Code for Earth - and what's in for you
This is a modal window.
The media could not be loaded, either because the server or network failed or because the format is not supported.
Formal Metadata
| Title |
| |
| Title of Series | ||
| Number of Parts | 156 | |
| Author | ||
| Contributors | ||
| License | CC Attribution 3.0 Unported: You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal purpose as long as the work is attributed to the author in the manner specified by the author or licensor. | |
| Identifiers | 10.5446/68443 (DOI) | |
| Publisher | ||
| Release Date | ||
| Language |
Content Metadata
| Subject Area | ||
| Genre | ||
| Abstract |
| |
| Keywords |
FOSS4G Europe 2024 Tartu25 / 156
6
33
35
53
55
59
61
67
70
87
97
99
102
103
104
105
107
111
121
122
123
124
125
126
127
128
134
144
150
151
155
00:00
Streaming mediaCoding theoryMachine codeCodeComputer configurationPresentation of a groupVideo gameGoodness of fitBitSelf-organizationLecture/ConferenceComputer animation
00:40
FreewareOpen setStandard deviationOpen sourceRange (statistics)Machine codeOpen setOpen sourceFreewareStandard deviationSoftwareRange (statistics)Computer animation
01:44
Service (economics)PrototypeDigital signalTwin primeData managementSelf-organizationMathematicsState of matterStaff (military)Numerical analysisPredictionEndliche ModelltheorieInformationCollaborationismComponent-based software engineeringPropositional formulaRange (statistics)Green's functionFood energyGroup actionGoodness of fitState of matterIntegrated development environmentState observerSelf-organizationService (economics)Observational studyNumeral (linguistics)Context awarenessProjective planeConnected spaceEndliche ModelltheorieComputer animation
04:00
CodeOpen setCollaborationismMachine codeGroup actionProjective planeGoodness of fitKey (cryptography)CodeComputer animation
04:29
Machine codeComputer programGroup actionComputerSoftwareOpen sourceVirtual machineStatement (computer science)CodeComputer programExpert systemBitContext awarenessGroup actionComputer scienceProjective planeSoftware developerPhase transitionOpen sourceReal numberKey (cryptography)Machine codeNeuroinformatikSoftwareComputer animation
06:22
Phase transitionSystem identificationStaff (military)Self-organizationOpen setComputer networkProcess (computing)System callNumberPhase transitionProjective planeComputer animation
06:57
Self-organizationStaff (military)Computer networkOpen setProcess (computing)Phase transitionSystem callMIDISoftwareSelectivity (electronic)Phase transitionMultiplication signIntegrated development environmentInformationStatement (computer science)Repository (publishing)Computer animation
09:20
Cone penetration testSoftwareLevel (video gaming)Machine codeSoftware developerComputerProjective planeSoftware developerSystem callStudent's t-testNeuroinformatikTouch typingHorizonComputer programLevel (video gaming)Computer animation
10:15
Machine codeTerm (mathematics)Open sourceCoding theoryPhase transitionCodeComplete metric spaceEvent horizonMachine codeGoodness of fitCodeProjective planeMultiplication signSoftwareLattice (order)Phase transitionEvent horizonPresentation of a groupGroup actionReading (process)Software developerCollaborationismOpen sourceResultantComputer animation
11:44
WebsiteImage registrationMachine codeTerm (mathematics)WebsiteResultantProjective plane5 (number)BitComputer animation
12:19
Machine codeProjective planeMachine codeImage registrationFile archiverCodeWebsiteWeb pageComputer animation
12:49
Web pageMachine codeComputer programSoftware developerComputerSoftwareOpen sourceGroup actionRange (statistics)FAQPhase transitionHTTP cookieDecision tree learningSoftware frameworkVirtual machineMachine learningFile archiverPresentation of a groupProjective planeInformationLink (knot theory)YouTubeDescriptive statisticsWebsiteComputer animation
13:31
Machine codePhase transitionMenu (computing)Machine codeComputer programPhase transitionProjective planeLevel (video gaming)Descriptive statisticsInformationRepository (publishing)DialectSystem callLaptopArithmetic progressionDifferent (Kate Ryan album)Constructor (object-oriented programming)CodeData Encryption StandardSoftwarePresentation of a groupTelecommunicationSocial classComputer animation
17:08
Projective planeVirtual machineSocial classComputer animation
17:28
Information retrievalMathematical optimizationProjective planeFood energyService (economics)Special unitary groupAreaMereologyGreen's functionConditional-access moduleVisualization (computer graphics)Computer animation
18:28
CodeTerm (mathematics)Machine codeWebsiteImage registrationGoodness of fitMachine codeSystem callComputer programMusical ensembleCodeComputer animation
19:00
World Wide Web ConsortiumVirtual machineProjective planeManufacturing execution systemCodeDialectBuildingSoftwareResultantEvent horizonMereologyMachine codeProduct (business)Operator (mathematics)PrototypeArithmetic progressionService (economics)Self-organizationLecture/Conference
23:51
Computer-assisted translationPlastikkarteComputer animation
Transcript: English(auto-generated)
00:00
Okay, Code for Earth. What is Code for Earth? Have you heard about it? I don't know. I will explain a little bit what it is, and I think there are many options for you to participate. And this presentation has been
00:20
came to life with the support of my dear colleague Esperanza Quatero, and then Tian-Zhen, who is here in the room again, one of my famous colleagues from ECMWF. Good. So, Code for Earth is supported and we have contribution by all the organizations, as you see the logos.
00:41
So, I go who I am. Most of you, or many of you, have probably seen the keynote on Wednesdays. I will not repeat. Ah, my other colleagues are also here. Hey, so I'm, I've been working in the GIS community for many, many years.
01:01
Various companies in Germany and Bonn. I'm half German, half Greek. You might wonder about the name. And then I've been over 15 years with OTC. Like two years ago, I decided to take a couple of months of sabbatical, and since one and a half years
01:22
I'm working now with the ECMWF, the European Center for Medium-Range Weather Forecasts. I'm a supporter of free and open source software, open standards and open data, and I've been, or I'm a charter member of OSG since 2008 and have been around pretty long in the community.
01:46
So, what is the European Center for Medium-Range Weather Forecasts? Who knows the ECMWF? Well, nice. So you, probably you learned something on Wednesday. That's, that's cool. So, we are a
02:04
member organization that has been established nearly 50 years ago. We have a 24-7 operational service where we do numerical weather predictions, and we're supporting our member states and cooperating states as well as businesses. We are also a research organization
02:25
where we experiment to continuously improve our models, and we do reforecasts and climate reanalysis, and additionally we operate two EU Copernicus services. That's the Copernicus
02:40
Atmospheric Monitoring Service and Copernicus Climate Change Service, and we support the Copernicus Emergency Management Service. And yeah, additionally, and it has been mentioned before, we are also involved in Destination Earth.
03:02
I would like to have a short shout out because in the context of Copernicus, we also have two new pilot initiatives. That's the Health Hub and the Energy Hub, and we are looking into Copernicus. The Health Hub is about environmental data and Earth observation data at the services of
03:23
the general public health, and the Energy Hub is about connecting environmental data and Earth observation to the green economy or green energy transition. And if you have
03:40
experiences with that, and if you have projects that relate Copernicus data to health or energy, please feel free to contact us, either directly my colleague Julie Letatre, or you can of course talk to me or to my colleagues as well. Good.
04:02
But we're here to hear about Code for Earth, and who of you have heard about Code for Earth before? Okay, good. So, this is the end. I will talk about what is Code for Earth, how does it work,
04:22
how you can get involved, and then I will showcase a few examples what the project has been doing. So, Code for Earth is a key innovation action run by the ECMWF and has been supported by Copernicus, Destination Earth, the Helmholtz Centrum
04:41
Herion in Gestach, close to Hamburg, and IFAB, an Italian-based foundation which is dealing a lot with machine learning. So, we want to enhance the visibility of what we do and what also what our partners do, and in the context of ECMWF, it's of course also, you know, we want to promote
05:01
the access of our data. Just think about a small research program or kind of a hackathon an extended hackathon, a small research program. We work with external talents, we call them young talents, that work over a couple of months with mentors and experts, and
05:25
here again, it's about working together, getting an understanding of what can be done, and solving mostly real world challenges, real world scenarios, and also make an impact in earth sciences, computer science, and applied data science, and of course
05:46
open-source software development. The projects that successfully end or finish their programming phase,
06:02
complete the project, they receive a 5,000 euro stipend. The program has been around since 2018 and was called before ESOC, like ECMWF Summer of Weather Code, so you might also, it was related a little bit also to Google Summer Code.
06:23
So, looking into detail, as I said, as we have since 2018, we have been running since 2018, we have been working, or we have supported nearly 60 projects,
06:46
and that have been mentored by 107 mentors, and we have an equally high number of participants. So, this is the timeline more or less, and I will go in the various phases. So, we usually start, or
07:01
now, and this goes until, let's say, mid of January, I walk around and talk with my colleagues and ask them if they have any ideas, or any problem statements, any challenges, things they always wanted to do, but never had the time, that can be dealt with in about four months.
07:22
And we call them challenges. So, I identify challenges, and meanwhile, we are working also with external partner networks. So, as I mentioned, and you will see this in more detail, so this is, we look for challenges with an innovative character, and then,
07:46
once I have collected all these challenges, we go in the next phase. But, as I mentioned, this year, we have worked with, or this edition, we have worked with partners, and
08:02
so, we have been working, we identified joint challenges with the European Environment Agency, with CSOC and the University of Bonn, University of Reading, and, as mentioned also before, Herion, Helmholtz, Zentrum Herion, and Germany. So, once the challenges have been collected,
08:24
my colleague Esperanza and I, we are preparing a call for participation, that means all the challenges go on GitHub, and I will show you later how, where you can find the information, and we issue the call for participation, usually beginning of
08:41
March, and we offer two Q&A webinars, where interested people can ask questions, we explain how it works, and people can browse through the GitHub repositories, get in contact also with the mentors, and see if they have a solution for a challenge.
09:04
After the call for participation ends, we have a selection phase, this is when the mentors, or the submitters of the challenges, they review the proposals for that challenge, and then we select, per challenge,
09:21
one proposal. So, who can participate in the call for participation? Well, the program is open to everyone who is interested, that can be an individual, we usually have teams of that submit a proposal, and they come from various backgrounds, that could be, you know, earth sciences, computer sciences, and software development.
09:45
It's really, it's really broadly, the horizon of the project is quite big. Then it's open for students at all levels, if you are a startup and want to work with ECMWF data, you can check if this is of interest for you, but we had also have, you know,
10:05
bigger companies, they wanted to get in touch with ECMWF and work with that, so literally everyone can submit a proposal. Once we have selected the teams, the coding phase starts, and that's usually in May,
10:24
so this is the time when it's about innovation, collaboration, and the development of open source software. We have a four month coding phase, that goes from May through August, and we had a fantastic kickoff meeting this year with many, with all the projects, the mentors, and the participants, and
10:45
what I would like to is the, yep, midterm webinars, but I have another slide. It's not so easy. Good, so the coding phase in 2024 started with 13 projects, and we covered 12 challenges.
11:06
We have 52 participants, and they are mentored by 40 mentors, which is a huge group to handle, but it's really, we're really excited because it's nice to see how the teams are working.
11:20
So, once the coding phase is finished, we, and if the successfully finished projects, we invite them to a final Code for Earth Day. That will happen in person for the invited people in Reading, and for the general public, the team, the event is
11:40
online available, so you can follow the presentations of the results. So these are the projects of the 24 edition. And before I go into, to GitHub, in our website,
12:00
please go on our website because there you can register for the midterm webinars. They will take place on 17th and 25th, 5th of July, so that's in two weeks. One and a half hours, the team will present, the teams will present what they have done so far. But let's look a little bit into the
12:22
projects. Yeah, so this is the Code for Earth website, and here again, if you, if you go on the website, you will find immediately the registration page, and you can have again, you know, an explanation, you can read how it works, and
12:45
what is interesting as well, you know, if you can go to the archive site, no, sorry, to the archive site, and you find all all the projects from previous editions. So if you, if you would like to see what, what have, what the projects have been doing, it's,
13:08
you know, last year we had, for example, fire forecasting. This links you to a YouTube presentation that is from the final day, and you have a project description, if you click here,
13:21
the project is described, surprise, surprise, and you can find also all the information on GitHub. So we have an ECMWF Code for Earth GitHub repository, and there you can browse all the challenges of previous years, and
13:44
let's look at at the 24th edition, you know, you have here the project descriptions, what the projects have been doing, and for example, the tales of dry land.
14:04
Here you have the description. This is how, do you remember, I talked about the call for participation. This is what, what people can read, you know, it's a description of the goal, mentors and skill, the challenge description, and
14:21
during the call for participation, people are interested, people can ask questions here, and you see there, there is communication taking place, and so if I go back, so that is, you can find all these descriptions for 2024, and also going back
14:42
on our website, like this is interesting as well, that's a regional to urban air quality mapper, and they are looking into downscaling regional pollutant data to urban levels, visualizing accurate air quality insights.
15:01
But it's not only the description that you can find on GitHub, but also there are repositories of those projects that actually developed software and actually did something. again, you know, just let's take again the tales of drought.
15:24
So they will craft from Python notebooks, they offer a way to understand climate data concerning droughts, and if you go here, you know, you have all the information. As I said, this is the project from this year, so it's quite, it's in progress,
15:47
so under construction, but they have a nice description, what they want to do, a roadmap, and everything you find on GitHub is mainly managed by the teams themselves, so we have
16:04
not a lot of restrictions, what they need to put up. Some projects decide that during the coding phase they work in private, so they are not, the repositories are not open to the public, as you could have seen here.
16:22
But if you go back, for example, to the repositories, so that is the FHIR forecast one, that is from from last year, it was about FHIR forecasting,
16:43
the name is program, and you have the entire documentation of the project here, and you can also access the teams, you know, and that is the idea of the program. So if I go back to my presentation,
17:10
yeah, so these are the other 13 projects of this year, and I did not go into detail, because we have also a lot of machine learning stuff,
17:21
yeah, maybe just the projects of this year, you know, we have here the CAMS visualizer with CAMS data, and then we talked about the tales of dry land,
17:42
then again CAMS, it's the Copernicus atmospheric monitoring service, then air quality monitoring, sun visor is nice, they are looking into which urban areas are good, or for photovoltaic parts, so it looks into the green energy aspects,
18:05
they are looking, that's a project from there, with AI, so it's really, really the topics of the challenges and the projects are really huge, and yeah, I would like to invite you to have a look at the projects, and also
18:24
look into, or join us for the midterm webinars, and get an idea what Code for Earth is, because, as I mentioned before, it's, you know, it's a yearly program, so I'm starting end of the year again collecting ideas, collecting challenges, and
18:42
early March 2025, or in 2024, or end of February, we will issue the next call for participation. Good. And then, I have shared that with you before, and
19:01
yeah, I like those closing thoughts. Coming together is a beginning, that's already nice, staying together is progress, and I see a lot of people here in the community, you know, they come not only once to FOSS4G, but more often, and what I really like is working together, that is success, and I think we have so many challenges ahead of us,
19:23
so let's work together, and be successful. Thank you. Thank you. Tina, are there any questions?
19:43
Thank you. How can we get one of these bags? No, that was another question. So, are there any of these prototypes that you developed during the
20:01
hackathon? Okay, is it better? Okay, any, do any of these prototypes that are, these codes, projects that are developed during the event, are later adopted through,
20:21
incorporated in ECMWF in a way, so do you pick up these software projects and take them to production sometimes? Yeah, so the question was, what happens if the project ends with the results? We have projects that just stay as they are, that's fine, that's nothing happens, but then we have projects that really go into, they're taken into
20:44
operational services of ECMWF and are further developed. So, for example, last year we had a regional reanalysis project with machine learning, that was really, we had actually two projects,
21:00
and they were not competing, it's not a competition, so each project that finished gets a 5,000 euro stipend. So, last year, this reanalysis project, I know that the colleagues picked it up, and they're building on that, they're working on that, they're developing it further, and another project
21:21
was very successful and picked up by the team. It really depends on the results of the project. I have a question, if I may. So, I wanted to ask, how do you pick mentors? Are they part of the team when they present the proposals?
21:44
Yeah, so So, I started last year, and how I picked mentors was the following, I went cup of coffee to each office, and I introduced myself and said, I'm running now Code for Earth, by the way, do you have a challenge?
22:04
Mainly, the colleagues come to us and say, well, of course, we inform them that we're looking for challenges, and then colleagues come to us and say, hey, Atina, Esperanza, we have an idea. So, it's mainly driven, well, we ask them, and they come back to us. So, the mentors are the people that propose the challenge.
22:25
Yeah, and so, what we do now with external partners, like the project we have with the University of Reading, or the Helmholtz Centrum, or EA, we approach them, or I talk with colleagues that, hey,
22:41
you're working together with EA, is there an idea or a topic that would be fit for a joint challenge for Code for Earth? And then I talk to colleagues at EA, and colleagues at ECMWF, and, you know, invite them to think about a challenge, and then, usually, they've come together, they have ideas.
23:04
What is important, if someone is interested to submit a challenge, we need always to find a partner in ECMWF, because it should be in the interest of the partner organization, as well as for ECMWF,
23:21
so that is a benefit for both, and from then on, these submitters, they become mentors. So, good, then I go back, so 15th, 17th and 25th of July, I want to invite you again,
23:45
join us for the midterm webinars, and thanks for your attention. Thank you, let's give a big applause to Athena. Thank you, everyone, for staying calm, and no red or yellow cards were given.