainly can say that the satellite event allowed more people to meet face-to-face than would have otherwise been able to.

With my social research hat on, I think we need to be careful about how we frame findings like this, and I’d be interested in knowing which data you’re drawing on here. If you ask someone “how can we make our conference more environmentally friendly” and they respond “better satellite events” that doesn’t necessarily mean that the respondent would prefer to go to a satellite event or even that they think satellite events are the best path forward for our conferences.

17

You make a fair point about the rebound effect of people attending mini-GUADEC who wouldn’t otherwise attend a main GUADEC. I think that’s probably a good thing (wider participation), but you are right to say it’s not necessarily reducing per-person or overall emissions compared to the counterfactual of there only being a Denver GUADEC.

This is indeed from looking at the responses to the “Do you have any suggestions for how to make the conference more environmentally friendly?” question over several years.* You are right that it’s hard to read too much into those question responses, since they’re highly depend on how the question is asked, and often the responses are very brief. I think my summary is accurate, though. The responses voiced support for satellite events. They didn’t say that satellite events were the best/worst/any specific kind of path forward.

* The actual responses are private, but a summary of them is available as the “Do you have any suggestions for how to make the conference more environmentally friendly” sheet in post-conference-survey-summary.ods for each year. As a board director you can probably have access to the raw data if you’d like to double-check I’ve summarised things correctly. You’re more qualified than me for that!