Presentation State of GeoNetwork
This is a modal window.
Das Video konnte nicht geladen werden, da entweder ein Server- oder Netzwerkfehler auftrat oder das Format nicht unterstützt wird.
Formale Metadaten
Titel |
| |
Serientitel | ||
Anzahl der Teile | 237 | |
Autor | ||
Mitwirkende | ||
Lizenz | CC-Namensnennung 3.0 Unported: Sie dürfen das Werk bzw. den Inhalt zu jedem legalen Zweck nutzen, verändern und in unveränderter oder veränderter Form vervielfältigen, verbreiten und öffentlich zugänglich machen, sofern Sie den Namen des Autors/Rechteinhabers in der von ihm festgelegten Weise nennen. | |
Identifikatoren | 10.5446/57230 (DOI) | |
Herausgeber | ||
Erscheinungsjahr | ||
Sprache |
Inhaltliche Metadaten
Fachgebiet | ||
Genre | ||
Abstract |
|
FOSS4G Argentina 202126 / 237
12
15
16
23
26
36
44
52
53
54
59
72
90
99
114
121
122
123
124
127
129
130
139
154
155
166
203
204
219
223
224
230
00:00
OISCRechter WinkelProzess <Informatik>VideokonferenzElementargeometrieProgrammierungTouchscreenSoftwareOpen SourceZustandSpeicherabzugSoftwareentwicklerDatenflussDienst <Informatik>Quick-SortDatenverwaltungSchnitt <Mathematik>Besprechung/Interview
02:19
ZustandElementargeometrieRPCZustandSoftwareWort <Informatik>DivisionForcingProgrammbibliothekZahlenbereichPeer-to-Peer-NetzCoxeter-GruppeXML
02:51
RechnernetzElementargeometrieOpen SourceSoftwareDivisionSpeicherabzugKoordinatenProgrammbibliothekMaschinenschreibenComputeranimation
03:24
RechnernetzSkalierbarkeitKontextbezogenes SystemExogene VariableCASE <Informatik>RichtungHypermediaOrdnung <Mathematik>Coxeter-GruppeSkalierbarkeitDatensichtgerätElastische DeformationInterface <Schaltung>FokalpunktLeistung <Physik>Güte der AnpassungVektorpotenzialBildverstehenBenutzeroberflächeServerSuchmaschineComputerarchitekturHochdruckElementargeometrieDatensatzCodeÄhnlichkeitsgeometrieSoftwareComputeranimation
05:39
MenütechnikGruppenoperationDatensatzÄhnlichkeitsgeometrieMetadatenDatenfeldKoroutineVerschlingungAbfrageOnline-KatalogSchnelltasteTrennschärfe <Statistik>MultiplikationsoperatorFeuchteleitungComputeranimation
06:44
VersionsverwaltungRechnernetzEreignishorizontCodeDienst <Informatik>Architektur <Informatik>Ideal <Mathematik>RichtungComputerarchitekturSoftwarewartungHochdruckCodeVollständiger VerbandSuite <Programmpaket>QuaderSchnitt <Mathematik>ResultanteComputeranimation
07:52
DatenverwaltungSkalierbarkeitRefactoringDienst <Informatik>RechnernetzArchitektur <Informatik>Automatische IndexierungComputerarchitekturKartesische KoordinatenDienst <Informatik>RefactoringComputersicherheitProjektive EbeneSpeicherabzugCodeGatewayVerschlingungQuellcodeArithmetische FolgeVerkehrsinformationVerknüpfungsgliedGruppenoperationDämpfungSchlussregelAutomatische IndexierungRechenwerkSoftwareDatensatzComputeranimation
10:22
Dienst <Informatik>RechnernetzProxy ServerStandardabweichungVersionsverwaltungTurtle <Informatik>Automatische IndexierungDateiformatAuflösung <Mathematik>TypentheorieWald <Graphentheorie>Produkt <Mathematik>Bitmap-GraphikÜberlagerung <Mathematik>Web-SeiteDigital Object IdentifierDualitätstheorieDistributionenraumMinkowski-MetrikAbstraktionsebeneCodeSchlussregelSuchmaschineStrukturierte DatenKonstanteDatensatzDigital Object IdentifierWeb-SeiteSpeicherabzugOnline-KatalogLanding PageFunktion <Mathematik>UmwandlungsenthalpieInformationAutomatische IndexierungDateiformatRechter WinkelBootenOrbit <Mathematik>AusnahmebehandlungNummernsystemDienst <Informatik>GoogolComputeranimation
12:03
VererbungshierarchieMeta-TagDienst <Informatik>VererbungshierarchieSoftwareentwicklerDienst <Informatik>Online-KatalogExogene VariableDatensatzProjektive EbeneComputerarchitekturGesetz <Physik>DifferenteAutomatische IndexierungNummernsystemComputeranimation
12:31
InformationsspeicherungFigurierte ZahlOnline-KatalogMetrisches SystemAusnahmebehandlungInterface <Schaltung>CASE <Informatik>RPCProjektive EbeneDiagrammFlussdiagrammComputeranimation
13:01
RechnernetzSelbst organisierendes SystemDesintegration <Mathematik>MetadatenWeb-SeiteAutomatische IndexierungPortal <Internet>Offene MengeInterface <Schaltung>Architektur <Informatik>Basis <Mathematik>RenderingServerDatenverwaltungZustandCross-site scriptingVariableWeb logDemo <Programm>Projektive EbeneSpeicherabzugInterface <Schaltung>DatenverwaltungCASE <Informatik>Demo <Programm>Front-End <Software>Offene MengeMetadatenZusammenhängender GraphBenutzerbeteiligungVariableDifferenteZustandRenderingTermLanding PageFramework <Informatik>Web logAutomatische IndexierungWeb SiteBasis <Mathematik>Online-KatalogImplementierungAdressraumKomplex <Algebra>Portal <Internet>BenutzeroberflächeVisualisierungPrototypingKartesische KoordinatenCoxeter-GruppeSchlüsselverwaltungAdditionComputeranimation
15:50
MIDI <Musikelektronik>AppletZeitbereichCOMTermKommensurabilitätGammafunktionTwitter <Softwareplattform>Online-KatalogOffene MengeInterface <Schaltung>VisualisierungForcingZusammenhängender GraphEinsKugelEntscheidungstheorieCoxeter-Gruppe
16:24
StichprobeNormierter RaumProjektive EbeneForcingZusammenhängender GraphMetadatenCoxeter-GruppeMakrobefehlDatenfeldInverser LimesDienst <Informatik>Computeranimation
16:51
RechnernetzKartesische KoordinatenBenutzerbeteiligungAutomatische IndexierungZusammenhängender GraphCASE <Informatik>RoutingDienst <Informatik>Natürliche SpracheVollständiger VerbandRechenwerkComputeranimationFlussdiagramm
17:34
StandardabweichungW3C-StandardMetadatenSpeicherabzugPlug inRechnernetzFunktion <Mathematik>TexteditorTemplateAutomatische IndexierungOffene MengePhysikalischer EffektPRINCE2CodeHochdruckElementargeometrieZusammenhängender GraphSoftwareentwicklerOffene MengeStandardabweichungInterface <Schaltung>DatenbankBildschirmmaskeUnrundheitArithmetisches MittelPhasenumwandlungSelbst organisierendes SystemCoxeter-GruppeComputeranimationXML
18:40
Offene MengeFokalpunktDatensichtgerätMetadatenKartesische KoordinatenVollständiger VerbandSoftwareentwicklerMaschinenschreibenComputerarchitekturProjektive EbeneInstallation <Informatik>Analytische FortsetzungVerschlingungComputeranimation
19:37
Web SiteRechnernetzOpen SourceGammafunktionKollaboration <Informatik>Web SiteZusammenhängender GraphProgrammierumgebungDifferenteOnline-KatalogAppletFunktionalMathematikChatten <Kommunikation>MultiplikationsoperatorVersionsverwaltungRichtungE-MailSoftwareOpen SourceKonfiguration <Informatik>App <Programm>Message-PassingService providerTranslation <Mathematik>StandardabweichungPunktwolkeSoundverarbeitungGüte der AnpassungKategorie <Mathematik>GraphfärbungNummernsystemSoftwareentwicklerOffene MengeSystemverwaltungSoftwaretestOrdnung <Mathematik>SpeicherabzugFokalpunktExogene VariableDienst <Informatik>BitRankingGradientGarbentheorieProjektive EbeneSkalierbarkeitGruppenoperationDatenflussFigurierte ZahlSchlussregelNatürliche SpracheWhiteboardAbstandCloud ComputingDivisionProzess <Informatik>EinflussgrößeNormalvektorPhysikalische TheorieMobiles InternetCASE <Informatik>Diskrete UntergruppeComputeranimationBesprechung/Interview
27:12
Computeranimation
Transkript: Englisch(automatisch erzeugt)
00:14
Actually, we'll play a video now, right? Yes, I'm here to answer questions only.
00:21
Easy job for you, so you can sit back. So the last talk will be about the state of geo network, given by Florent Graven, and he has recorded the video.
00:41
And we'll play shortly. Florent, he's from France and is a front-end and technical leader, but also a geospatial full stack developer at Camp2Camp. I think he's one of the sponsors as well at this conference and a very innovative open source geospatial development company.
01:09
And Paul from the Netherlands also studied in Wageningen. And he recently started a new job at ISRIK and he already gave the URL in Wageningen as well.
01:28
Probably most of you know him as he worked the last 10 years at Geocut, also a very innovative open source geospatial company in the Netherlands,
01:40
although I should say the main company behind geo network and also one of the main sponsors. And Paul is called a core contributor and he's also in the program steering committee of geo network, but also in many other OSGEO projects, including by geo API.
02:02
So without further ado, I will play... Take some popcorn, yeah. Yeah. Play a video. Let's see what Florent has prepared for us. It will cover the whole screen now.
02:22
Probably we will go to the back side. Hi, everyone. I'm going to present the state of geo network 2021 at Phosphorgy, Buenos Aires. I'm glad you are here remotely. Unfortunately, I'm not there, neither physically, neither remotely. So I've prepared this talk with Paul, who is going to be there, hopefully, to answer your question at the end of the presentation.
02:50
Thanks to him. A few words about me. I'm Florent Graven. I work at Camp2Camp in the geospatial division as a technical leader. I'm also a member of PSC of geo network and a contributor to many open source geospatial libraries and softwares.
03:11
And it's the core of what we are doing at Camp2Camp. Here are my coordinates. If you want to get in touch, I would be pleased to talk about geo network or OSGEO in general.
03:24
So what's new in geo network ecosystems? Actually, many things. Yes, it was a very prolific course. New projects, new direction, new architecture, new UI, many things, many code prints.
03:40
I'm very excited to talk about that today. So let's discover everything during this presentation. But first, let's introduce the new release, the major one, the fourth release of geo network. The principal focus was the move from lutein to Elasticsearch. Elasticsearch is a very powerful search engine.
04:05
And it was really a great step forward to move to that because it ensures a new potential for the geo network search. First, it's faster. The response are faster. But also, the display in the UI is faster because we can customize what we want in the response of Elasticsearch.
04:27
So it means for each search in the UI, depending on the use case and the context, we can just fetch small and few record fields, which is great for the performances.
04:41
It's scalable, so it was very crucial for now because lutein was not duplicable. And it's great, Elasticsearch, you can have clusters. So you can scale Elasticsearch in Kubernetes or other rancher ecosystem. And it's great if you want to have good performances and heavy requests on your server.
05:06
It's heavily configurable with facets, with boost, with many things. And it brings new features like better suggestions, similarity record, etc. You can find the full presentation in there.
05:21
Let's quickly boost to this new version. So beside all of that, there was still many user interface improvements to make the interface clean, clear, and usable. The search has been improved. The suggestions are really improved in there and faster.
05:46
We focus on the title. There is a boost. You can customize that. So the research is really better. The facets, you can have new kinds of facets. You can have nested facets.
06:01
So here it's a value in the field. And in there we group them. It's like an aggregation. And we can jump in there. There is the negative selection, which is new. You can say, I don't want S3 shapefile. There is some facets on routine queries.
06:23
So you can just bind the facets on queries. And it gives you the amount of records that match that. On the metadata page, the links have been improved. And you can find the similarity record in there and jump in there very easily to boost your catalog.
06:44
It's not the only thing during this year. Actually, many releases, four or five per year, maintenance of the third version, 3.12, and 4.05 has been released very recently.
07:04
We have a lot of meetings, technical meetings, to decide the direction. Lots of code prints this year, even if there was remotely. We tried to really keep active on the community. So, yes, it was a very prolific year.
07:21
And, yes, pretty exciting things to show. Everything quite started at the meeting last year, 2020. Very good success. Many participants, many ideas. We dream about what we would like GeoNetwork to be. And two directions came out from these meetings.
07:42
We wanted to dream about a new architecture based on microservices. And to really focus more on user experience concerns. I will speak about the microservice architecture first. So the idea is to break the monolith. So GeoNetwork has a big core with a lot of code.
08:04
Some of it is deprecated. And it's very hard to split. So the idea, instead of doing a huge refactoring, is to start from scratch. New services, pure, clean. And to get rid of old dead code.
08:23
The idea as well is that GeoNetwork is a mono application. So it means when you run background tasks, like harvesting, exporting, and things like this, it consumes a lot of resources. And the search, public search, is slow down, which is really a pity.
08:45
So the idea is to split that in many microservices, new services, clean. And you can scale each services accordingly. So you would have harvesters. You allow a lot of resources for that.
09:02
You allocate a lot of resources. And the search is never impacted. So there is a project on GitHub. You can find the sources and the progress. The idea of the architecture is about that. So in the middle, in green, you have all the services, the microservices behind the gateway
09:23
that is provided by the project. And you can, for now, the transition, it still lives beside GeoNetwork for API. And you can use the microservices API from the GeoNetwork for UI as well.
09:40
So it's a transition, and we are doing more and more services and improving the link between the services, the gateway, the discovery service, the authentication, and the security. At the moment, there are three services. Searching, SCSW, GRSS, indexing, and OGC API records,
10:08
which we are talking about right now. It's in progress, so more service will come, and the refactoring will be better. Harvesting services is on the rail as well.
10:23
So OGC API records. So I think you have heard about OGC API for the features, for the layers, the style, many things. There is also a specification for the records. So it's new OGC standards, and the OGC API record is microservices
10:44
and is just implemented off top of Elasticsearch. So all your GeoNetwork for catalog is indexed in Elasticsearch, and the OGC API record microservice just needs Elasticsearch and extracts the information from the index to the records document.
11:05
The idea is to improve the discoverability of GeoNetwork so we can find the records on Google. We use common formats like schema.org, dcat, dcatap, inspire.
11:20
So we try to cover most format we can, so it can be used from everywhere. We have the DOI for the landing pages that you can find. So this is the example of the page on the right. You can choose the schema and the output.
11:41
So it's the way it's done so far. We have JSON-LD for schema.org, but also for other services like dcat, etc. And it's really great to have this API in front of GeoNetwork for index
12:00
so you can serve different purposes than the core. Super Harvester. It's a new microservices. It's in development. And the goal is to have a new European meta-catalog that harvests all the catalogs in CSW. So it's a very new, brand new project
12:22
whose responsibility is to harvest, ingest, index, and validate the records of the whole Europe. A very brief schema on the architecture. So yes, you have Elasticsearch, you have the Harvester,
12:40
you have the different schema, etc. Some work has been done also to provide an inspired dashboard. So based on Elasticsearch and Kibana, you have lots of metrics, charts, data figures. So you can really monitor your catalog from this UI.
13:03
GeoNetwork UI. I was talking about improve the user experience. So we started a new project called GeoNetwork UI. It's really use case centric. We want to focus on user experience and bring modernity in the interface of GeoNetwork.
13:20
We don't want to have a too complex user interface, but something simple, inspired from open data catalog to bring innovation in the world of metadata. You can find the project on GitHub. We try to address new use cases, more modern,
13:40
like open data portals with some data visualization. We want simple addition, landing page indexing. We want to integrate some components in third-party websites. We want to search both in the metadata and in datasets. We want to have pure UIs.
14:01
So we try to address all these use cases. And in the end, we develop GeoNetwork UI, which is actually a framework that provides a toolkit to ease the creation of interface, application, components, and website around catalog, datasets, and metadata.
14:21
So from this, we can easily implement new stuff, new components. So here are the basis of the architecture. We use a framework based on components. We really focus on web component exports for third-party integration, server-size rendering for the landing pages. We want to use CSS variables to customize the rendering
14:44
and the layout for the different use cases. We use a state management. We use an API generator to interface with GeoNetwork for backend and many things. So there is a blog post to talk about that.
15:00
You can find it here. And now, briefly, some demo about that. So first, the Storybook. Storybook is great because it allows to really separate the concern, the presentation, and the logic. So people can contribute to GeoNetwork
15:22
even without big skills in terms of the core of GeoNetwork but just doing presentational components. So there is the Storybook. You can find different components that we have.
15:41
You can test it. You can change the input, et cetera. So it's a great showcase. Another example of implementation is this prototype towards an open data catalog. So it's a really pure interface.
16:00
You can pick metadata here. And you have kind of open data visualization. With data visualization, you can visualize the data in a table. You can visualize data on the map. You have to export. So it's really in the trend of open data catalogs.
16:22
And there is another thing that has been developed with GeoNetwork UI is the data feeder. There is a presentation during the force. So the idea here is to bring wizard component to fill the metadata. And then, because it's in GeoNetwork UI,
16:42
this is shareable and this allow you to publish and create a new metadata. So many things, very interesting thing in this project. So with the microservices and GeoNetwork UI,
17:01
it's the new constellation of GeoNetwork. So you have GeoNetwork UI. You have many applications. You have web components. You have many things. GeoNetwork 4 in the middle as usual. So GeoNetwork UI can live beside GeoNetwork 4 for specific use cases.
17:21
It call the GeoNetwork 4 API or the GeoNetwork 4 microservices. And the microservices in there are narrowed to the index or other pieces of GeoNetwork. There have been code prints about Dica2, many things to try to embrace the open data world
17:45
and not to stick only on Geo and Inspire ISO data sets. So we have done code prints and we are trying to do a new interface called DataHub that is going to gather within GeoNetwork data,
18:03
geo data, non-geo data, and open data. So good work has been done on the Dica2 standard. So there is a very straightforward and simple edition form in there based on suggestion, on thesery, and on shared resources.
18:25
We have also database components that you can see in the storybooks to interface on the data. So we really try to embrace the open data world and it's an ongoing development.
18:43
So what's coming next? Yes, as you can see, everything from this year, from the GeoNetwork meeting is new idea, new concept, new projects. So everything is under development. We will release new... The Data Feeder application has been released
19:00
but we will release application continuously from the UI and we will improve the microservices. So we are really at the early ages of this new constellation and this new architecture of GeoNetwork. So yes, keep in touch to follow what came out from that project,
19:24
very exciting project. And you can keep an eye on what is next. Link in there to follow our continuous roadmap and things about what's coming next with GeoNetwork. How to participate and collaborate?
19:40
You can find all the resources of the community. So the website, the Forge, the mailing list, etc. We are very active. So more and more contribution is great. We can see more and more downloads on the Forge. So yes, it's very great to see that GeoNetwork is still worldwide used.
20:02
And we really hope that all this new direction, this new project, the GeoNetwork UI and the microservices will really convince you in the near future. But really keep an eye on that. And if you want to help the community, you can contribute as you can on the translation with the trans effects
20:27
or give your ideas on the issues on GitHub. Yes, thank you very much from all the community. I hope you are excited about GeoNetwork and what we propose for the next future.
20:42
Thank you. And thank you, Paul, now to take the lead for the questions. Bye-bye. Yes, thank you very much, Laurent, from a distance. Yeah, Paul will be here to answer your questions.
21:03
And we already received some questions in the chat. I will start with an easy one. They're all voted in the same category. So the first is, is GeoNetwork 4 compatible with JDK 1.1?
21:22
Isn't that unfair? Or with JDK 11? Sorry. I left Java JDK 1.1. So that's not true. It's good that I didn't ask for JDK 17. 17, okay. I don't know what's upcoming. No, but the core, the GeoNetwork 4 core,
21:46
so that's kind of the current version, is JDK 8 only. So I really recommend the OpenJDK container. However, the microservices project,
22:01
which will soon replace most of the components of the core also, is JDK 11 only. So in order to run both in an infrastructure, you already need kind of a Docker orchestration with microservices
22:21
where you have the newer components on the JDK 11 and the older components on the JDK 8. And now these components work together, but they have a different Java environment. But since we now have Elasticsearch
22:41
and deploying this on Docker or Kubernetes is anyway a better option than it was also for scalability. Of course, yeah. We keep that hearing back that we all should use Docker.
23:02
Maybe that's a message I hear to many presentations. And okay, there's, thank you. And there's another question you probably can answer. So is the GeoNetwork compatible with the EU accessibility directive on the accessibility of the websites and mobile applications of public sector bodies?
23:22
Yes, so it's a bit of a complicated answer. It's yes and no. It has our constant focus because we get these questions a lot from our users. On the other hand, it's also a shared responsibility between the developers of the software
23:41
and how it is implemented because it's also dependent, for example, which color scheme you pick as administrator already can fail this test. So in theory, the software itself is compatible, but it also requires some attention from you as an administrator.
24:06
Yeah, it's a complicated issue. We see it in the Netherlands as well. And the third and probably the last question is, are there any issues with licensing using Elasticsearch?
24:20
Yeah, that's a hot topic for almost a one and a half year already. So in normal usage, when you deploy Elasticsearch yourself and in your own environment or even on the cloud infrastructure,
24:42
you're totally fine with the license from Elasticsearch. However, if you go to a service provider that offers you this as a hosted service, then there may be issues. In that case, I would actually recommend to use the paid version of Elasticsearch.
25:02
On the other hand, there is Amazon that has created this open search project, which is based on the Elasticsearch version that was still fully open source. And we try at least to be compatible with that version.
25:21
There may be some functionality not available, but at least it will work. It will not break. Yeah, I just looked at the website. I see there are some changes in 2021. So let's see if there are any other questions in the chat.
25:47
I think we're there and we're in time. So I see a question from Jeff related to ISO 10110. That's the feature catalog standard.
26:04
So yes, that is supported. Because that's functionality that was available since a long time. Or maybe you referenced some recent improvements on that, but I'm not fully aware of that.
26:26
Okay. Well, I think that concludes this session, these six sessions. And I think we have the icebreaker now?
26:44
Or is there a keynote still? Okay, thanks, Paul and Floral and all the previous speakers in this session for their well-prepared talks and keeping us up to date. And well, we'll see many of you in upcoming sessions.
27:06
So thanks very much. And have a good conference.