The LIBER Research Data Management Working Group webinar, ‘Data Management Plans: Use and Reuse’, took place on the 1st of February 2024.
The webinar invited participants to think about how the content of data management plans can be used in libraries or other institutions, and what kind of information DMPs contain that can be reused for other purposes.
Speakers:
– Dr. Kerstin Wedlich-Zachodin (KIT) and Kerstin Soltau (FIZ Karlsruhe). Linking a DMP tool to a repository: an update on the use of RDMO and RADAR
– Elisa Rodenburg (VU Amsterdam). Reusing DMP content for other purposes.
– Mari Elisa Kuusniemi (Helsinki University Library). Targeted marketing of institutional long-term preservation archive.
Moderator: Kirsten Krogh Kruuse (AU Library, Royal Danish Library)
More information about the RDM Working Group: https://libereurope.eu/working-group/research-data-management/
Good afternoon my name is Kirsten ker and I’m from the oos University Library which is a part of the Royal Danish library on behalf of the Liber research data management working group it’s a pleasure for me to welcome you all to this webinar on the use and reuse of data management
Plans as you probably know libraries can play a crucial role in supporting the content of data management plans just to mention a few examples libraries can provide guidance and training to researchers on creating effective dmps we can offer DMP review services to help researchers ensure that their plans meet funder requirements and
Institutional policies and we can help to ensure that data are well documented and organized according to the information in the DMP today’s webinar will focus on how libraries can use the content of data management plans to develop better Services also a warm welcome to our three speakers the first speaker is Dr castin
V sakin from the K Institute of Technology the title of her presentation is linking a DMP tool to a repository an update on the use of rdmo and radar the second speaker is Elisa Rodenburg from the F University Amsterdam Elisa will talk about reusing DMP content for other
Purposes the third speaker is Maria Lisa Ki from the Helsinki University Library the title of her presentation is targeted marketing of an Institutional long-term preservation archive there will be time for questions after each presentation after the presentations we’ll send you into breakout rooms and we have prepared some discussion points
But you are free to add other topics as well each group should pick a reporter and report on their main thoughts we’ll get back to the discussion points later as for housekeeping rules there are five things you should be aware of keep your cameras and microphones closed during the
Presentations use raise hand under reactions at the bottom of the screen or the chat for questions or comments please participate actively in poll and group discussion join designated Discussion Group use camera and microphone in there and the session is being recorded now it’s time for poll and we have
Prepared some questions for you and I’ll Just sh stop my sharing of the screen so that the leaper office can introduce the PO questions for you okay so all of you have some understanding of what a data management plan is few people people have a general idea
But have never seen a DMP many people have made or read several dmps and also quite many people are experts in dmps so thank you very much yes and now it’s time for the first presentation and as you can see on the screen there’s a short abstract and it is presented by
Dr Cas B zakin from H School uh and I stop my share so that casine can share her her slides so hello everyone um I hope you can see everything um I sorry you can’t see me because my my camera is not working and I didn’t want to start a
Computer again so we just have to deal with that now um yeah I’m presenting together um with castin Salto from the Fitz C who is the um product manager of radar the um repository and um I’m talking about how we are linking a DMP tool to this
Repository and as a DMP tool we are using the research data management organizer um it’s the German development um which is so the questionnaires and everything is usually more tailored to the German funders but we also have of course the Europe The Horizon Europe as an example and
Um yeah and it’s also available in a few languages so actually any body can use it if he wants to um yeah so I’m just going to start right now giving you a little bit of a detail on the research data management organizer um for us it is more than just another DMP
Tool um because it’s really more organizer and so we’re shifting from the static um DMP templates we usually know to actually um some active ones and so the focus shifts from planning to organizm research data management um it is very nice because you’re structuring the information on research data management based on
Projects and also on collaborations you can Import and Export projects between different research management organ instances so there are a few ones now available in Germany and um also we have freely programmable project exports and imports these are called so rdmo plugins we have a few ones that’s it’s
The RDA DMP common standard for machine action of DMP um we have also one for data side XML for data sets and then we have also the one for radar metad data adjacent for data sets and we have a few other ones but um um we are going to focus on the radar
One and you can also make use of external apis that’s also possible with this tool then just another more detailed view of um rdmo itself so it kind of works like uh through a structured interview so we have various question cataloges U generic but also discipline specific ones you can have with
Institutional ones um we have various um fidgets controll vocabularies help test tax navigation system we have tabs for different project Partners data set Etc and there’s also a possibility skipping questions um with different conditions so the conditions basically tell you if there’s a yes or no question and if you
Say no you’re avoid the next one um I already mentioned it’s multilingual so we have German English French Italian and Spanish by now but the default one is always English and we usually have a German translation um yeah we can create different versions so by snapshots
Um we can display various tasks such as what something else what I still need to do um and display and Export questions VI in common formats and according to requirement of specific funders such as the Horizon Europe for instance um just a few quick overview um
Um how many there are right now you see on the left hand side a map we even have um so it’s not just Germany we also have an instance uh in Austria in Vienna but there’s also one in Italy and another one I think in France so
Um yeah so there’s there’s a big Community out there um and also this ensures kind of the continu of the software because we came together as a community um we funded a working group together which consists of three different groups like a governance group who uh governs the group itself um
Then we have um a software Group which is responsible for the um development of course and we have a Content group um and the content group usually takes care of something like new add think new attributes um which link the questions um to The Domain and we have
Like like um we also do we have different subw workking groups which kind of do different things such as um de developing content of a new questionnaire Etc or working on the website of the um Community Etc we use um so for the users we use slack channels for exchanging um opinions and
Help so if somebody wants get in contact with us the fastest way is always through select Channel but we also have a dfn mailing list on which we can put you on to and now I would just like to give you a quick um demonstration I think that’s
Better than than showing you just some screenshots so I will stop the sharing again and um go right into the tool because we really actually want to show you how this actually works um connecting why this plugin um so just one second again um yes so this is this one
So guessing you can see this I’m just going to quickly look in why orchid this is one thing it’s just just one moment so um this should be working so um usually this this is uh the main page in the beginning once you logged into the
System with Orchid um you have you see the projects here usually this would be empty but I already uh added two projects of mine um we can usually create here a new project um I don’t want to show everything but usually this is how it works you create a new project
You give a title you choose a catalog and depending on which instance you are you’re Ling on to you have different catalogs available um the rdmo catalog here is the main one the default one you always have a possibility also to add a parent project um to your current
Project so you kind of and connect them um so I could choose here if I wanted to anyway I just um want to show you the the project we put together just for today called it the test Liber Workshop so once you established your um your um your project basically this
Is this is the display you get so you get a description of of the um questionnaire the catalog you see you can always edit things by this pencil here on the right hand side and on the right hand side you have a lot of options of updating the catalog the
Project the task whatever um there’s there’s uh a big part here called the export so this is also a way also to make sure that you always um are updated with or you have to write uh files all the time time because you can export your um
DMP as an XML file um which makes it possible for you also to import it into any other instance um because this is how it works but what we also have down here is you can see you can also export it as Rader XML and we also have something like
Called directly to radar or directly to codo and we we are going later to show you how this here works but just a few more things which are nice about this tool like I already said tasks are displayed here then we have different views so you can um have a view um from
Horizon Europe here for instance I can show that later and also you can add members so you can see me here is castin within and you can see casan Salto um she’s the at the moment and I’m just a manager um so if you if you want um to add somebody else
You can do that um yeah so usually usually you can um it’s probably not me I cannot show you this because um I’m not the owner this is why the button is not here sorry um but um usually I would be able to um add um somebody else
To the project but in this case only the owner can do this but usually you are able to um like um give different roles so I can if I’m the owner I can tell I can invite somebody and give him the function as a just as a as a author or just um
Somebody who can read yeah and cannot change anything so this is possible here snapshots are usually displayed here and um this now works that we um start with just answering the questions um yeah so starts with the general questions usually with a topic and
Um so it goes now through all a lot of question I don’t want to go all through of this because we don’t have so much time I just wanted to show you where you add basically the um um just one second um when you set the you add the data sets so
Here um there’s a section where you can add the data sets um so per data set you add a extra data management plan so so to say so per data set you give a different name which we’ve done here we have data set one and we have data set
Two and these are the ones which we should uh at least the media data which should be later be transferred to the repository yeah so as you can see here on the right hand side just to give you a quick um update here here’s the project we have the catalog we can
Always go back to our project CST progress bar which kind of tells us how many question we asked uh we answered and the navigation here shows us uh where we are so this is how it works usually um so if I go now to the last one and I
Go to complete questionnaire it says done thank you for filling out the questionnaire and I can view my answers um they’re being displayed here on the right hand side I have export options in different various formats um PDF can be also just txt but it can be also latex or markdown even
Um I’m just going to go back to the project and I just want to show you quickly um with the view of the Horizon here so so this is the Horizon template and these are the questionnaires with the with the format of the Horizon Europe
Now so this is how it works and now I would stop it quickly and let my colleague casan soter continue to show you how this uh how the transfer works so I’m stopping sharing now yes and now I skp in I hope you see now my
Screen because uh uh before I follow up with this little demonstration I would like to uh say pref a few words about radar itself what it is uh it’s a cross disciplinary repository service for the archival publication and peer review of research data from already completed scientific studies and projects its
Target audience uh this is a publicly funded research institutions and universities and with help of R they can establish an own generic institutional uh research data repository R was developed uh in a project funded by the German research sech organization called DFG and since th 2017 it’s an productive uh uh cloud
Service called radar Cloud and it’s available this cloud service is available exclusively for German academic institutions software development and operation is done by my Institute Fitz car for Li nits Institute for information infrastructure however um three years ago in 2021 we introduce an additional business variant which is called Ro radar local
And as the name implies here we run the radar software locally on institution’s Own It infrastructure and this offering can also be licensed outside Germany as well um as you might have noticed there are some similarities between rdmo and radar both are DFG funded they have the
Same uh target audience and both both are generic Services however opposite to rdmo radar as data repository is currently at the end of the research data life cycle and to remedy this for the benefit of researchers the idea of linking these two services with each other was
Born and um by linking these two Services those descriptive metad dat which are already available in rdmo since researcher entered those during the questionnaire those can be transferred into the data repository radar and the advantage for researchers are of course duplicate entries are avoided and errors might also be
Avoided and um here more in detail how does it work there are two options for this data transfer and I will show both methods in a moment the first option is that users can export Radder uh Radder metad data from adom o as an XML file and import those
Meta data into an already existing radar data set which is not yet archived or published and the second option is um called directly to radar that means that users can directly uh directly transfer disc of metadata to radar with only a few clicks and create a new data set there and this
Is done via the radar rest API and as next steps they take place in radar as soon as these descriptive metadata are there uh that means that other metadata can be entered in radar afterwards this must be done at least for those metad data which haven’t been
Entered in rdmo before but but are mandatory or optional Fields within radar and of course the research data must be uploaded into radar in a separate step and finally of course the data set can be created archived peer reviewed or published within radar and I also would like to
Demonstrate this now uh for this I use our radar test uh system which is linked with the rmo development system and I just close my presentation I hope that you still see it do you still see my presentation or do I have to open it again no we see
Your website website you see the website of the rmo uh we see yeah you’re right right yeah yes you see it uh and you see that I’m marking up uh the the the project at the moment yeah okay my my my uh edge here um is not visible any longer okay good
Then we will continue it’s the same um project cassine already showed and she already showed that we have this export section here on the right navigation frame and we see two options this s radar XML this means the XML op uh export option um this would give me an a simple
XML of the descriptive metad data the user entered uh however would like to focus on the directly to radar option the direct transfer using our rest API which I will show now simply I click simply on this link here and then in the next screen I have to choose the data
Set which I would like to transfer to radar and I also have to select a workspace in radar to which I would like to transfer the uh the descriptive metad data then I click export to radar and in the next screen I have to authenticate with Rader I enter my test credentials
Here and then in the next step radar will create hopefully a landing page for this new data set and uh displays those descriptive metadata which have been transferred from rdmo to radar these are in this case the data set name uh an abstract descriptions some keywords have
Been entered in rmo and R transfer toor and production year subject areas the rights information and the funding information however you see there are some empty Fields those which are not covered by the rdmo questionnaire and those have to be entered then in the next step via our
Um form based manual based um metadata editor for example here then I include author Creator and right holder and these other descriptive metad data necessary according to the radar metad data schema and if I would have chosen um export to XML with XM XML then I
Would have the option to drag and drop this XML file for example in this section here yes I think this is the end of my little demonstration I just skip here thank you very much and one moment yes I would like to thank you of course for um for your attendance and
Um maybe also for your questions thanks thank you very much if you have questions uh then please use the reactions um at the bottom and I can see that there many applauses uh Cora rman has a question yes thank you thank you castin and castin um could you evaluate a bit
More on the meta data that described the data sets on how vocabularies they are based on um and the second question that I would have uh relates to the workspace in radar uh has that to be allocated in advance or is that something that is there anyway per
Instance yes be c i I will um at least answer the second uh question uh yes um before an institution is R they have to um make an agreement with us with Fitz C and they have to um Define uh a person who is a so-called administrator who administrates this contract for the
Using institution and then this administrator he sets up different uh work groups within rer which correspond to project teams to institutes and so on and here we have a rights and roll system and this administ rator defines who is a data provider within um certain workspace so this has of course to be
Set up before in advance well the the media data itself um it is a bit more it’s not so let’s say it this way um it’s is usually this this plugin now is focused on the main catalog so the m o catalog and there are just a few attributes like
I said it works by the attributes which uh connect to to this informations so only this more information is then transferred um what you but what you could do probably is then to go into the code and um basically tell it to um also add other attributes
So yeah so this is what needed to be done so you need to be a bit more sevy with these things I guess with the coding and with a bit of um the computer language so but that that would mean also that I would have
To do the work double yeah I mean if I have the metadata like for example in the sensor management system system which should be connected to the rmo I already endered all the meta data related to my data set for example like instruments or something like that and
Then how is that I mean variable names and uh instrumentation and so on how is that then translated to finally like in radar for kit or whatever um if I would have to do it again in in the code then I have to I’m not I’m not saying that you need to
Put it in this it’s just this is how it works at the moment this plugin is now made only for a few attributes so if you want to change this yes you need to go and change the code this the way it is um but um anything else how you connect another
Another system by the AP to the rdmo I think you need to talk really to developers here um you know I’m sorry but you’ve been already in contact with them right so um yeah I’m just waiting I have not to stud move so far I’m waiting
For yeah you’re waiting for the for the test test system to be usable I know I know I know it shouldn’t take so much longer really it’s just a few more technical things and then it should be working soon yeah okay great great thank you very much thank you welome
There’s also a question from Christina hner yes thank you for this uh yeah very Advanced I would say um integration system it’s nice to see um I was wondering what is your experience um with the usability uh because a DMP is usually uh of course set up in the
Beginning of a project and then you want to enter a data set or like publish a data set at the end of a project so usually a lot of things are happening in between and things change and evolve and so what uh what sort of information in practice can you actually link um
Between these two documents except the persons or the yeah I don’t know I was wondering how would you how is your experience in real life well I’m afraid there hasn’t been so much experience yet with the system we we just been testing it all themselves at the moment and we don’t
Really know of anybody who’s actually using it at the moment there’s all there’s a project uh in potam which will take this more into Focus but I don’t know how far they are um this is all I can tell you at the moment sorry oh no it’s it’s fine I just
Didn’t uh I missed that I guess like the the status of the of the project but uh thank you welcome yes there’s still a question from Yasmin toat and I think after that um we will stop the questions for now there are a couple of questions in the
Chat and we we may get back to them after we have had the discussions but um time is running and and we’ better proceed with the program but uh please Jasmine thank you very much yeah it’s quite interesting thanks for sharing and I had exactly the same question as
Christina and I understand of course it’s at an beginning stage so I would be very curious to see how it will evolve because it’s also a question we have in general that researchers do it only at the beginning and what happens later on and follow up on that is quite
Interesting and I was also just I’m trying to understand because you are gathering all this metadata but when I think of the data management plans we have we use DMP online we are in Netherlands at the University of Technology and there uh we have one question that asks all sorts of data
Sets that that data that are involved but then the rest of the questions are all General so how do you link it per data or did you say maybe I kind of miss it so that you really have like mini dmps within the rdmo per data set so that you can really export
Metadata per data set because when I think of a project there are all sorts of data that are involved so not just one type of data yes um so once in this tool when I put up a data set that then usually I need to answer all questions for each
Data set and now the media data is that is transferred is transferred for each data set you saw that um Pro Miss Salto um only took chose one data set but there was another data set she could have chosen to transfer it to radar but
At the moment the The Meta data which is transferred is just a small part which um resembles the the media data template in in in radar so this is this is what it is at the moment this is also what I meant with you want to have more
Attributes you want to have more media data you need to change it in the code at the moment thank you very much here welome thank you very much for an interesting presentation and now it’s time for Elisa wbor to give her presentation so please Elisa thank you Ken uh I will share my
Screen I hope you can hear me all right uh shout please shout if if if something’s wrong with the audio um so I will speak uh for about 10 minutes uh and then we’ll have five minutes for Q&A and um this is going to be quite a different uh presentation from the
Previous one uh I will talk about a bit more about work in progress uh and I’m also afraid there’s there’s not really a demo involved uh so very briefly about me um because it will help you to understand uh what we’re aiming for and what the
Context is so I’ve been at the uh V libraries RDM support desk since uh 2020 uh I’ve been a data Steward officially since 20121 and uh since last year I’ve been a coordinator of the RDM support desk and also a member of the Lio working group for research data
Management and um the RDM support desk oners researchers questions about RDM and open science and if we don’t know the answer to people’s questions we can uh refer researches to Specialists and this might be a legal officer an IT consultant or someone in their faculty and we also help help
Researchers with writing a DMP we give them advice we review the DMP if they ask us to and um based on the outcomes of the DMP we might refer a researcher to some uh other division in the University this could be a privacy officer uh it could be an IT consultant
It could be about it security or just computer environments um and um this is we are kind of the the central point in the spiderweb of different um uh of different RDM services in which the researcher then gets called and can never escape from anymore
Uh so about uh those dmps um uh we use DMP online uh uh indeed and this is managed by the the library we uh manage the the templates we make sure that the the thing works that we that the templates have the right um uh questions that that that yeah that bugs are
Reported Etc uh we also have we review dmps but we also have data students and faculties who can review dmps and they have those rights in DMP online to see the the dmps our institutional DMP template is accepted by Dutch funders um so uh if researchers want to
They can use our DMP instead of the funer uh our template instead of the funer template and our DMP template also includes questions that are relevant for gdpr registration and I will talk a little bit more about that in the next slide uh uh but I wanted to show here
That we also have a template for those who do not write a DMP or who have already written a DMP in a different template so in DMP online if you clicked that you are um um connected to Affiliated to the view then you will see two templates our regular DMP online
Template and the gdpr registration form uh and so uh as the gdpr is is the general data protection regulation it’s the EU Privacy Law and as most of you will know it is a requirement under the gdpr to write uh uh to to record in a in
A register what you what what processing steps you will do to uh personal data uh and there is much overlap sort of naturally there’s much overlap with the dmps in terms of the information that the gdpr registration needs to record uh but researchers in general don’t like filling out a lot of forms
That then all ask for the same thing so uh uh we added a few questions to our DMP online template uh that were especially for this gdpr registration and we also in in the D in the in the online template we also already had questions uh that related to gdpr so uh
Add those added together we can uh create a pretty good gdpr registration and obviously uh uh there’s also questions in the DMP template that do not relate to gdpr um so we used the API from uh DMP online to extract the information uh from the relevant questions there s of
Questions relevant to gdpr registration and Rew write that to a Json file uh and this uh this this Json file is then the gdpr registration for that project and as I uh uh showed in the previous slide we also have a gdpr template uh for those researchers who
Don’t write a DMP uh or uh they wrote it before we had this in place or they they used a different template for example if a funer absolutely demands that that template is used and um this uh form has only the questions that are relevant for the gdpr
Registration so it’s a lot shorter and this entire form is also written to the Json file and and and um this becomes a registration as well uh organizational challenges mean that it’s pretty possible to develop this stuff especially if you have a colleague who’s e for a hobby
Programming project and and we did um uh but embedding these things in the organization is an entirely different cup of tea so I think I have a photo of two young women here and uh they they’re wearing what we call in D how Yas and that is a Monte coat and the the
Fastening in the coat is is a is you have to put a a little piece of wood through a piece of string and it’s it’s kind of it’s It’s called a frog fastening in English in Dutch it’s called how and that is also an expression for something that is
Built uh uh not very well structured it’s it’s it’s used for a solution that works but it’s also a bit shaky it’s a bit improvised so we have the colleague who uses the script manually uh but nothing is automated and if he’s on holiday the script doesn’t run on it on
Its own and he told me on Tuesday that he has to open his laptop when he is on holiday every day to run the script so this not ideal and we’re we’re working with it to set up a server to automate the exports and and and automatically
Save the Json file in the right location but due to sort of University bureaucracy this is now taking every year uh so this is probably not news or surprising to any of you working in universities but it’s good to keep in mind uh don’t be disappointed uh you can
Try to prepare for it but that doesn’t always help uh things so in terms of a general warning developing stuff is much easier than embedding it um so despite all my protestations though about lack of automation this is actually a pretty successful project because we actually achieved our goal uh
We created um uh gdpr registrations for research and researchers hopefully don’t even notice it’s happening uh and we were actually hoping to uh replicate uh this success in a new project and this we call the RDM administrations project uh and here we aim to reuse DMP
Content in other forms such as the IT service request forms uh it uh consultation forms ethical selfcheck ethical review forms uh privacy uh related forms uh because they always have a big overlap in terms of the questions that are being asked and then researcher has to fill the right now
Researcher has to fill them each individually and repeat the same information five or six times and our aim is that uh questions in these follow-up forms are prefilled with the answers that researchers have already given in the DMP uh that is where the questions align so we hope to reduce the
Workload for the researchers um at this point we wrote a DMP for this and we are looking for an application uh that can support our goals uh and besides obviously the the this technical challenge of finding the right tool we also have you know organizational challenges I I outlined
One of them in the previous slide um but we’re also quite burden capacity wise you know the time that people can put in for this project are limited and also we are going to need the support of all the uh University faculties to eventually get on board with this project
And agree to have their you know their ethical review forms in in the tool that we end up choosing and this will also be a bit of a challenge and that’s OB obviously normal because they have to change the way that they are working and and that’s a big
Ask um we also have uh two faculties on board right now in the project while we’re still still in the development phase so that is really good and we hope that uh this can convince other faculties to get on board as well um and obviously there’s an there’s
A financial challenge in the sense that our budget obviously has limits as all budgets do so that’s it for now I think I’m doing okay for time and I will stop sharing so uh we have about we have a few minutes for Q&A if anybody has questions thank you thank you very much
Elisa yes and it’s time for questions and to those of you who asked some questions in the chat uh for the previous presentation presentations uh there are some answers there already but um if anybody have questions for Elisa then please Mara yeah thanks for this very interesting presentation I would like to
Know because I believe to know and uh but maybe the colleagues from telf they can otherwise correct me that at T Del they have created some kind of integration between DMP online and other administrative systems among others I believe pure but also regarding gdpr this gdpr registry and I I would just
Like to know um and at Mast they have created an entirely different uh solution so I I would like to know have you considered um collaboration between the the Dutch universities because uh we are all looking in awe to the Netherlands and what you accomplished but I’m seeing quite some fragmentation
And that’s of course what one usually would want to avoid or where where you think okay can we fool our um uh resources and create something shared so I I was just curious yeah great question you’re right uh in the Netherlands we’re all Reinventing the wheel at the same time
Um uh and and you know we also inventing different wheels so that’s indeed not ideal uh you I mean uh yes the the answer to your question is yes we are actually uh starting to um combine efforts uh in this uh uh in in in this area and there are
Several ironically there are now several working groups of of people combining efforts but uh we are we we are on each other’s radar uh by now yeah uh and but but uh uh for as far as I know the expectation is that this will take a few more years uh because it
Would have to you know the the a possibility is that that we would have to have something developed and and that that always takes time thank you thank you very much again Elisa for your presentation and for your answers so now it’s time for the third speaker who’s Maria Lisa who’s
Neari there I hope you can see it we can yes yeah so uh yeah so it’s it’s very nice to see you so many of you here today uh my name is mariaa I come from University library and resarch dat management team um so this is very modest example
Uh of how you can use use content of dmps um as a background information I’d like to tell you where we get the dmps we use uh of researchers or consor who has received a positive funding decision from research of finot must submit a data management plan within eight weeks
Of the funding decision and the university needs to ensure that the data management plan can be implemented at the site of research and the plan complies with good data management practices in our University the research data management team of the library is coordinating the reading and committing
Dmps of those researchers or conss who has received a positive funding decision from resarch cing of being that’s why we have some dmps to uh work with uh today I’m uh tell you about the small study we are doing based based on content of the dmps in this study we collected uh
Sorry uh in this study we concentrated to dmps from research Council of Finland we also read dmps for foru and other funders domistic or International but the research Council of Council of Finland is the bigger research funer at least if we count the number of dmbs we read uh
Uh yeah let’s see I think this is the wrong yeah yeah uh here are here you can see the number of dmps our team read during last summer we did not include all of them to study just because study was conducted before we got all of these
Dmps but we read 90 dmps for this study and why we did the study there were three main points first we are setting up the new data preservation service and we need information about preservation needs of research projects second reason was that we got new staff
Members last sprek and not all all of them were familiar with dmps the study was a way to get new employees to read to DM PE and get familiar with the content and the reason was that uh we wanted to know whether data preservation Services should be advertised to all
Researchers who received positive funding decision or only the to those who indicated a need for preservation services so targeted marke marketing tends to be more effective but how many research project who got funding does not produce research data which should uh be preserved that was our
Question uh the study was carried out by new members of our staff during the summer an AUM of 20 uh 23 study questions where what are data types mentioned in the dmps what are common file formats what are data repositor of storage services mentioned uh what are data size
Estimates and what are needs mentioned for data preservation mentioned by uh by researchers or mention it by our data support experts when reviewing to DMP uh our results then only 15 dmps of this uh 90 mentioned need for longterm preservation in this case long-term preservation means over 25 years
Preservation and 12 dmps of uh this my 190 mentioned for Middle term preservation and in this case middle term preservation time is 5 to 15 years so these are the options for preservation Services we have in our University then we sent uh those who were mentioning these needs uh uh Target
Email mentioning that they have reported the preservation needs in the dmps so we asked if they would like to discuss preservation options with us because uh we know that preservation is not something that you can do after the research project but you need to plan it carefully uh before you start collecting your
Data so we are now organizing these meetings with those who said yes and most of them uh also have some previous data sets they want to preserve uh so we have actually found a lot of more data sets that we present presumed so we got a quite a uh good
Emotion of uh yeses how however we don’t yet know exactly how many of the emails email recipients answered yet it is over 60% but how much over that we don’t know yet um our next step then uh we are planning to modify the DMP template so that we can more easily
Get information about preservation uh at this point uh at this time we needed to pick uh up this information from a free Tex fields and it was a time consuming work um and we don’t do not yet know if sending an email to all researchers who have received a positive funding
Decision is equally effective effective than this type of targeted marketing we just resum it is uh we might try to send email to all who has have re received a positive funding decision in the next funding call just to compare the results but if only about 15% of the research projects
Really think they need preservation Services PR Services it might be a good idea to try targeted marketing uh with the future CA as as well uh and at least we are quite happy with the results we got this time so that’s it from my part any questions or
Comments if that’s not the case then we will proceed to the group discussions and I’ll just share my screen for a moment we have prepared a couple of discussion points for you but you’re very welcome to add other discussion points as well uh the first one is um any ideas what information from
Dmps we should use to provide better services and the second is where or to which Services could we integrate information from dmps and uh I hope that someone will share the link to these discussion points and the document behind it in the chat uh because then you will be sent out into breakout
Rooms and I think the link is already there thank you very much so that we can proceed with the discussions there and you will have 25 minutes in the breakout room uh which means that we will meet again uh halfast 2 so it seems that we are all back in
The main room I hope that we are and I’ve shared my screen with you uh so we will go through all the groups and start with group one which was L by Meck are you also reporting from group one yeah we didn’t uh have time to discuss who we report so I start
And others can uh add if I forget something so uh we were discuss discussing about the what kind of uh information to use uh to provide better services and there was examples like estimation of data storage needs could be uh seen from the dmps and we could spot out the resarch
Project which use very sensitive data to offer them some extra support and uh then we could estimate also what other resource are needed some kind of extra help from the data support something uh and uh then you could also ask feedback about dmps at the end of
The D DMP template so that you can actually get a uh can use it as a feedback for for as uh well and uh we one discussion point was that DMP is kind of um formulating the discussion between researchers and support Personnel so it’s that kind of tool for discussing about data management
Planning and then about uh with Services we could integrate information uh we got uh uh here you can see the list here so that’s um that’s uh our thoughts about these this question thank you very much Mar Elisa are there others from the breakout room that would like to add some things
To the answers from group one no okay then we’ll proceed with group three there’s no apparently no group two so let’s take group three led by Jordan Goen and are you also the reporter we didn’t discuss that either so I’ll be I’ll report and if people in
The the group feel something wasn’t said but maybe just feel free to to tell us um so yeah we were thinking about um what the information for the gmps we could use for better better Services we were thinking of course like making dmps more operable and more machine
Actionable make them easier easier to use um that the dmps content should also be considered as data itself so to have good metadata that describes these dmps um it it can also the dmps could also help for uh developing some preservation plan um and to alert IT
Services for when to delete some data what to delete how to delete it what to destroy and to delete when there is gdpr concerns um maybe see how we can make the researcher update the DMP and how to incorporate tasks into these dmps to educate the researcher and get maybe
Some outcomes out of these the dmps um then for the second question we kind of got off got out topic a bit but it was really interesting interesting point of view so we kept on going on that on that on that way um so it’s the same thing as to which
Criteria to set on which data to preserve or not because that is a real um problem we we thought that there’s no criteria on which data we we should preserve um also on the elements of dmps um thinking about which is scientif ific Al driven so that the researchers
Could really think about these scientifical point of views and not only the administrative part of the DMP so that their data is really more interoperable in their community and so these are the aspects where the researcher should really be involved in and have some aspects where the data Librarians data stewards the IT
Department gets involved as well we should also maybe think about organizing Community discussions to agree on standards that are relevant for research domain for example so that means that are scientifically relevant the dmps could be used for different purposes as well education educating the researcher in uh research
Data management and using the DMP to get their attention but not all not as um not really just by the DMP and then it’s all over it’s really to help them get um more understanding on the data we can extend the dmps for researchers that are
Using AI tools as well so that they use them well and they are aware of what they are they are using we thought about maybe does any everything really need to be in the DMP it’s good to help the researchers think about the concerns even if everything isn’t in the
DMP uh finding elements that are scientifically relevant as is in the finished DMP template whether is data quality control even though it is really difficult to write but it’s important to add that on it and finally how can we check if a DMP is actually followed up so yeah that’s about everything I
Think we discussed thank you very much that was no problem a lot of issues you came over yeah yeah anyone else from breakout room that like to comment on this okay then we have group four led by Manas satama yep we didn’t either decide on who’s going to report
Our discussion so I might just say a few words and the others can add quite many similar things that previous groups already have discussed um we perhaps I think that what I could uh pinpoint at this place was that we discussed about the sort of the idea of
A DMP being a living document and everybody knows that nobody updates the DMP in the meanwhile so it’s just sort of a sort of a words that don’t come into flesh so we all also discussed because of this we discussed about the idea that should it be should be more talk about
Sort of a dmps in different phases of the research it’s the first phase of uh when the project is beginning so when you sort of a describe that what you are planning to how what kind of data you’re planning to have what what kind of services you you figure you need and
Then it’s the the sort of the the last DMP would be the when the project has ended and then we also discussed about the snapshot um tool or the snapshot um thing that was in the rdmo which sounds quite interesting and good um then what else did we discuss ah we
Discussed these uh RDM plugins in software that researchers already are using for in producing or handling their data processing their data so like SP SPSS and other so in the analogy of of uh citation or or that kind of plugins in that we have for example
Z and then we of course of course we ended up in in taking up the idea that we cannot solve every discipline every so every discipline uh uh in one solution that we need to have these prototypes and and then work from their own and the mapping comes into play at this this
Point I think that’s that was pretty much we discussed yeah thank you Mana comments from the breakout room no and then we have group five and we selected that a reporter and that’s Claudia F okay thank you kisten so just a brief summary um uh the group uh anytime can add
Something if I skipped important things so uh at the beginning or in the middle we we agreed that in general uh that um dmps can improve services because they um collect needs and structure the work of researchers and there is really lots of important information in in data
Management plans that’s worth reusing so now I try to align our answers a little bit to the questions I put it in italic um letter so first uh is the idea to uh enrich content uh of EG archive so there was example from Kon from the National Archive of Denmark to um
Deliver information from dmps to the archive um um targeted services so the software management is a big challenge so um dmps can also help to structure this and to think of what is needed in in regard to managing software then the the part of um the services where DPS could um help or
Where the information is is very um much um putting the the output forward this is Administration service it’s for example the GDP gdpr relevant information what is already um regarded in in dmps so this could be used um elsewhere because you have to answer those question usually very
Thoroughly then the next example is um you can use DP information for creating uh metrics and evaluation there is an example from Lin um where the colleagues are going to use dmps for an internal audit that’s um aligned along the research data life cycle and they put a a score for um the
The level of all their services so the DMP is very helpful to to go through those steps um so we uh will discuss in that it’s worth this researchers take a look back once they have finished a project and they have to ask um themselves or to have do
Kind of evaluation if they have achieved what they um was trying to do or what they promised in the um DMP I don’t know if there has been ever any kind of punishment or something else from a third party funded project if it’s not the case would be interesting to her about
That and um so as we learned from the presentations at the beginning uh there’s a need to connect those serves shared by the participants so to discuss about the tools and maybe to to share those tools and to make connections to for for um exchanging data out of dmps um into other services
And tools at the universities so there was that was output from our breakout real breakout room five so are there any questions thank you Claudia any comments or additions from the breakout room no then we will proceed with group six lit by Anna pichos thank you and I will summarize it
And please the others from the group feel free to add on uh I think many points were already mentioned in a similar way what uh are ideas were on what information that could be used to provide better Services um in general what we see in dmps is what kind
Of expertise is needed um by the researchers and that means not necessarily expertise in the way of like we need certain technical tools to to create of course those would be helpful but really personal expertise like what kind of specialist could improve um our services how could they um help
Researchers to write better dmps or have it easier in the entire funding process of creating dmps as well so mostly legal and ethical expertise seems to be missing at least in our group um also I was a moderator but also from my side I can really say that this is the same
Here so this is like a big gap that was established the need for legal expertise which is on the one hand concerning uh questions of ethics on the other hand also concerning uh compliance with the gdpr so this is something that was really seen from dmps um one possible solution uh apart
From hiring really people would be to set up data management support units at The Faculty so that would be a idea we all agreed that we want integration and basically the more the better so it’s not like uh to integrate only one thing but to have it in an ideal world
Everything under the same umbrella and interchanging information would be ideal but there is a general lack of money and personal however in this Ideal World all something we are working towards uh the services that could be integrated uh of course Legal Services if there are any
Uh at least some legal expertise it can also be just one person doesn’t need to be an entire department of legal support same for ethics um research support office is a possible candidate because depending on the institution you work at certain um questions are solved there then they data protection officers they
Could be within Legal Services they could be separate it Services patent office again could be part of legal service Etc and also not to forget State libraries I think that was really interesting because not all universities have their own Library many do but not all and this sometimes goes forgotten so
Also look outside uh of your own garden and try to incorporate Services which do cater to researchers even if they’re not in the same campus like the state’s libraries for example so those were these two main questions um what was generally um discussed apart from those two focuses
Um I already said something about yeah ethics is a big question and generally personal data needing to be uh protected um yeah DMP has a living document that was also uh you said that cloudia in your summary I think or before Mana so researchers don’t see the DMP as a
Living document they fill it out once in the beginning and then they forget about it unless uh the funer asks for an update of it so for example the Swiss National Foundation does ask for an update at the end but this is only one funer uh also in Switzerland like one
Example the rest doesn’t ask for it so really um it could be that could be a possible Improvement was also mentioned that ERC dmps are usually filled out a bit more proper because when it’s afraid of the review from the European Union but the other DMS are usually just checking
Boxes and it’s not just a fear how to fill out or not knowing how to fill it out it’s also not being able to how to formulate uh um answers for the DMP that’s a big trouble for the researchers so another possibility would be that um so in
Denmark that uh was I think an example by sarari uh they harvested nice formulations and put it all together in like in best practice kind of body of knowledge sens to fill out dmps properly so we shouldn’t forget that languages could be issue because dmps are mostly
In English so that that could be a effect um and last thing DMP trainings for students I think that was a nice example for Netherlands that um some faculties but not all have uh mandatory trainings so the the policy as it is is mandatory from the Tha University but the training is really
Individual and uh the need for that came bottom up so really the students pushed for that and wanted to have this training in order to be able to properly fill out the dmps and in general deal with uh all questions data management so it was a nice example I think of how
Even you know Grassroots action um could do something I think that would be it and please the group feel free to add on if I missed something thank you Anna and time is already four minutes to three um but before we end we have another small
Poll question for you and I think the Liber office will present it while you’re writing I want to thank you all for participating in this event and I particularly want to thank this speakers for the presentations it was indeed interesting to learn about your experiences in this
Matter and uh thanks also to all of you who participated in the breakout rooms and joined the discussion I think it it’s really fruitful to have all this input from all of you so thank you very much and I hope to see you again at other events Leo events