INTEROPERABILITY - Using linked data notifications to assemble the scholarly record on the decentralised web
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 | 14 | |
| Autor | ||
| Mitwirkende | ||
| Lizenz | CC-Namensnennung 3.0 Deutschland: 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 | 10.5446/60263 (DOI) | |
| Herausgeber | ||
| Erscheinungsjahr | ||
| Sprache |
Inhaltliche Metadaten
| Fachgebiet | ||
| Genre | ||
| Abstract |
|
00:00
Computeranimation
Transkript: Englisch(automatisch erzeugt)
00:00
And I'm happy that the first talk is a good fit to the lightning talk, to the talk we heard before. We heard that the current research infrastructure is broken because of large companies. And Patrick Kostenbach and his team from Belgium and Netherlands will show an alternative to it with an open research infrastructure.
00:28
So, we welcome Patrick. The stage is yours. Thank you, Gregor. Well, the first thing I need to say is that I really miss to be with you all in these big rooms and have a chance to talk with all of you.
00:49
This is a very long two years without real library social interaction, but I hope this talk will give a bit of an alignment to what I did the last year.
01:02
And indeed, like Jacob said, this talk, well, I didn't need a very long introduction because what Professor Laardman said, this is really what we did in the last year. And trying to find a solution, what can be done with this fact that we are living in
01:26
this world of big temperatures that do all kinds of things with visual data you really don't know. And this is in the title of our talk, using linked data with vacations to assemble the scholarly record on the decentralized web.
01:48
And that is something I am doing with my co-workers and co-researchers.
02:01
Next slide. So this is research that is sponsored by the Mellon. And this started in 2020, but because of the corona crisis, we got an extra year. And it's all based on the research that Herbert van der Semple did since the 1990s.
02:26
And because from a very early stage, it was already known for library techies that something was going in the wrong direction. How publications got on the web, how it was very fast and huge concentration of all these servers in a few library vendors.
02:53
And that was called the serial crisis with the cost of journals rising every year with crazy amounts.
03:06
So if you want to learn more about the vision that Herbert has, and still is based on what we are doing in this Mellon project, I advise you to watch his 270 CNI lecture.
03:23
There are also the latest things, what we are doing, and that's what the people are doing at CoreRutify. It's all around peer review. And the EFCUT sector is about memory institutions that want to share data, but also in the subject of architecture and buildings.
03:48
So what we are trying to do can be very well reflected in what the core vision is of what the next generation repository should look like.
04:03
And I copied here just a piece of text from the mission statement. What do we really need in the future to solve this, to solve the problems that Professor Landon talked about.
04:23
And I highlighted the important words here. It's about distributed and globally networked infrastructure that should be made for this quality. And on top of these things, we need to add layers and added services that we are deploying ourselves for our own world.
04:48
These things have to be research-centric and managed by us, the scholarly community. Of course, this is a very utopian thing that we are thinking about.
05:06
But in the light of what we heard in the keynote, it is important that we start already experimenting with these technologies and see how far we can go to solve these very dear problems we have.
05:23
These problems are really not new. It's already addressed in the keynote. Already, it's very interesting, not the decentralization, but really the decentralization of the web.
05:47
Because this all happened suddenly. When the internet was started in 1968, the internet was built in this open community without borders.
06:03
And everyone could have access to this internet. And it also was in the World Wide Web that Tim Berners-Lee invented in the 1990s. Also there, it was a very open web and it had to give all the people
06:24
the chance to share the data with the vision of a very utopian leveling of society. But what we saw is that it's not really happened.
06:41
You saw already the big portals that are dominating the web. The server is of course very free, but you are being sold on the web. If you are already in one of these portals, it's very hard to move
07:01
the data and there's also hardly any kind of interoperability between all these services. So there's an example where it says, if I want to move my LinkedIn data to Facebook and connect it with the images I got on Instagram or Flickr, it is very hard because all the data is really locked in these platforms.
07:29
And you can't move this. So if you say, well, Instagram has a very nice way to display all my artworks. Can they also use this in LinkedIn? No, it is really tied into one of these platforms.
07:45
That is not different from what we see in our world, which is already addressed in the keynote. We are really dominated by these massive publishers or we can't say that there are publishers these days.
08:08
Of course, and the service is very smooth, but free, I wouldn't say it is a very free service. You see the spirals everywhere. You have to pay the APCs to get things published there. There's a transfer of all your rights to the publishers.
08:32
There are no standard APIs for reviewing, annotating, indexing, searching and archiving. All these services are really tied to one of these platforms. You can't easily search between these services.
08:50
This thing you already saw in the keynote. The question is, is this picture okay? Is this what we really want? Is this the most ideal solution?
09:05
I hope you say no. What we are trying to do and Melon is trying to see if we can find a way to re-decentralize this web. The idea is there that we are going to give to all the researchers and
09:23
researchers about the environment where they are in full control of the data they are producing. Their publications, their datasets, everything you can mention. We also are building a decoupled way so we can do all the
09:45
core functionalities of what you need in scholarly publication like registration, peer review, archivation. Everything could be built with decentralized services and all need to be standard based. Of course it will make a bit
10:05
more constraining because standards means you need to have a way to have an interoperable way to talk with other people. But still we are trying to do this. Of course this is very technical
10:24
focused. There is also a social aspect but we are working on the technical focus. So what does it look like? Well you have all these researchers here, Alice, Bob and Charlie, they deposit their publications or their datasets in the personal data spot here on their personal DNS name.
10:48
So if you move to another university, you still keep all your data on the same URI. And then we are trying to build, based on this data, create interoperable service between your
11:06
pods and the rest of the world that are creating these service hubs in a decoupled way. So for example, when the researcher can say, well I need to have a registration of my publication or the dataset. Well this can be done by your institutional way pool.
11:27
It can be reviewed and there are now all over the world that are starting this decoupled review service.
11:44
Also you can say I want to have my publication in a web archive and all the fictionalities can be on different parts of the network. My video is still okay, I hope. How we do this is we are sending linked data notifications to all these service hubs and we
12:13
are creating now the standards how all these different actors should talk to each other to fulfill all these services and make this really happen.
12:27
So I had prepared some demos, but because we are short in time I can't show them. But hopefully on Thursday I will give a boot on how these things could work.
12:42
Also we want to have a publication of lifecycle events because when all these things are happening in these service hubs, there are things registered, there are things indexed and there are overlay journals and there are things archived and things in preview, what is left there?
13:04
And there is some extra data added to the network that is not yet written there. You need to express in an event log the performance of all the events that are happening on the data that is created in this researcher pod here.
13:23
So if you would open one of these event logs, you will see it is registered and certified and published and all the technical metadata, when it is done, who did it and in what way. Based on this event log data, you can add a lot of services like applications, can readers, and create back links between all
13:50
these services. You can give extra recommendations like if you publish data sets, please go to your institutional repo and add it there.
14:04
Please send it to the DMP to create data management things. There are also other kinds of services like the people outside of this network. They read online, on a social network they read some paper or data set and
14:28
they don't know if it is this one or this one or this one or this one or this one. Based on these event logs, you can show this metadata that shows exactly what has happened
14:42
to this data set or to this publication. It is quite a nice thing to do. So what we are doing is creating all these specs here. You can read this all in the open. We are also creating software, research software. It is not very good documented, but it
15:06
works. We have dashboards, we have orchestrators, and we have demonstrators for all these use cases here. I hope this all will happen. If it will not happen, I think this could be one of our futures if we don't solve this in the next few years.
15:31
So this is my last slide. I hope there will be some questions or not. I wish you a nice rest of the conference.
15:44
If you want to learn more about the tools I am creating on Twitter, I am creating a daily cartoon. It is called Solvembro. You can learn more about all the techniques we are using in this project.