D4.2 An approach to steer the behaviour of non-government data holders towards open data through a technical strategy

Towa ds a sus ainable Open Da a ECOsys em
D4.2
An app oach o s ee he beha iou o non-go e nmen
da a holde s owa ds open da a h ough a echnical
s a egy
This p ojec has ecei ed unding om he Eu opean Union’s Ho izon 2020 esea ch and inno a ion
p og amme unde he Ma ie Skłodowska-Cu ie g an ag eemen No 955569.
The opinions exp essed in his documen e lec only he au ho ’s iew and in no way e lec he
Eu opean Commission’s opinions. The Eu opean Commission is no esponsible o any use ha may
be made o he in o ma ion i con ains.
An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
P ojec Ac onym
ODECO
P ojec Ti le
Towa ds a sus ainable Open Da a ECOsys em
G an Ag eemen No.
955569
S a da e o P ojec
01-10-2021
Du a ion o he P ojec
48 mon hs
Deli e able Numbe
D4.2
Deli e able Ti le
An app oach o s ee he beha iou o non-
go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
Dissemina ion Le el
Public
Deli e able Leade
Uni e si y o he Aegean (UAEGEAN)
Submission Da e
22-10-2024
Au ho (s)
Cha alampos Alexopoulos - UAEGEAN, Ma ia
Ioanna Ma a si - UAEGEAN, Mohsan Ali –
UAEGEAN, Geo gios Papageo giou - Fa osne
S.A., Abdul Aziz – UNIZAR, Dagobe o He e a
Mu illo - UNIZAR
Documen his o y
Ve sion
#
Da e
Desc ip ion (Sec ion, page
numbe )
Au ho & O ganisa ion
V0.0
06-06-2024
Kick-o Mee ing
Cha alampos Alexopoulos -
UAEGEAN
V0.1
05-06-2024
Fi s d a able o con en s
Cha alampos Alexopoulos -
UAEGEAN, Ma ia Ioanna Ma a si -
UAEGEAN
V0.1a
06-06-2024
Re iew o able o con en
And ea Polini - UNICAM,
Cha alampos Alexopoulos -
UAEGEAN, Ma ia Ioanna Ma a si -
UAEGEAN, Mohsan Ali –
UAEGEAN, Geo gios
Papageo giou - Fa osne S.A.,
Abdul Aziz – UNIZAR, Dagobe o
He e a Mu illo - UNIZAR
V0.2
20-08-2024
Fi s d a
Cha alampos Alexopoulos -
UAEGEAN, Ma ia Ioanna Ma a si -
UAEGEAN, Mohsan Ali –
UAEGEAN, Geo gios
Papageo giou - Fa osne S.A.,
Abdul Aziz – UNIZAR, Dagobe o
He e a Mu illo - UNIZAR
V0.2a
13-09-2024
Second d a . Res uc u a ion
a e he eedback om he
i s pee - e iew.
Cha alampos Alexopoulos -
UAEGEAN, Ma ia Ioanna Ma a si -
UAEGEAN, Mohsan Ali –
UAEGEAN, Geo gios
Papageo giou - Fa osne S.A.,
Abdul Aziz – UNIZAR, Dagobe o
He e a Mu illo – UNIZAR
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
Ve sion
#
Da e
Desc ip ion (Sec ion, page
numbe )
Au ho & O ganisa ion
Re iewe s: Ja ie Nogue as -
UNIZAR, F ancisco Ja ie López
Pellice - UNIZAR
V0.3
18-10-2024
Final epo . Edi o ial wo k
a e he eedback om he
second pee - e iew
Re iewe s: Bas iaan an Loenen -
TUDELFT, Ja ie Nogue as,
UNIZAR, F ancisco Ja ie López
Pellice - UNIZAR, Bi ge La sen -
AAU
V1.0
22-10-2024
Final edi ing.
Dani sja an Heusden- an Winden
– TUDELFT
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
Table o Con en s
Abb e ia ions ..................................................................................................................................................................... 7
1. In oduc ion ............................................................................................................................................................... 9
1.1 P oblem De ini ion .................................................................................................................................... 9
1.2 Role o his deli e able in he ODECO p ojec .............................................................................10
1.3 S uc u e ......................................................................................................................................................10
2. Me hodology ..........................................................................................................................................................11
3. Ba ie s and Mo i a ions o Da a Sha ing o Non-Go e nmen al Da a Holde G oups ........13
3.1 Non-Specialis Ci izens ..........................................................................................................................13
3.2 Da a Jou nalis s .........................................................................................................................................13
3.3 Elemen a y School S uden s ................................................................................................................15
3.4 Non-Go e nmen al O ganisa ions ....................................................................................................16
3.5 Comme cial O ganisa ions ...................................................................................................................17
3.6 Open Da a In e media ies ....................................................................................................................18
4. Technical Requi emen s o Non-Go e nmen Da a Holde G oups ...............................................20
4.1 Open Da a Ac ion P inciples ................................................................................................................20
4.2 Technical Requi emen s o OD Holde s and Use s ...................................................................21
4.3 Syn hesis ......................................................................................................................................................24
5. Challenges o OD C ea ion and Sha ing ....................................................................................................28
5.1 Enhanced da a c ea ion p ocess h ough be e quali y, in eg a ion and alida ion ools
28
5.2 Enhanced da a sha ing h ough be e licensing and anonymisa ion ................................29
5.3 Towa ds inclusi e, use -d i en and ci cula open da a pla o ms h ough be e
in e aces and eedback mechanisms ................................................................................................................29
6. Technical S ee ing Mechanisms pe Challenge ........................................................................................30
6.1 Exis ing ele an solu ions ....................................................................................................................30
6.1.1 Gene al oolki s o open da a sha ing ......................................................................................30
6.1.2 In eg a ion o non-go e nmen da a on open da a eposi o ies ...................................33
6.1.3 Open da a gene a ed by olun ee s h ough mic o asking: he case o he HOT
Tasking Manage ...................................................................................................................................................35
6.2 P oposed echnical solu ions wi hin he ODECO P ojec .........................................................38
6.2.1 P oposed solu ions o Challenge 1 - Enhanced da a c ea ion p ocess h ough be e
quali y, in eg a ion and alida ion ools ......................................................................................................38
6.2.2 P oposed solu ions o Challenge 2 - Enhanced da a sha ing h ough be e
licensing and anonymisa ion ............................................................................................................................47
6.2.3 P oposed solu ions o Challenge 3 - Towa ds inclusi e, use -d i en and ci cula
open da a pla o ms h ough be e in e aces and eedback mechanisms .................................50
6.3 Technology Readinesso he P oposed Solu ions .......................................................................58
7 Discussion and Conclusion ................................................................................................................................59
7.1 Sha ed pe spec i es ................................................................................................................................59
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
7.2 Limi a ions ...................................................................................................................................................60
7.3 Fu u e esea ch di ec ions and ac ions ...........................................................................................60
Acknowledgemen s .......................................................................................................................................................61
Appendix A – The ODECO TW5 Idea hon S uc u e ........................................................................................63
Appendix B – Use -cen ed Design .........................................................................................................................65
Re e ences .........................................................................................................................................................................68
Lis o igu es
Figu e 1: The me hodological p ocess o Task 4.2 ............................................................................................11
Figu e 2: Main componen s o Epicollec 5 ...........................................................................................................31
Figu e 3: Key a ibu es o QField. ...........................................................................................................................32
Figu e 4: Sc eensho o he p ocess o inges ing indi idual da ase s in o he F ench open da a
po al ...................................................................................................................................................................................34
Figu e 5: Sc eensho o he dashboa d o he da ase s inges ed by a use on he F ench open da a
po al ...................................................................................................................................................................................34
Figu e 6: The p ojec asks sc een in HOT-TM displays key in o ma ion in wo sec ions. On he le
sideba , use s can ind p ojec de ails, ask s a uses in lis o m, s ep-by-s ep ask ins uc ions, and
con ibu o s a is ics......................................................................................................................................................35
Figu e 7: Sc eensho o he mapping phase in HOT-TM. Sou ce:
h ps:// asks.ho osm.o g/p ojec s/15476/ asks .................................................................................................36
Figu e 8: HOT-TM API ha allows access o da a on campaigns, p ojec s, o ganiza ions, asks,
commen s, among o he s. Sou ce: h ps:// asks.ho osm.o g/api-docs ...................................................37
Figu e 9: A simple example o di e en da a schema use in a Cul u al He i age case ......................39
Figu e 10: The p oposed implemen a ion p ocess ...........................................................................................40
Figu e 11: Indica i e sc een p o o ype o he mapping ecommenda ion upon da ase upload..41
Figu e 12: P oposed me hodology o he au oma ed hema ic anno a ion o open da ase s .....42
Figu e 13: BRYT Keywo d Ex ac ion A chi ec u e .............................................................................................43
Figu e 14: Example Scena io o he Applica ion ...............................................................................................44
Figu e 15: Minimum iable p oduc o Applica ion unc ionali ies and wo king ................................44
Figu e 16: The echnology s ack equi ed o de elop he Applica ion .....................................................45
Figu e 17: Building blocks o he Open Da a D i en Ci izen Science Applica ions gene a ion model
...............................................................................................................................................................................................46
Figu e 18: The de elopmen p ocess o he p oposed pla o m ................................................................48
Figu e 19: Key componen s o he sys em ...........................................................................................................49
Figu e 20: Th ee-pa ame o he p oposed use in e ace ........................................................................49
Figu e 21: The de elopmen p ocess o he p oposed pla o m ................................................................50
Figu e 22: An example o a Cha bo assis an on he G eek open da a po al .....................................51
Figu e 23: Channel 1 o a mul i-channel inclusi e eedback mechanism o di e se communi ies
...............................................................................................................................................................................................55
Figu e 24: Channel 2 o a mul i-channel inclusi e eedback mechanism o di e se communi ies
...............................................................................................................................................................................................55
Figu e 25: Example case o in e ac ion wi h a da ase ....................................................................................56
Figu e 26: An example o a use in e ac ion wi h a da ase on da a.gou . ..........................................57
Figu e 27: The use -cen ed design p ocess based on ISO 9241-210:2010 ...........................................65
Lis o ables
Table 1: Ba ie s o non-expe ci izens’ con ibu ion o OD ecosys ems ...............................................13
Table 2: Mo i a ions o non-expe ci izens o con ibu e o OD ecosys ems.....................................13
Table 3: Iden i ied ba ie s ..........................................................................................................................................17

D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
Table 4: Mo i a ions o NPOs o con ibu e open da a ................................................................................17
Table 5: Ba ie s o comme cial o ganisa ions o con ibu e open da a ..................................................18
Table 6: Mo i a ions o comme cial o ganisa ions o con ibu e open da a .......................................18
Table 7: Ba ie s o open da a in e media ies o con ibu e open da a .................................................19
Table 8: Mo i a ions o open da a in e media ies o con ibu e open da a .........................................19
Table 9 - Technical Requi emen s o NGO Open Da a use s ( aken om ODECO deli e able D2.2)
...............................................................................................................................................................................................22
Table 10: Technical equi emen s, ac ion p inciples, ba ie s, and mo i a ions pe Open Da a
Holde G oup ...................................................................................................................................................................25
Table 11: E alua ion o Open Da a Tools o he Key Challenges...............................................................32
Table 12: Technology Readiness o he P oposed Solu ion ..........................................................................58
Table 13 - GenAI ools o use in e ace design ...............................................................................................67
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
7
Abb e ia ions
AI A i icial In elligence
API Applica ion P og amming In e ace
AR Augmen ed Reali y
D Deli e able
ESR Ea ly-S age Resea che
GDPR Gene al Da a P o ec ion Regula ion
GenAI Gene a i e A i icial In elligence
HOT Humani a ian OpenS ee Map Team
HOT-TM Humani a ian OpenS ee Map Team Tasking Manage
IDIP In elligen Da a In e ac ion Pla o m
IOT In e ne o Things
LLM La ge Language Model
LODC Linked Open Da a Cloud
MS Miles one
MVP Minimum Viable P oduc
NGD Non-Go e nmen Da a
NGO Non-Go e nmen al O ganisa ion
NLP Na u al Language P ocessing
NPO Non-P o i O ganisa ion
OD Open Da a
ODP Open Da a Po al
ODK Open Da a Ki
ODECO Open Da a ECOsys em
OSM OpenS ee Map
PII Pe sonally Iden i iable In o ma ion
RAG Re ie al-augmen ed gene a ion
TAM Technology Accep ance Model
TRL Technology Readiness Le els
TSLO Technical, Seman ic, Legal, and O ganiza ional
VR Vi ual Reali y
TW T aining Week
UCD Use -cen ed design
VGI Volun a y Geog aphic In o ma ion
WAI Web Accessibili y Ini ia i e
WP Wo k Package
N
Pa ne
Pa ne sho
name
Coun y
Bene icia y
1
Technische Uni e si ei Del
TU Del
Ne he lands
2
Ka holieke Uni e si ei Leu en
KUL
Belgium
3
Cen e Na ional de la Reche che Scien i ique
CNRS
F ance
4
Uni e sidad de Za agoza
UNIZAR
Spain
5
Panepis imio Aigaiou
UAEGEAN
G eece
6
Aalbo g Uni e si e
AAU
Denma k
7
Uni e si à degli S udi di Came ino
UNICAM
I aly
8
Fa osne S.A.
FAROSNET S.A.
G eece
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
8
N
Pa ne
Pa ne sho
name
Coun y
Pa ne o ganisa ions
1
7eDa a
7EDATA
Spain
2
Digi aal Vlaande en
DV
Belgium
3
Ci y o Copenhagen
COP
Denma k
4
Ci y o Ro e dam
RDAM
Ne he lands
5
CoC Play ul Minds
CoC
Denma k
6
De ilinx
DERI
I eland
7
ESRI
ESRI
Ne he lands
8
Maggioli S.p.A
MAG
I aly
9
Na ional Cen e o Geog aphic In o ma ion
CNIG
Spain
10
Open Knowledge Belgium
OKB
Belgium
11
SWECO
SWECO
Ne he lands
12
The go e nmen lab
GLAB
Uni ed S a es o Ame ica
13
Agency o Da a Supply and In as uc u e
ADSI
Denma k
14
GFOSS Open Technologies Alliance
GFOSS
G eece
15
Inno3 Consul ing
IC
F ance
16
Regione Ma che
RM
I aly
17
Open Da a Ins i u e
OCI
Uni ed Kingdom
18
Swedish Na ional A chi es (Riksa ki e )
SwNA
Sweden
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
9
1. In oduc ion
1.1 P oblem De ini ion
Open Da a ini ia i es o en ocus on a na ow selec ion o p o ide s and use g oups while
neglec ing o he s akeholde s in he ecosys em. This exclusion is e iden in he dominance o
go e nmen al ac o s in p o iding open da a. Sus ainable open da a ecosys ems ad oca e o
inclusi i y, sugges ing ha non-go e nmen ac o s should also sha e hei da a in open o ma s
(Cha alabidis e al., 2018, Runeson e al., 2021). Howe e , encou aging non-go e nmen ac o s o
sha e hei da a as open da a emains a la gely unexplo ed challenge. The ODECO p ojec aims o
add ess his issue h ough a mul idisciplina y app oach ha includes: (1) unde s anding he
ba ie s and mo i a ions o non-go e nmen ac o s o con ibu e open da a o he ecosys em,
(2) de eloping he echnical mechanisms o acili a e his p ocess, and (3) es ablishing go e nance
s uc u es o suppo and sus ain i .
Speci ically, Task 4.2 examines he echnical mechanisms ha can be used o encou age non-
go e nmen da a holde s o sha e open da a. Task 4.2 conside s he ba ie s and mo i a ions o
ask 4.1, and echnical equi emen s o non-go e nmen ac o s o become ac i e con ibu o s o
he open da a ecosys em (by eleasing hei da a in o he open da a ecosys em) and p oposes
he co esponding echnical s ee ing mechanisms o sui ably p epa e he g ound owa ds his
di ec ion. The echnical s ee ing mechanisms p esen ed in his Deli e able a e o a ying
Technology Readiness Le els (TRL), hose de eloped wi hin he ODECO p ojec being o highe
TRL.
Go e nmen al ac o s ha e his o ically been he dominan p o ide s o open da a. In many cases,
such esponsibili y is eng a ed in policy documen s, om non-binding s a egic plans o binding
laws and egula ions such as he EU Open Da a Di ec i e. On he o he hand, while he e ha e
been some ini ia i es by non-go e nmen al ac o s o publish hei da a as open da a, i is s ill a
limi ed p ac ice and emains la gely desi ed ( an Loenen e al., 2018). Non-go e nmen al ac o s
a e ac o s ha a e ou side he public/go e nmen al sec o , such as companies, ci il socie y
o ganisa ions, and he media. Explo ing he mo i a ions and challenges o non-go e nmen al
ac o s o con ibu e open da a is hus essen ial o de elop sus ainable open da a ecosys ems ha
inco po a e bo h go e nmen and non-go e nmen open da a.
Aligned wi h he ODECO’s Desc ip ion o Ac ion, his deli e able aims o answe he ollowing
main objec i e:
• Iden i y and u he de elop echnical mechanisms (solu ions) o acili a e OD sha ing
by non-go e nmen da a holde g oups.
These solu ions will en ail quali y assu ance me hods, he de elopmen (o embedmen ) o
da a in eg a o se ices, as well as new use (da a holde ) in e aces om a leas he wo king
p o o ypes o he engaged ESRs. These solu ions will help he non-go e nmen al da a holde s
o mo e easily c ea e and sha e hei da a as open da a using hese no el and au oma ed
solu ions. Fi s and o emos , he de eloped solu ions could be embedded o he al eady
exis ing open da a spaces (e.g. Zenodo, OpenAIRE, wikida a e c.) ha enable he upload o
da ase s om any non-go e nmen al s akeholde . Secondly, hey could be used owa ds he
( u he ) de elopmen o hei own open da a po als (o websi es sha ing open da a) wi hin
hei o ganisa ions.
In o de o add ess he main objec i e, h ee sub-objec i es a e necessa y:
• Sub-objec i e 1: Analyse he iden i ied mo i a ions and echnical equi emen s o non-
go e nmen al da a holde s o con ibu e open da a. The iden i ied mo i a ions a e b ie ly
p esen ed in sec ion 3 o his deli e able and in mo e de ail in D4.1.
• Sub-objec i e 2: Analyse he echnical equi emen s and ba ie s which hinde non-
go e nmen al da a holde s o con ibu e open da a. The de eloped and p oposed echnical
mechanisms should no dis ega d he equi emen s o he da a use s o build “use -d i en”
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
16
Mo i a ions
• Being ac i e ci izens. Celis Va gas e al. (2023) ha e iden i ied ha specially OD lea ning
ac i i ies seeking he de elopmen o compe encies o ac i e ci izenship add ess ac i i ies
o collec ion o own da a. In hose cases, s uden s ha e been in ol ed in bo h c ea ing simple
sp eadshee s and collec ing mo e complex da a by using ools such as senso s, games, o
mobile applica ions (Badioze Zaman e al., 2021; Chicaiza e al., 2017; Saddiqa, La sen, e al.,
2019; Vallejo-Figue oa e al., 2018). The mo i a ion is ac i ely pa icipa ing as ci izens o c ea e
a be e wo ld. Fo example, s uden s du ing he ocus in e iews wonde ed abou hei school
p ojec s "How is his going o c ea e a be e wo ld?".
• Raising awa eness o local issues a ound s uden s' con ex and daily li e. Conside ing
s uden s as expe s o hei own local expe ience, hey can po en ially c ea e and sha e local
da ase s add essing aspec s o hei en i onmen and daily li e expe iences. F om hei
pe spec i e, hey wan o aise awa eness and p o ide con ex ual unde s anding o local
issues. Thei mo i a ion behind is aising hei oice and being hea d. Making "Child en's oice
as impo an as o he s".
• Helping he communi y a ound he school, conside ing s uden s as impo an ac o s in
local ecosys ems. S uden s a e mo i a ed by helping he communi y a ound hem, implici ly
o a eeling o belonging, building hei iden i y and hei place. C ea ing and sha ing da a
has been iden i ied as an oppo uni y o also building ne wo ks in hei local communi ies by
add essing p oblems om o he ac o s and con ibu ing o sol e hem wi h da a.
• Seeing wha s uden s lea n in schools as use ul in he eal wo ld. Fo young pupils in
elemen a y school, i was ele an o see wha hey do in school being used in he eal wo ld.
I inc eases he au hen ici y o hei lea ning expe ience.
• Making school ac i i ies mo e ele an , in e es ing, and un. To sum up, c ea ing and
sha ing hei da a migh inc eases hei mo i a ion o lea ning by ul illing hei in insic
mo i a ion o making some hing ele an , being hea d and connec ing o hei communi ies.
O e all, s uden s a e mo i a ed by ac i e lea ning expe iences, we e hey ha e eedom o
expe imen and lea n by hemsel es.
3.4 Non-Go e nmen al O ganisa ions
Non-Go e nmen al O ganisa ions (NGOs), also in e changeably called Non-P o i O ganisa ions
(NPOs) in his sec ion, ake up an in e media y ole in he open da a ecosys em, whe e hey b idge
he gap be ween open da a p o ide s and use s (González-Zapa a and Heeks, 2015). NPOs a e
unique as in e media ies because he e a e speci ic use communi ies hey a e ocusing on o
add ess a social issue (Enaholo, 2017) while also no seeking o gain any p o i s om i (Salamon
and Anheie , 1992). As such, NGOs can be seen as g oups o indi idual use s ha a e wi h one
mo e le el o explainabili y and echnical assis ance in he use o a ailable ools. The di e ence
wi h non-specialis ci izens is ha he se ices and guidelines could be d a ed in a mo e echnical
o ma . I he aim o NPO is o imp o e o e all openness and anspa ency, i can push hem o
aim o a a ie y o p ojec s and ha e open da a and open sou ce on ha p inciple (Baack, 2015).
Howe e , NGOs may ace ba ie s ha p e en hem om con ibu ing (Cha apadhyay, 2014). The
iden i ied ba ie s and mo i a ions a e p esen ed in Table 3 and Table 4 espec i ely.
The ba ie s ound a e summa ised in Table 3 ; he e a e wo ba ie s o he mo i a ions ha s op
NGOs om sha ing he da a as open: (1) he lack o inancial esou ces, limi s NGOs abili y o
pu sue open da a p ojec s, (2) he lack o knowledge and echnical skills o NGOs’ employees,
a ec ing he NGOs capaci y o co ec ly license and publish da ase , and (3) he lack o an exis ing
common po al o NGOs o sha e hei da a.

D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
17
Table 3: Iden i ied ba ie s
Ba ie s
Desc ip ion
The lack o inancial
esou ces
The na u e o non-p o i o ganisa ional model means ha
NGOs need o secu e unding in a o m o g an s o indi idual
dona ions. Some NGOs may no be able o ob ain enough
unding o hei open da a p ojec s.
The lack o knowledge and
echnical skills o NGOs’
employees
The lack o knowledge and echnical skills o NGOs’ employees
p e en s hem om publishing he da a hey ha e as open, o
hey do no know how o p o ide i p ope ly licensed as open.
NGOs ha a e sho on hei inancial esou ces a e unable o
hi e addi ional employees wi h he equi ed skills.
The lack o an exis ing
common po al o NGOs o
sha e hei da a
NGOs canno add o he open go e nmen da a po als and
he e is no exis ing common po al o NGOs and ci il socie ies.
Howe e , c ea ing such a pla o m and/o main aining i would
equi e a lo o inancial and human esou ces ha he a e age
NGO canno a o d.
Table 4: Mo i a ions o NPOs o con ibu e open da a
Mo i a ion
Desc ip ion
Show he powe o open da a
C ea ing a p ojec wi h enhanced o eused open da a
can engage o he s akeholde s o make hem mo e
awa e o he open da a a ailable.
Help wi h he local and global socie al
issues
The e is open go e nmen da a ha migh be
a ailable on socie al issues, bu no u ilised. By inding
ways o enhance he da a and publish i as open, he
NPO can highligh i and help hose a ec ed by he
issue.
Recei e posi i e eedback om he
communi y
The communi y can gi e eedback on he issues wi h
he da a and gi e sugges ions when hey a e
in e es ed and in ol ed which mo i a es NGOs o
con inue he p ojec .
Follow o ganisa ional goals and he
pe sonal belie s o he employees in
openness and anspa ency
Some NPOs ha e anspa ency and openness o da a
and
knowledge as hei main goals and employees
join NPOs because hey sha e he ision.
C ea e oppo uni ies o o he
s akeholde s
NGOs wan hei da a o be eused widely by and
bene i a a ie y o communi ies, NGOs, p i a e and
go e nmen al o ganisa ions.
3.5 Comme cial O ganisa ions
Comme cial o ganisa ions a e de ined as hose whose goal is o make an economic p o i . This is
illus a ed in con as o use s who in end non-comme cial use, which C ea i e Commons
(C ea i e Commons, 2023) de ines as “means no p ima ily in ended o o di ec ed owa ds
comme cial ad an age o mone a y compensa ion.” Comme cial o ganisa ions may ha e di e en
mo i a ions o con ibu e o OD ecosys ems, depending on hei ield and own necessi ies. The
iden i ied ba ie s and mo i a ions o comme cial o ganisa ions o sha e da a a e p esen ed in
Table 5 and Table 6 espec i ely.
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
18
Table 5: Ba ie s o comme cial o ganisa ions o con ibu e open da a
Ba ie
Desc ip ion
Technical/ Tools
Lack o da a in eg a ion ools, and ools o
impo ing la ge da ase s.
License compa ibili y
Da ase s gene a ed by comme cial o ganisa ions
a e some imes a combined p oduc o di e en
sou ces, o which he license may no be
compa ible.
Insu icien esou ces [SMEs]
Limi ed money, labou and ime, compa ed o
la ge o ganisa ions.
Resis ance by o he communi y membe s
[Big co po a ions]
Resis ance by communi y membe s as hey ea
big co po a ions may domina e he p ojec
landscape.
Table 6: Mo i a ions o comme cial o ganisa ions o con ibu e open da a
Mo i a ion
Desc ip ion
Imp o ing he da a quali y
By imp o ing he da a quali y o ganisa ions can
deli e be e se ices o hei cus ome s.
Communi y building
Some o ganisa ions a e in e es ed in each opic
o domain. They kicks a communi ies in ha
opic o domain o main ain da a con inui y and
eshness.
S anda diza ion and alignmen wi h an
open p ojec schema
Fo some o ganisa ions, his is a mo i a ion o
con ibu e da a o open p ojec s
, o lowe he
subjec i i y o he da a.
Being g a e ul and wan ing o gi e back
Some o ganisa ions eel ha open p ojec s ha e
p o ided alue o hem and hei p oduc s, and
eel i is only ai o gi e back and p o ide alue
o he p ojec , o keep i going.
3.6 Open Da a In e media ies
Open da a in e media ies a e de ined as “ hi d-pa y ac o s who p o ide specialised esou ces and
capabili ies o (i) enhance he supply, low, and/o use o open da a and/o (ii) s eng hen he
ela ionships among a ious open da a s akeholde s” (Shaha udin e al., 2023). Examples a e
de elope s who p ocess and include open da a in apps/so wa e, c owdsou cing pla o ms ha
ga he and publish da a as open da a, and o ganisa ions ha ans o m open da a in o easily
diges ible in o ma ion such as isual o ms. Open Da a In e media ies ca y ou a wide ange o
asks depending on hei specialised esou ces and capabili ies such as compiling da a, alida ing
da a, and imp o ing echnical openness o da a (Shaha udin e al., 2023). Thus, h ough some o
hei asks, some open da a in e media ies ( e-)p oduce da a ha could be con ibu ed back o
he open da a ecosys em.
As all ac o s in he open da a ecosys em, including open da a in e media ies, ha e hei own
agency, and hus sel -in e es ed (Da ies, 2011; Poikola e al., 2010), he mo i a ions o hem o
con ibu e open da a na u ally ha e o be aligned wi h hei in e es s. In o he wo ds, open da a
in e media ies ei he mus be con inced ha hey can di ec ly o indi ec ly cap u e alue o
hemsel es by publishing open da a (in insic mo i a ion) o hey mus be o ced by ex e nal
condi ions such as h ough law and egula ions o do so (ex insic mo i a ion). The iden i ied
ba ie s and mo i a ions o open da a in e media ies a e p esen ed in Table 7 and Table 8
espec i ely.
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
19
Table 7: Ba ie s o open da a in e media ies o con ibu e open da a
Ba ie s
Desc ip ion
P o ec ing business in e es s
Some open da a in e media ies a e hesi an o
elease (some o ) hei da a as open da a because
hey wan o p o ec hei business in e es s.
No compelling legisla ion
The e a e limi ed o no legal equi emen s ha
compel open da a in e media ies o p o ide open
da a.
Addi ional cos s o de elop and
main ain open da a pla o ms
De eloping and main aining da a pla o ms o
elease open da a would incu addi ional cos s o
open da a in e media ies.
Table 8: Mo i a ions o open da a in e media ies o con ibu e open da a
Mo i a ion
Desc ip ion
Suppo he isibili y o hei
o ganiza ions
By p o iding open da a, open da a in e media ies can
inc ease he isibili y o hei o ganiza ion; hence, i is a
o m o ma ke ing o hei p oduc s and se ices.
Suppo o he pa ne s wi hin
hei ne wo ks
Some open da a in e media ies p o ide open da a o
suppo he business o ope a ion o hei pa ne s, which
hey would also ge he bene i om. This may u he
s eng hen hei posi ion wi hin hei ne wo k.
The desi e o con ibu e o
socie y
Some open da a in e media ies a e d i en by philan h opic
o al uis ic mo i a ions o con ibu e open da a ha could
bene i socie y.
The a ailabili y o open da a
pla o ms o sha e open da a
Open da a in e media ies could be mo e mo i a ed o
elease open da a i he e a e open da a pla o ms ha
could acili a e hem o do so. This is because no all open
da a in e media ies ha e he capabili y and esou ces o
de elop and main ain hei own da a pla o ms.
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
20
4. Technical Requi emen s o Non-Go e nmen Da a Holde
G oups
Fo many non-go e nmen ac o s s udied, mo i a ions ela ed o own bene i we e iden i ied in
D4.1. Such mo i a ions include suppo ing o he pa ne s wi hin an o ganiza ion ne wo k o hei
own bene i , he p i a e alue o con ibu ing, he eeling o belonging, and enjoymen .
Aligning hese own bene i mo i a ions o he di e en s akeholde s is key o achie ing he goal
o open da a sha ing. These own bene i s ma ch wi h he in insic mo i a ions om he Sel
De e mina ion Theo y de ined in Task 3.3 (D3.3). Own bene i mo i a ions a e suppo ed by he
desi e o c ea e an impac , men ioned in he clus e s o suppo ing o he pa ne s, suppo ing he
communi y, and c ea ing social impac . These can be classi ied as ex insic mo i a ions in he Sel
De e mina ion Theo y.
Finally, he a ailabili y o and desi e o imp o e he echnical skills and solu ions a e men ioned
as mo i a ions, and po en ial mo i a ions. This means ha c ea ing he co ec echnical
en i onmen can be an enable o open da a sha ing, as well as including he po en ial
s akeholde s in he p ocess, wi h he desi e o imp o e he echnical en i onmen .
4.1 Open Da a Ac ion P inciples
Based on he indings o ODECO Task 3.3, whe e he commons-based go e nance model o he
open da a ecosys em has been e ined, six ac ion p inciples ha e been un eiled ela ed o
bounda y-making, communi y suppo , pa icipa ion and decision-making, legal mechanisms,
in e ope abili y and in e ope able p ojec s, and sus ainabili y. In ou analysis, we ha e ocused on
he ollowing p inciples which a e he mos ele an o he non-go e nmen al da a holde s:
1. Encou aging pa icipa ion and sha ed decision-making,
2. Conside ing app op ia e legal mechanisms,
3. Designing an ecology o in e ope able p ojec s, and
4. Ensu ing sus ainabili y o open da a ecosys ems.
The aim is o ex ac he main key poin s and p opose he de elopmen and de elop echnical
solu ions ha can mo i a e each g oup o be an ac i e pa icipan in he open da a ecosys em.
Encou aging pa icipa ion and sha ed decision-making
The key poin o his ac ion p inciple is he oppo uni y o less powe ul use s o he open da a
ecosys em (pa icula ly he ones ha in a hie a chical sys em ha e no con ol o e he open da a)
o aise conce ns, and needs, and p o ide eedback. I is impo an o men ion ha o public and
non-p o i o ganiza ions, echnical solu ions may mo i a e hem o gi e oice o ci izens and
unde ep esen ed g oups, bu o comme cial use s (in e media ies, companies, e c.), hey mus
ha e p o i as an incen i e o o he ex insic mo i a ions (laws, egula ions, e c.).
Jou nalis s, NGOs, local go e nmen s, and open da a in e media ies can all bene i om
collabo a i e pla o ms ha enable public con ibu ion and eedback on open da a. Fo
jou nalis s, such pla o ms o e aluable c owdsou ced insigh s, hough obus mode a ion is
c ucial o main ain da a in eg i y. NGOs equi e simpli ied, low-main enance eedback
mechanisms o e icien ly ga he inpu wi hou s aining esou ces. Local go e nmen s can
le e age ci izen eedback o imp o e se ices and decision-making. Open da a in e media ies play
a key ole by p o iding echnical solu ions o hese eedback sys ems, b idging he gap be ween
da a p o ide s and use s. While s uden s may no di ec ly implemen echnical solu ions, hey can
d aw inspi a ion om hese key poin s o unde s and he impo ance o public engagemen in
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
21
open da a ecosys ems. Ul ima ely, a well-designed eedback mechanism can enhance da a quali y,
inc ease anspa ency, and os e g ea e ci ic pa icipa ion ac oss a ious sec o s.
Conside ing app op ia e legal mechanisms
Two in eg al issues we e unco e ed in his legal mechanism ac ion p inciple. The p edominan
one e ol es a ound he lack o da a li e acy and he impac ha legal mechanisms can ha e,
especially in go e nmen s uc u es. The second impedimen is he disco e y and adop ion o
sus ainable business models o in e media ies and comme cial use s. This obs acle in ol es
inding economically iable ways o hese en i ies o ope a e while adhe ing o open da a
p inciples and legal equi emen s.
The p ima y challenge aced by all use s is he equi emen o compa ible licenses. A po en ial
solu ion o his issue could be he de elopmen o an on ological model o open da a licensing.
This model could be applied o open da a se s, he eby au oma ing he p ocess o checking
compa ibili y be ween di e en licenses.
Designing an ecology o in e ope able p ojec s
The in eg al ocus o his ac ion p inciple is he adop ion o in e ope able s anda ds be ween
da ase s so hey can be exchanged, combined, and analysed e en when hey a e sou ced by
di e en o ganiza ions. This ini ia i e can boos he use o open da a ac oss a ious use ypes as
i will signi ican ly educe he echnical skills and expe ise cu en ly equi ed o use s o
e ec i ely u ilize he open da a hey can acqui e and use. The mos c ucial p oblems men ioned
in his ac ion poin in D3.3 a e, as al eady no ed, he lack o pe sonnel wi h he app op ia e skill
se and knowledge o u ilize open da a. Secondly, he e a e conce ns among use s ega ding
po en ial secu i y and p i acy issues ha may a ise om combining di e en da ase s, which could
indi ec ly expose da a ha was obscu ed h ough anonymiza ion echniques in he o iginal
da ase s.
Ensu ing sus ainabili y o open da a ecosys ems
The p ima y emphasis o his ac ion poin is o ensu e he sus ainabili y o he open da a
ecosys em and he con inued pa icipa ion o use s wi hin i . This mainly in ol es he s uggle o
unds and esou ces, pa icula ly o non-p o i o ganiza ions. The ocus is on ad ocacy o public
unds, while o comme cial use s, he emphasis is on he iabili y o business models ha cen e
on ac i e pa icipa ion in he open da a ecosys em.
4.2 Technical Requi emen s o OD Holde s and Use s
In his Sec ion, he echnical equi emen s o Open Da a holde s we e mainly de i ed om
ODECO Task 2.2 (D.2.2) and we e mapped a e wa ds wi h he da a holde g oup hey conce n.
Analysing he equi emen s ega ding hei impo ance (numbe o NGD use g oup epo ing
his equi emen ) and connec ion o he NGD holde s (au ho s’ in e p e a ion), he mos ele an
equi emen s de i ed om D2.2 a e hose ela ed o he Reusabili y and In e ope abili y aspec s.
They a e mapped ega ding hei espec i e s akeholde g oup shown in Table 9. Since, NGD
holde s canno sha e da a h ough he open go e nmen da a po als, we speci ically men ion he
al eady exis ing open da a spaces (e.g. Zenodo, OpenAIRE, wikida a e c.) ha enable he upload
o da ase s om any non-go e nmen al s akeholde . In his case, he a ailable ools and egis ies
o NGD holde s should comply o he FAIR p inciples.
As al eady men ioned, he de eloped and p oposed echnical mechanisms should no dis ega d
he equi emen s o he da a use s in o de o build “use -d i en” solu ions. Fu he mo e, o
achie e ci cula i y in he open da a ecosys em, we need o conside he dual ole o he non-
go e nmen al s akeholde s o he ecosys em se ing bo h as use s and p o ide s o open da a.
This means ha speci ic open da a use equi emen s a e also impo an o he de elopmen o
echnical mechanism o s imula e NGD Holde s o p o ide hei da a as open da a. Fo ins ance,
one equi emen om he use pe spec i e is some hing ha needs o be esol ed by he p o ide

D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
22
o da a holde . In his case, we use hese equi emen s o iden i y he challenges ha occu ed o
he NGD da a holde s.
Table 9 - Technical Requi emen s o NGO Open Da a use s ( aken om ODECO deli e able
D2.2)
Requi emen s
Non-
speciali
s
ac o s
Da a
jou nalis
s
Elemen
a y
school
s uden
s
Non-
go e n
men al
o ganisa
ions
Comme
cial
o ganis
a ions
Open
da a
in e m
edia ie
s
R1 - Facili ies o da a p ocessing
and o checking he quali y o da a
X X X X X
R2 - Da ase s should ha e
schema.o g ags o p o ide a
seman ic meaning
X X X
R3 - Visualize a da ase on he da a
po al
X X X X X
R4 - Da a p o enance
X
X
X
R5 - Technical s anda ds o da a
o ma s, me ada a, APIs
X X X X
R6 - Easily publish da a in he
equi ed o ma s
X X X X X X
R7 - Engage wi h da a holde s o
unde s and hei echnical needs
and challenges
X X X X X X
R8 - Feedback mechanisms o da a
holde s o sugges imp o emen s
X X X X X X
R9 - Reducing he gap be ween
NGO use g oups and hei
echnological needs can help
iden i y he common equi emen s
o all s akeholde s
X X X X X X
R10 - High quali y da a p o ision
X
X
X
X
X
X
R11 - The de elopmen o da a
in eg a o s o enable he echnical
in e ope abili y o open da a
s emming om di e en domains
X X X X X
R12 - Widely adop able da a
o ma s
X X X X X X
R13 - Me ada a sui abili y checke
o da a publishe s
X X X X X X
R14 - Quali y enhancemen ools
(da a augmen a ion ools) should
be a ailable
X
R15 - Use s should be allowed o
upload upda ed e sions o exis ing
da ase s
X X X X X X
R16 - Consis en and eliable
deli e y o da ase s should be
p o ided o he a i icial use s
X
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
23
These equi emen s ela ed mainly o eusabili y and in e ope abili y a e desc ibed in mo e de ail
as ollows:
Use needs o Reusabili y o an Inclusi e Open Da a Ecosys em om a echnical
pe spec i e
• R1: Facili ies o da a p ocessing and o checking he quali y o da a: Cu en ly, po al
adminis a o s and da a use s mus ha e he basic skills o da a p ocessing and o checking
he quali y o da a. Fo open da a eusabili y, da a quali y, alidi y and comple eness o he
da a may be e alua ed wi h he use o echnologies a ailable ia he da a po al. This
equi emen add esses he se ices ha need o be de eloped o educe he da a p ocessing
capabili ies o he po al adminis a o s.
• R2: Da ase s should ha e schema.o g ags o p o ide a seman ic meaning: All he
da ase s a ailable on he open da a po al, e en hi d-pa y da ase s, should ha e schema.o g
ags o p o ide a seman ic meaning because wi h his equi emen we can p io i ize in
sea ches he commen s ha can be mo e ele an o hei p o ile. I can also help in eusing
he da ase .
• R3: Visualize a da ase on he da a po al: I a use wan s o isualize a da ase on he da a
po al, hen he da a po al mus ha e isualisa ion ools and esou ces. Fo be e insigh s o
he da ase , i will be be e o isualize i in he po al be o e downloading and p ocessing i
o u he asks.
• R4: Da a p o enance: Da a p o enance wi h espec o he eusabili y o open da a wi hin
open da a po als ensu es da a au hen ici y and quali y. This in ol es clea ly documen ing he
da ase ’s o igin, owne ship, and upda es. Me ada a and euse s anda ds like open da a
licenses p omo e da a sha ing and accessibili y.
• R5: Technical s anda ds o da a o ma s, me ada a, APIs: Implemen ing CSV, JSON, and
XML da a o ma s p o ides pla o m po abili y. Sea ching o da a and con ex
comp ehension imp o e signi ican ly by using Dublin Co e o schema.o g me ada a
s anda ds. S ong APIs like REST o G aphQL p o ide au oma ed da a access, in eg a ion, and
scalabili y while e aining secu i y and con ol. Da a is eadily accessible, in e p e ed, and used
by a ied people and sys ems because o hese s anda ds.
• R6: Easily publish da a in he equi ed o ma s: An Open Da a Ecosys em equi es da a o
be publishable in he igh o ma s. Da a endo s may al e and dissemina e hei da a in CSV,
JSON, and XML using s aigh o wa d ools and se ices ha simpli y da a ans o ma ion and
publica ion. This s eamlines he p ocess, lowe s echnical ba ie s, and ensu es da a o ma
compliance, imp o ing end-use accessibili y and usabili y.
• R7: Engage wi h da a holde s o unde s and hei echnical needs and challenges:
Engaging da a holde s o unde s and hei echnological equi emen s and di icul ies is
essen ial o an inclusi e Open Da a Ecosys em. Conduc ing su eys, semina s, and eedback
sessions helps da a p o ide s de e mine hei needs and challenges. This knowledge enables
he c ea ion o cus omized assis ance, aining, and echnological solu ions o help da a
holde s publish and dis ibu e high-quali y da a.
• R8: Feedback mechanisms o da a holde s o sugges imp o emen s: Con inuously
imp o ing he Open Da a Ecosys em equi es da a holde s o p o ide inpu . Discussion
o ums, eedback o ms, and w i ing o da a publishe s enable da a supplie s o sha e hei
knowledge. Feedback is c ucial o imp o ing da a s anda ds and ools, ensu ing he open
da a ecosys em mee s use s demands, and iden i ying a eas o de elopmen .
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
24
Use needs o Technical In e ope abili y o an Inclusi e Open Da a Ecosys em om a
Technical Pe spec i e
Technical equi emen s ou lined in Table 10 om ODECO Deli e able 2.2, ele an o he open
da a use s om a echnical s andpoin , a e p esen ed. Some pa s o hese equi emen s a e lis ed
below wi h hei explana ion, ela ed o he non-go e nmen s akeholde s:
• R9: Reducing he gap be ween NGO use g oups and hei echnological needs can help
iden i y he common equi emen s o all s akeholde s. A one-size- i s-all echnological
app oach would be mo e bene icial o s akeholde s om go e nmen o non-go e nmen
sec o s. The solu ions should be eady o enhance bo h ypes o pla o ms (na ional o
go e nmen al da a po als bu also o he open da a po als like Zenodo and OpenAi e).
• R10: High quali y da a p o ision – F om he go e nmen o he non-go e nmen
s akeholde s, and ice e sa. The c ea ion o open da a has h ee di e en pe spec i es: (a)
he ac ual c ea ion o da a (e.g. om IoT senso s), (b) he c ea ion o DCAT-complied da ase s
( la and con ex ual me ada a le els) and (c) he linked open da ase s (de ailed me ada a le el)
c ea ion. Solu ions should a ge all o he abo e.
• R11: The de elopmen o da a in eg a o s o enable he echnical in e ope abili y o open
da a s emming om di e en domains. In his s eam, non-go e nmen s akeholde s can
in eg a e hei da a in a be e way. The p oposed solu ions should a ge easie (au oma ed
i possible) c ea ion o da a and me ada a om hei sys ems.
• R12: Widely adop able da a o ma s – deciding upon uni o m adop able da a o ma s can
esol e se e al p oblems in he cu en open da a ecosys ems and can make open da a
in e ope abili y much easie .
• R13: Me ada a sui abili y checke o da a publishe s – me ada a selec ion is a e y
impo an phase o he open da a publica ion. Fo non-go e nmen al o ganiza ions, he e is
also a need o a me ada a sui abili y checke be o e adding he da a o main s eams o open
da a po als (Na ional o go e nmen al open da a po als).
• R14: Quali y enhancemen ools (da a augmen a ion ools) should be a ailable – Wi h
he ise o AI and LLMs, da a augmen a ion ools can enhance he quali y o open da a.
Le e aging AI o achie e seman ic and syn ac ic in e ope abili y is highly sough a e in he
ma ke . An example o his scena io is AI o da a and da a o AI.
• R15: Use s should be allowed o upload upda ed e sions o exis ing da ase s. I has been
no ed ha use s cu en ly canno upload hei imp o ed da a back o he po al, whe he he
imp o emen s pe ain o da a quali y, me ada a, o applica ions.
• R16: Consis en and eliable deli e y o da ase s should be p o ided o he a i icial
use s – Is he e a way o b idge he gap be ween LLMs and open da a po als? By es ablishing
a echnical in e ope abili y channel, LLMs can e ec i ely u ilize da a om open da a po als.
4.3 Syn hesis
This subsec ion includes he summa y o he echnical equi emen s (D2.2), ac ion p inciples
(D3.3), ba ie s, and mo i a ions (D4.1) o non-go e nmen al da a holde g oups o sha e hei
da a as open da a, g ouping he a ious elemen s acco ding o an expe s’ g ouping as shown in
Table 10. As explained in he me hodology Sec ion, he syn hesis o he Challenges was conduc ed
by he co e au ho s eam o his Deli e able. The Challenges and hei de ailed desc ip ion a e
p esen ed in he nex Chap e (Chap e 5).
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a h ough a echnical s a egy
25
Table 10: Technical equi emen s, ac ion p inciples, ba ie s, and mo i a ions pe Open Da a Holde G oup
Requi emen s + Ac ion P inciples /Non-
Go e nmen al
Da a Holde G oup
Challenge
G oup
Non-
Specialis
Ci izens
Da a
Jou nalis s
Non-
Go e nme
n al
O ganisa
-
ions
Schools
Comme ci
al
O ganisa
-
ions
Open Da a
In e medi
a ies
Impo anc
e coun
Easily publish da a in he equi ed o ma s
1
X
X
X
X
X
X
6
The in e ope abili y needs o a
go e nmen body a e di e en om he in e ope abili y
needs o a company ( agmen ed open da a ecosys ems)
1
X
X
X
X
X
X
6
Use ul da a disco e y ools (e.g., Con ex ualized disco e y)
1
X
X
X
X
X
X
6
High quali y da a p o ision
1
X
X
X
X
X
X
6
The ODP should p o ide echnical assis ance o he
con ibu o o p epa e he da a esou ce o be added o
he po al (da a p epa a ion, cleaning, labelling)
1
X
X
X
X
X
X
6
Widely adop able da a o ma s
1
X
X
X
X
X
X
6
The ODP should help he da a con ibu o disco e use ul
connec ions om he da a esou ces i con ibu es o o he
esou ces al eady p esen in he po al
1
X
X
X
X
X
X
6
Basic skills o da a p ocessing and o checking he quali y
o da a.
1
X
X
X
X
X
5
The de elopmen o da a in eg a o s o enable he
echnical in e ope abili y o open da a s emming
om di e en domains
1
X
X
X
X
X
5
Technical s anda ds o da a o ma s, me ada a, APIs.
1
X
X
X
X
4
Da ase s should ha e schema.o g ags o p o ide a
seman ic meaning.
1
X
X
X
3
Add essing echnical da a challenges
1
X
X
2
Face challenges wi h da a quali y, imeliness, and esou ces
1
X
1
Bene i om seamless da a exchange
1
X
1
Face challenges adop ing new in e ope able o ma s
1
X
1
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
32
• QField
QField is a mobile applica ion ha is mean o in eg a e pe ec ly wi h QGIS, which is a popula
desk op GIS so wa e p og am. I s pu pose is o collec geog aphical da a. In addi ion o
p o iding a a ie y o geog aphic da a ga he ing and edi ing ope a ions di ec ly on mobile
de ices, i enables use s o ake hei QGIS p ojec s ou in o he ield ha hey ha e c ea ed.
The ollowing Figu e 3 highligh s he key a ibu es o QField:
Figu e 3: Key a ibu es o QField.
Beside he key a ibu es he e a e a ew use cases:
• U ban planning and in as uc u e mapping.
• Ag icul u al da a collec ion and a m managemen .
Below, Table 11 compa es he ou da a collec ing and publishing ools (Da ase e, Epicollec 5,
ODK, and QField) o he speci ic use equi emen s men ioned ea lie . Table 11 also p esen s
a echnical assessmen o each ool in co espondence o he Challenge i add esses:
Table 11: E alua ion o Open Da a Tools o he Key Challenges
Challenge
Da ase e
Epicollec 5
ODK
QField
Challenge 1: Enhanced
da a c ea ion p ocess
h ough be e quali y,
in eg a ion and
alida ion ools
Suppo s basic
SQL que ies o
da a alida ion
and analysis
Basic da a
alida ion
h ough o m
design and
cons ain s
Suppo s da a
alida ion ules
and cons ain s
in o ms.
Suppo s
expo s o CSV,
JSON, XML
Relies on QGIS
o da a
alida ion,
some in-app
alida ion ools
Challenge 2: Enhanced
da a sha ing h ough
be e licensing and
anonymisa ion
Me ada a ields
o licensing,
equi es cus om
anonymiza ion
No buil -in
suppo , expo
o ex e nal
anonymiza ion
Me ada a
suppo ,
po en ial o
anonymiza ion
h ough o m
design
Managed
h ough QGIS
me ada a and
ex e nal
anonymiza ion
Challenge 3: Towa ds
inclusi e, use -d i en
and ci cula open da a
pla o ms h ough
be e in e aces and
eedback mechanisms
Gi Hub issues,
communi y
o ums, Buil -in
isualiza ion ools
Communi y
eedback
h ough app,
Basic
isualiza ion
o da a
collec ion
Feedback ia
communi y
channels,
su eys,
Limi ed in-app
isualiza ion,
be e wi h
ex e nal ools
Gi Hub, QGIS
o ums o
eedback,
Limi ed in-app,
elies on QGIS
o ad anced
isualiza ion

D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
33
6.1.2 In eg a ion o non-go e nmen da a on open da a eposi o ies
Open Da a po als a e online pla o ms ha enable he disco e y, access, and managemen o
me ada a eco ds o da ase s (Kuble e al., 2018). These da ase s may be a ailable o download
in a ious o ma s. These po als play a c ucial ole in making da a accessible o he public. In he
Eu opean Union, da a.eu opa.eu is he leading open da a ini ia i e ha b ings oge he da ase s
om communi y p o ide s a all le els o go e nmen (Ca a a e al., 2015). In u n, all membe
s a es ha e a na ional open da a po al (He e a-Mu illo e al., 2022).
Cu en use in e aces in open da a ecosys ems do no acili a e he addi ion o da a by ex e nal
ac o s ou side go e nmen da a p o ide s ( an Loenen e al., 2021). The la es edi ion o he Open
Da a Ma u i y Repo (Publica ions O ice o he Eu opean Union, 2023) o e s se e al
ecommenda ions o le e aging collec i e in elligence by engaging he b oade open da a
communi y in na ional open da a p og ams. Key sugges ions include allowing e-use s o upload
hei own da a and showcase hei ideas and c ea ions on na ional po als. Addi ionally, he epo
ad oca es o enabling use s o commen on and a e da ase s, wi h hese inpu s being in eg a ed
in o sea ch algo i hms.
We conduc ed a e iew o he Eu opean Open Da a Po al and 27 na ional po als, as compiled
by He e a-Mu illo e al. (He e a-Mu illo e al., 2022), o alling 28 po als. Du ing his inspec ion,
we speci ically assessed whe he hese po als explici ly and isibly o e use s he ollowing
ea u es: a) he abili y o eques he addi ion o speci ic da ase s, b) he op ion o submi use -
gene a ed use cases o applica ions, and c) he oppo uni y o non-go e nmen al use s o
con ibu e hei da a o he po al.
Ou o he 28 po als e iewed, 17 (61%) p o ide use s wi h he op ion o eques he addi ion o
da ase s. Howe e , he implemen a ion o his ea u e a ies signi ican ly ac oss po als. In some
ins ances, a gene ic help eques o m is used, whe e use s selec he ca ego y o hei eques ,
wi h da ase inco po a ion eques s being one o he op ions. O he po als o e mo e
sophis ica ed mechanisms; o example, he Es onian po al displays all ecei ed da ase p oposals
and allows use s o o e on hei a ou i es, while he Li huanian po al p esen s a eques his o y
ha includes he ou comes—whe he a ou able o un a ou able—along wi h explana ions o
each decision.
Eigh een o he 28 po als (64%) allow use s o submi applica ions o use cases o be showcased
on he si e. Howe e , he me hods o submission a y widely. Some po als simply in i e use s o
sha e hei use cases ia email, o e ing li le guidance on he equi ed s uc u e. In con as , o he
po als p o ide a dedica ed o m o submissions. A no able example is he Aus ian po al, which
ea u es a comp ehensi e o m wi h mul iple ields, including i le, desc ip ion, na u e o he
applica ion, open da ase s used, links o he applica ion, links o he sou ce code, and
adminis a o in o ma ion.
Only 6 o he 28 po als (21%) allow p o ide s ou side he public adminis a ion o inges hei
own da a in o he po al. These coun ies a e F ance, Spain, Aus ia, Poland, Luxembou g and
Po ugal. The po als o F ance, Aus ia, Luxembou g, and Po ugal pe mi uploads om all ypes
o use s. The Polish po al ex ends his capabili y o companies. In con as , Spain es ic s he
uploading o non-go e nmen al da a o selec p i a e ins i u ions, p ima ily ounda ions,
associa ions, and p o essional bodies ha gene a e da a om publicly unded p ojec s.
To illus a e how a non-go e nmen al p o ide can add i s da a o an open da a po al, we will
use he F ench po al as a e e ence. The po al o e s a ange o esou ces de ailing bes p ac ices
o open da a publishing, as well as speci ic guidelines on how he pla o m ope a es. Da a
p o ide s begin by egis e ing an accoun , which allows hem o add da ase s. The p ocess s a s
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
34
wi h including he me ada a o he esou ce, including he i le, desc ip ion, license, upda e
equency, keywo ds, ime co e age, and spa ial g anula i y. A e his, he ile i sel is uploaded
(See Figu e 4). Finally, he publishe is encou aged o sha e he newly c ea ed esou ce. Once he
da ase is published and accessible o he public, he da a p o ide gains access o a dashboa d
whe e hey can edi he en y and moni o a ious me ics ela ed o he da ase 's usage and
in e ac ion. These me ics include iews, subsc ibe s, use cases, commen s, communi y
con ibu ions, quali y indica o s and ecommenda ions o imp o e da a and me ada a quali y (See
Figu e 5). Non-go e nmen p o ide s can also use he po al API o au oma e da a inges ion in o
he po al.
Figu e 4: Sc eensho o he p ocess o inges ing indi idual da ase s in o he F ench open
da a po al
Figu e 5: Sc eensho o he dashboa d o he da ase s inges ed by a use on he F ench open
da a po al
The addi ion o hi d-pa y da ase s is no limi ed o go e nmen open da a po als; he e a e also
nume ous open-access eposi o ies ha enable a wide ange o use s, pa icula ly esea che s, o
sha e da ase s gene a ed om hei wo k and esea ch p ocesses. A no able example is he
Zenodo eposi o y, managed by CERN, which cu en ly allows he c ea ion o en ies con aining
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
35
up o 100 iles and 50 gigaby es o da a. These en ies can include bo h inal and in e media e
esea ch ou pu s, such as da ase s. O he gene al eposi o ies include Figsha e, D yad, and he
Open Science F amewo k. In addi ion, he e a e domain-speci ic eposi o ies, such as ICPSR o
social sciences, as well as ins i u ional eposi o ies hos ed by a ious academic and esea ch
ins i u ions.
6.1.3 Open da a gene a ed by olun ee s h ough mic o asking: he case o he HOT
Tasking Manage
This sec ion e iews he main lessons lea ned om he ODECO p ojec esea ch on use in e aces
o open da a gene a ion, speci ically in he con ex o humani a ian mapping He e a-Mu illo e
al. (2024).
Volun a y Geog aphic In o ma ion (VGI) is a philosophy ha ha nesses he collabo a i e
con ibu ions o olun ee s o collec , analyse and dis ibu e geog aphic da a. This is one o he
mos success ul examples o open da a con ibu ion by non-go e nmen al ac o s. VGI plays wi h
he idea o “ci izens as senso ” which appeals o he abili y o human beings o syn hesize and
in e p e local in o ma ion in an in elligen way (Goodchild, 2007). In u n, he Humani a ian
OpenS ee Map Team (HOT) is one o he mos ecognized VGI ini ia i es. I le e ages he
OpenS ee Map (OSM) pla o m o enable he de elopmen o eely accessible global maps, wi h
a special emphasis on egions in c i ical need o humani a ian assis ance. A ele an ool in his
p ocess is he HOT Tasking Manage (HOT-TM), which coo dina es a wo ldwide ne wo k o
olun ee s o suppo ocused and e icien mapping e o s. Figu e 6 and Figu e 7 shows he
p ojec asks sc een whe e olun ee s selec he mapping ask o be pe o med and he sc een
co esponding o an indi idual ask.
Figu e 6: The p ojec asks sc een in HOT-TM displays key in o ma ion in wo sec ions. On
he le sideba , use s can ind p ojec de ails, ask s a uses in lis o m, s ep-by-s ep ask
ins uc ions, and con ibu o s a is ics
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
36
Figu e 7: Sc eensho o he mapping phase in HOT-TM. Sou ce:
h ps:// asks.ho osm.o g/p ojec s/15476/ asks
As pa o he ODECO p ojec , He e a-Mu illo e al. (2024) conduc ed a sys ema ic analysis o he
humani a ian mapping p ocess wi hin HOT-TM, aiming o unde s and he in e ac ions be ween
olun ee s and he use in e ace o he pla o m. Based on his esea ch, he au ho s de eloped
a se o p ac ical ecommenda ions o designing pla o ms and use in e aces ha acili a e da a
con ibu ions om non-go e nmen al use s, especially hose ha seek o in ol e he public,
ega dless o whe he geog aphic in o ma ion is in ol ed:
Technology is a powe ul enable o building a communi y o da a con ibu o s: HOT-TM
se es as a model ha can be eplica ed by o he non-go e nmen al open da a ini ia i es. I s
echnological pla o m ac s as a b idge be ween a obus olun ee communi y, like OSM, and
humani a ian o ganiza ions esponding o global disas e s. The pla o m equips bo h g oups wi h
he necessa y use in e ace, ools and in o ma ion o c ea e and alida e open da a.
Volun ee expe ience is a key asse : when olun ee s sign up o HOT-TM, hei p e ious
mapping expe ience wi hin OpenS ee Map is au oma ically sha ed h ough indica o s. This
enables p ojec manage s o se expe ience equi emen s o pa icipa ion in p ojec s based on
hei complexi y. The ask alida ion phase is ese ed o mo e expe ienced use s, who possess
he necessa y judgmen o e alua e and co ec mapping p oduc s e ec i ely. Resea ch shows
ha his dis inc ion is well- ounded, as ad anced use s consis en ly demons a e highe
p oduc i i y, be e ask success a es, and sus ained pa icipa ion o e ime. Open da a ini ia i es
by non-go e nmen al ac o s can g ea ly bene i om iden i ying key a iables o assess da a
con ibu o expe ience. Doing so allows o mo e e icien ask assignmen , ensu ing ha p ojec s
a e ma ched wi h he igh le el o expe ise. Managing he expe ise o da a con ibu o s also
in ol es add essing he challenge o e aining and mo i a ing no ice use s, many o whom s op
con ibu ing a e hei i s p ojec . This sugges s ha ini ia i es should ocus on la ening he
lea ning cu e o newcome s, p o iding hem wi h cons uc i e eedback and iendly use
in e aces, and sha ing indica o s o he eal-wo ld impac o hei con ibu ions.
Mic o asking can simpli y pa icipa ion o da a con ibu o s: HOT-TM employs a mic o
asking app oach ha , unlike egula OSM dynamics, di ec s use s owa d speci ic, simple
mapping objec i es. Mos asks can be comple ed in jus a ew minu es, allowing e en
inexpe ienced use s o con ibu e meaning ully o he p ojec goals. This mic o asking model can
be applied o o he non-go e nmen al open da a agg ega ion ini ia i es, whe e pa icipan s a e
asked o make small and manageable con ibu ions. When combined, hese indi idual e o s can
ha e a signi ican impac , making la ge-scale p ojec s mo e accessible and e icien h ough he
powe o c owdsou cing.
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
37
Da a quali y ma e s: a common conce n wi h open da a gene a ed by non-go e nmen al
s akeholde s is whe he i mee s he necessa y quali y s anda ds o u u e use. While many o
hese s akeholde s possess aluable da a, hey o en lack he skills and ools o e ec i e da a and
me ada a quali y managemen . O e bu dening hem wi h addi ional quali y con ol asks could
discou age hei pa icipa ion o di e hei ocus. To add ess his, HOT-TM has de eloped a wo-
s age quali y con ol mechanism. In he i s s age, du ing he mapping phase, con ibu o s ocus
on achie ing he mapping objec i es as quickly as possible, ensu ing ha geog aphic in o ma ion
o eme gency esponse is a ailable immedia ely. In he second s age, du ing he alida ion phase,
p ojec o ganize s and expe ienced con ibu o s alida e and co ec he mapping asks, ensu ing
ha he inal p oduc mee s quali y s anda ds. This app oach demons a es how any open da a
ini ia i e in ol ing non-go e nmen al ac o s can bene i om simila quali y con ol mechanisms,
whe he h ough au oma ed ule-based checks o e iews by human expe s.
P ocess da a is a use ul ool o con inuous imp o emen : HOT-TM collec s de ailed da a on
con ibu o s, he wo k p ocess, and he esul ing ou pu s. This in o ma ion is sha ed wi h ele an
communi ies ia a well-documen ed API (see Figu e 8). This app oach enables da a-d i en
managemen o non-go e nmen al open da a ini ia i es, allowing ex e nal con ibu o s o engage
in p ocess imp o emen wi hin an open inno a ion amewo k. Key imp o emen decisions, such
as hose ela ed o use in e ace design, can be e ec i ely in o med by analysing in e ac ion da a
collec ed om use s.
Figu e 8: HOT-TM API ha allows access o da a on campaigns, p ojec s, o ganiza ions,
asks, commen s, among o he s. Sou ce: h ps:// asks.ho osm.o g/api-docs
Ha nessing he collec i e in elligence o non-go e nmen al open da a con ibu o s:
Resea ch indica es ha HOT-TM is highly e icien in achie ing b oad mapping objec i es wi hin
a sho ime ame. Howe e , i s success elies hea ily on expe ienced mappe s, while no ice
mappe s—who make up mos con ibu o s— end o ha e a lesse impac , pa ly due o
subop imal in e ac ion wi h he pla o m. As pla o ms like HOT-TM de elop mo e sophis ica ed
mechanisms o agg ega ing use inpu , hey can ap in o he "wisdom o he c owds," whe e he
collec i e con ibu ions o a di e se g oup esul in an ou come o highe quali y han he sum o
indi idual inpu s.
A e iden i ying he ools o da a sha ing desc ibed abo e, i is no iceable ha hese ools allow
o he da a c ea ion and sha ing ypes o pla o ms, howe e , he need o make he c ea ed
da ase s mo e accessible in gene al open da a po als emains. Many o he echnical s ee ing

D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
38
mechanisms (o solu ions) p oposed in he ollowing Sec ion, a e no limi ed o (e en hough hey
could imp o e hose as well) go e nmen da a sou ces (e.g., go e nmen open da a po als), bu
hey a e applicable o use on o he non-go e nmen al da a sou ces and p o ide s, which is
mainly he ocus o his Deli e able.
6.2 P oposed echnical solu ions wi hin he ODECO P ojec
In his Sec ion, p oposed echnical solu ions co esponding o each o he Challenges desc ibed
in Chap e 5 a e p esen ed. The p oposed solu ions a e ei he a esul o ongoing esea ch wi hin
he ODECO p ojec ( hus no inalised bu unde p og ess), o a esul o a p oposed solu ion om
he ODECO TW5 Idea hon.
An impo an aspec o conside is ha he e is a dis inc ion be ween he echnical solu ions wi hin
he ODECO p ojec and he ones p oposed du ing he OD Idea hon in Samos. The o me ones
a e ongoing esea ch o he ODECO eam, and hey will con inue o p og ess while he p ojec is
s ill ongoing, bu he la e a e solu ion ideas o be conside ed and migh be (usually he case) a
a much ea lie s age o concep ion, whose p og ess in o deepe le els o p og ess is ou o he
scope o his deli e able.
Ano he impo an no e o he ollowing Sec ion o his documen is he dis inc ion be ween he
e m “use ” in i s con en ional use and meaning (so anyone, including ci izens, non-specialis use
g oups e c.) and he use o his deli e able. In his Sec ion, whene e he e m “use ” is
encoun e ed, i is no limi ed o he use as simply a da a use , bu a he , i e e s o a sys em
use , so he a ge sys em use who will make use o he p oposed echnical solu ions p esen ed.
In ou case, and as he main pu pose o his deli e able, his o cou se e e s o he non-
go e nmen da a holde g oups which we aim o mo i a e o sha e hei da a, using he p oposed
and en isioned echnical solu ions.
6.2.1 P oposed solu ions o Challenge 1 - Enhanced da a c ea ion p ocess h ough be e
quali y, in eg a ion and alida ion ools
6.2.1.1 Augmen ing seman ic in e ope abili y o da ase s
This p oposed seman ic in e ope abili y solu ion aims o a ge one o he common p oblems ha
a ises when non-go e nmen da a mus be combined and in eg a ed wi h o he he da a sou ces:
he seman ic he e ogenei y be ween he schemas (concep ual models) o wo da a sou ces ha
mus be combined. The main challenge ela ed o seman ic in e ope abili y ha o ganisa ions
and da a po als ace is ela ed o he inconsis en use o da a ocabula ies and me ada a schemas
(Ma a si e al., 2024 a); O ganiza ions use di e en (cus om) da a and me ada a schemas ha do
no ma ch exis ing s anda dised ocabula ies and on ologies o exp ess domain-speci ic
knowledge.
Fo ins ance, aking an example om he Cul u al He i age domain, and assuming we ha e wo
o ganisa ions (e.g., a galle y, museum, e c.), he a i ac s exhibi ed in A Galle y 1 and A Galle y
2 o Figu e 9, a e no au oma ically in e ope able when exchanging da a om hese wo
ins i u ions, due o he lack o linking he cus om schemas o known, s anda dised ones o his
speci ic domain (e.g., CIDOC-CRM and mo e, in his case). The ques o s anda disa ion and
in e linking exis ing on ologies and schemas mainly in he cul u al he i age domain, bu also in
o he domains (e.g., legal domain, psychology), has been ac i ely esea ched wi hin he ODECO
p ojec , he esul s o his s udy o be ound in (Ma a si e al., 2024 b, Lou sa is e al., 2023, Ma a si
e al., 2024 c, Ma a si e al., 2023).
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
39
Figu e 9: A simple example o di e en da a schema use in a Cul u al He i age case
I he cus om schema ields o he wo di e en ins i u ions could be mapped in an au oma ed o
semi-au oma ed way o exis ing, s anda d schemas upon da ase upload o ins ance, hen he
new da a would ins an ly become seman ically in e ope able wi h mo e da a sou ces, and mo e
easily disco e able by seman ic sea ch engines.
The p oposed solu ion aims o use an LLM (e.g., LLAMA) o align and map hese a ious schemas
o s anda dized, ocabula ies. This model would be ained using expe -cu a ed da a o ensu e
accu acy and would ollow he p ocess o Figu e 10.
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
40
Figu e 10: The p oposed implemen a ion p ocess
Ini ially, he LLM model will be ained on cu a ed expe - alida ed da a (examples o expe s’
mappings), and hen will go h ough a p elimina y sys em alida ion p ocess which will be
supe ised (app o ed o ejec ed) by a human. This is going o be an i e a i e p ocess un il
consensus is eached ega ding he p oduced mappings o s anda dised ocabula ies and
schemas. On a simila no e, he use o Linked Open Vocabula ies, o he acili a ion o his p ocess
o on ology and s anda d domain-speci ic ocabula y e-use o imp o ed seman ic
in e ope abili y is desc ibed by (Ma a si e al., 2024 d). Some echnologies o be used o he
implemen a ion o his solu ion a e he ollowing:
• Da a Inges ion & S o age: AWS S3
• Model T aining and Deploymen :
o Model T aining: PyTo ch, Tenso Flow
o T aining En i onmen : GPUs and TPUs on AWS SageMake
• Model Deploymen : Docke
• Da a In eg a ion: Apache Ka ka
• Moni o ing & Logging: P ome heus
The ou come o his p ocess is a cohe en da a in eg a ion which ensu es all da a is aligned wi h
he p oposed sys em's s anda d ocabula ies. The o m o ou come is a p oduced
ecommenda ion o mappings o s anda dised ep esen a ion schemas o a speci ic domain (o
mo e gene ic i his is no necessa y). The esul s o his p ocess will hen be con ibu ed o he
Linked Open Da a Cloud (LODC).
In p ac ice, when an o ganisa ion aims o sha e hei da a (e.g., upload o an OD po al) o
elsewhe e, he p oduced sys em solu ion will p o ide hem wi h a se o ecommenda ions on
how hei cus om schema can be imp o ed in e ms o seman ic in e ope abili y. Figu e 11 shows
an indica i e sc een p o o ype o he mapping ecommenda ion p ocess o he end use (e.g.,
o ganisa ion o da a holde g oup which wan s o sha e hei da a).
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
41
Figu e 11: Indica i e sc een p o o ype o he mapping ecommenda ion upon da ase
upload
This p oposed echnological solu ion can be u ilised, o ins ance, as a SaaS da a s anda disa ion
pla o m, opening a ious oppo uni ies o be in eg a ed in exis ing open da a po als, o as a
sepa a e se ice. This solu ion aims o imp o e he s a us on seman ic in e ope abili y o da ase s
and ep esen a ion schemas, bu also consequen ly imp o es da ase indabili y, while a he same
ime makes ac i e con ibu ions o he Linked Open Da a pa adigm.
6.2.1.2 Au oma ic Thema ic Anno a ion o Open Da a
Whene e a non-go e nmen da a asse is in eg a ed in an open da a po al, i is essen ial o
classi y and anno a e i app op ia ely o acili a e i s indabili y and access by po en ial use s. This
p oblem is simila o he p oblem aced by Open Da a po als such as he Eu opean Da a po al
(da a.eu opa.eu) ha mus agg ega e and classi y he con en s ha es ed om he e ogeneous
na ional and local po als o p ede ined lis o hemes. Al hough ideally he o iginal da ase s
should be al eady anno a ed wi h a heme ocabula y and la e ma ched wi h he Eu opean da a
hemes, based on ou desk op e iew o he da ase s, i is e iden ha assigned hemes a e o en
ei he missing o inco ec . Inco ec heme assignmen can hinde he indabili y o he da ase s
by o he s akeholde s.
To add ess hese issues, we de eloped a solu ion ha assigns p ope hemes o da ase s, e en
when a da ase belongs o mo e han one ca ego y. This solu ion is a hema ic classi ica ion
sys em designed o esol e he p oblem o inco ec hema ic assignmen . We c ea ed a mul i-
class classi ica ion model using a aining da ase o 29,793 eco ds in RDF o ma . The sys em
employs bo h classi ica ion echniques and cosine simila i y, analysing he me ada a ( i le and
desc ip ion) o da ase s agains he hemes de ined by he Eu opean Open Da a Po al.
The de eloped sys em can assign hemes o da ase s based on hei i le and desc ip ion wi h
92% accu acy. Addi ionally, he model can sugges a second heme ha could be assigned o a
da ase , imp o ing i s indabili y on he po al. All he code and da a a e a ailable on he gi hub
eposi o y.
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
48
Figu e 18 shows he p oposed solu ion de elopmen p ocess o c ea e a i s Minimum Viable
P oduc (MVP) o he p o ided se ice:
Figu e 18: The de elopmen p ocess o he p oposed pla o m
This p oposed solu ion educes he isk o da a exposu e du ing de elopmen and es ing phases,
p o ec ing sensi i e in o ma ion while i also imp o es da a quali y by ensu ing high da a quali y
and in eg i y, leading o be e decision-making and ope a ional e iciency. Mo eo e , i simpli ies
compliance wi h da a- ela ed egula ions by es ablishing clea policies and p ocedu es o da a
managemen , in eg a ing da a licensing sand compliance unde he same pla o m.
6.2.2.2. An Open License Decision Suppo So wa e
One o he p oblems ha sys em use s (in his case, including in e es ed non-go e nmen da a
holde g oups o sha e da a) a e encoun e ing is he complexi y o licensing cu en ly used in
open da a se s. Now, he e a e mo e han 300 open da a licenses in use. Choosing he igh license
equi es specialized knowledge o unde s and he a ious licenses accompanying open da a se s
and o p ope ly euse hem. I becomes e en mo e complica ed in cases whe e use s mus
combine da a se s, especially om di e en p o ide s and om di e en coun ies wi h di e en
legal amewo ks. This can be a se e e impedimen and discou ages open da a use s no only
om using open da a bu , mo e impo an ly, om becoming ac i e con ibu o s o he open da a
ecosys em. The e o e, he aim o his p oposed solu ion is he c ea ion o an open license decision
suppo sys em ha can help open da a c ea o s and use s o handle he di icul decision o
picking a da a license.
Al hough he e a e some ools ha p o ide his unc ionali y, some d awbacks ha e been
de ec ed. The design o hese ools is s a ic – hey ei he p o ide a oo much in o ma ion in a
abula o m, o hey exis as mul i-s ep ques ionnai es whe e he esponden has no idea abou
hei p og ess in he ques ionnai e o abou how he ques ionnai e i sel is s uc u ed.
In his p oposed solu ion, he goal is o de elop a ool ha can (Figu e 19):
• P o ide a decision-suppo sys em o da ase c ea o s o choose he igh license acco ding
o hei speci ic needs.
• Build a sys em ha will se e as a lea ning ool o da ase c ea o s on how o na iga e he
di e en conside a ions when hey mus selec a da a license.

D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
49
Figu e 19: Key componen s o he sys em
Fi s , he eam aims o design a simple ques ionnai e o be answe ed by da ase c ea o s o e-
use s. This ques ionnai e will p omp esponden s o e lec on he sou ce and p o enance o each
indi idual da a elemen in hei da ase , hei objec i es o c ea ing and sha ing he da ase , and
hei p op ie a y o o he in e es s in he da ase . Each esponse will be mapped o nodes wi hin
he backend decision ee, gene a ing a lis o licenses ha mee he esponden 's pa ame e s.
Second, he eam aims o esponden s o be able o "see" he in e nal wo kings o he decision-
suppo so wa e. This ea u e enables he so wa e o se e as a lea ning ool o esponden s as
well. I also allows o mo e anspa ency o he so wa e and mo es away om a echno-
solu ionis "black box" design. The goal is o c ea e a use in e ace whe e he esponden can
always see each sec ion o he ques ionnai e as well as hei p og ess. The eam also wan s o
enable esponden s o oggle be ween ques ions and al e answe s. Finally, hey in end o he
license decision ee o be dynamic o he esponden . In o he wo ds, as he esponden
p og esses h ough he ques ionnai e, he license decision ee will be gene a ed wi h each
esponse and modi ied wi h each al e a ion o a esponse. Ins ead o being isible o he
esponden only a he end o he ques ionnai e, his dynamic license ee will always be isible
as hey go h ough he ques ionnai e, allowing hem o unde s and he unde lying logic o he
ecommenda ion sys em. Figu e 20 and Figu e 21 explain he key componen s o he sys em and
h ee ame o he p oposed use in e ace o he sys em espec i ely.
Figu e 20: Th ee-pa ame o he p oposed use in e ace
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
50
In Figu e 21, he de elopmen p ocess o he sys em is depic ed. The aim o he eam was o adop
a p ocess o he c ea ion o he decision sys em ha would "design solu ions 'wi h' s akeholde s,
and no ' o ' hem." The e o e, a ci cula design o he de elopmen was adop ed so ha he
sys em could be adjus ed based on he equi emen s and needs o he use s.
Figu e 21: The de elopmen p ocess o he p oposed pla o m
The p oposed open license decision suppo sys em add esses he complex challenge o
na iga ing he mul i ude o open da a licenses, o e ing a use - iendly solu ion o bo h c ea o s
and use s o open da a. By combining a simple ques ionnai e wi h a dynamic, anspa en
in e ace, he sys em no only ecommends app op ia e licenses bu also se es as an educa ional
ool, empowe ing use s o unde s and he decision-making p ocess.
6.2.3 P oposed solu ions o Challenge 3 - Towa ds inclusi e, use -d i en and ci cula
open da a pla o ms h ough be e in e aces and eedback mechanisms
Imagine an open da a po al in which a a ie y o use g oups, such as go e nmen agencies,
esea che s, de elope s, and o ganiza ions ha ep esen ci il socie y, a e ac i ely engaging wi h
da ase s. The eedback equi emen s o hese use g oups a e di e se; o ins ance, de elope s
a e looking o mo e echnical da a o ma s, while esea che s may seek imp o emen s o
me ada a o acili a e da a in e p e a ion. Despi e he a ailabili y o eedback channels such as
discussion o ums, emails, and su eys, he e is equen ly a lack o anspa ency ega ding he
p ocessing and implemen a ion o his inpu . Consequen ly, use s a e le wi h a lack o cla i y
ega ding he enhancemen s o modi ica ions ha a e he ou come o hei con ibu ions.
A possible eedback scena io migh in ol e a use accessing a clima e da ase h ough he
na ional OGD po al. A e wo king wi h he da a, hey no ice inconsis encies in he da e o ma s
and p o ide de ailed eedback ia he discussion o um o ill in he eedback o m. Howe e , due
o he agmen ed na u e o eedback p ocessing ac oss di e en channels (emails, eedback o m,
social media, discussion o ums), hei inpu is no add essed p omp ly, o he po al lacks
anspa ency ega ding wha ac ions a e aken based on such eedback. As a esul , he eedback
loop is incomple e, hinde ing da a eusabili y and us among use s.
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
51
6.2.3.1. A Cha Bo Guide o Open Da a Po als
The mo i a ion behind his p oposed solu ion is he need o ci izens wi hou echnical skills on
open da a o access in o ma ion, ools and da a a ailable, knowing whe e and how o ind hem.
Ci izens wi hou echnical skills migh need open da a o unde s anding an issue, o example,
clima e change, compa e Open Da a and c ea ing a solu ion o sol e a p oblem.
The idea pe ains o a Cha bo assis an which p o iles he use owa ds he esou ces hey need
on he open da a po al and ex e nally. Figu e 22 shows he sc eensho o cha bo .
Figu e 22: An example o a Cha bo assis an on he G eek open da a po al
Fo ins ance, as also shown in Figu e 22 he Cha bo could ask ques ions o unde s and he use
needs and help na iga e hem acco dingly h ough he a ailable da a. Below a e some indica i e
ques ions he Cha bo could ask:
• Wha is you p oblem?
• Do you wan me o sugges opics o in e es ?
• Wha da a a e you looking o ?
• Do you need me o di ec you o he ools and guides o help you analyse he da a?
• The e exis s esea ch on his opic using his da a. Do you wan me o di ec you o i ?
The e can also be sugges ions o ques ions ha migh be o in e es o he use .
The echnologies o be used o he implemen a ion o his p oposal a e mainly Gene a i e AI and
Na u al Language P ocessing (NLP). C ea ing a da abase o exis ing ools/ isualisa ions and da a
esea ch on he po al o om ex e nal s akeholde s could enhance he us wo hiness o
o ganisa ions, while he in o ma ion o he Cha bo can be e i ied and/o p o ided by a lis o
o ganisa ions ha he da a po al owne s deem as us ul.
The alue o he p oposed solu ion o ci izens (use s) is ha i helps hem gain unde s anding o
a opic, while ha ing da a and esea ch in he same place. The p ocess o accessing and using
da a is acili a ed. F om he open da a po als’ side: new da a use s a e being a ac ed. In addi ion,
missing da a o componen s can be disco e ed, as well as ools and o he esou ces cu en ly no
a ailable on he po al. The alue o his solu ion migh also be add essed owa ds p i a e and
hi d pa ies, such as h ough he c ea ion o a da abase o s akeholde s (NGOs, NPOs,
olun ee ing o ganisa ions and mo e).
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
52
The de elopmen p ocess o he Cha bo would include he a ixing o an app op ia e license, he
p epa a ion o he da abase, ools and esea ch a ailable o use, he e i ica ion o sou ces om
ex e nal s akeholde s, he c ea ion o he algo i hm and models, he design o he in e ace and
guide cha ac e , he au oma ion o solu ions, conduc ing he app op ia e usabili y es s, and
o ming a human suppo g oup o expe s ha would be a ailable upon eques .
6.2.3.2. Enhancing be e in e aces and use in e ac ion unde eedback suppo ed by AI
This p oposed solu ion add esses he challenge o be e in e aces and use in e ac ion wi h open
da a, solu ions ocused on c ea ing mo e in ui i e, engaging, and accessible ways o use s o
in e ac wi h da a. The p oposed in e ace will be hos ed by a local da a po al and will be
add essed o all ci izens (including ulne able g oups). The a ionale behind his is inding he
app op ia e mechanisms o suppo and de elop mo e in ui i e, AI-powe ed in e aces ha allow
use s o que y and in e ac wi h da a in na u al language. These in e aces could p o ide
pe sonalized ecommenda ions, isualiza ions, and insigh s based on use que ies, also allowing
o AR (Augmen ed Reali y) in eg a ion o c ea e imme si e da a in e ac ion expe iences, whe e
use s can explo e complex da ase s in a 3D en i onmen , making i easie o unde s and pa e ns
and ela ionships. Fu he mo e, oice-ac i a ed assis an s ha guide use s h ough da ase s could
be in oduced, o e ing explana ions, answe ing ques ions, and helping wi h da a analysis asks.
Mo e speci ically, he key ea u es o he IDIP (In e ac i e Da a In elligence Pla o m) include:
1. In e ac i e Da a Visualiza ions
o Dynamic Dashboa ds: C ea e cus omizable, in e ac i e dashboa ds ha allow use s
o isualize da a in eal- ime. Use s can d ag and d op da a elemen s o c ea e hei
own isualiza ions (cha s, g aphs, maps) and explo e he da a dynamically.
2. Augmen ed Reali y (AR) and Vi ual Reali y (VR) In eg a ion
o In e ac i e Da a Models: In AR, use s could in e ac wi h holog aphic da a models
using ges u es o oice commands. This is pa icula ly use ul o educa ion, allowing
s uden s and p o essionals o engage wi h da a in a hands-on, in ui i e manne . The
VR employs 3D nea -eye displays and pose acking o gi e he use an imme si e
eel o a i ual wo ld and he e o e he use s can ha e a u u is ic idea engaging also
wi h da a.
3. Voice-Ac i a ed Assis an s p o iding eedback
o Pe sonalized Da a Assis an : In oduce a oice-ac i a ed assis an ha helps use s
na iga e he open da a pla o m, answe s hei que ies, p o ides summa ies, and e en
pe o ms da a analysis asks. The assis an could also lea n om use in e ac ions o
p o ide mo e pe sonalized expe iences o e ime.
o Hands-F ee In e ac ion: This ea u e is especially aluable o use s who need o
mul i ask o o hose wi h accessibili y needs. The oice assis an could execu e
commands, e ie e da a, and explain complex da ase s wi hou he need o manual
inpu .
The echnologies he ea u es will be based on include:
• An LLM Fine uning - p o iding an LLM wi h addi ional and specialized da a o e ine he
aining.
• Re ie al-augmen ed gene a ion (RAG) a speci ic a chi ec u e in which que ies a e
“augmen ed” wi h addi ional and domain speci ic/con ex ual in o ma ion. Implemen ing RAG
in an LLM-based ques ion answe ing sys em has wo main bene i s: I ensu es ha he model
has access o he mos cu en , eliable ac s, and ha use s ha e access o he model’s
sou ces, ensu ing ha i s claims can be checked o accu acy and ul ima ely us ed.
• Technology Accep ance Model (TAM) (pe cei ed use ulness, pe cei ed ease o use)
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
53
o Impac T acking: T ack he impac o eedback o e ime, showing how use
sugges ions ha e led o angible imp o emen s in da a quali y, usabili y, o
ea u es. This anspa ency encou ages con inued use pa icipa ion.
O he echnology cha ac e is ics which need o be aken unde conside a ion a e:
• Accessibili y: Easily ob ainable by e e yone.
• Licensing: No es ic ions on usage.
• In e ope abili y: Usable ac oss di e en pla o ms and sys ems.
The pe cei ed bene i s o such an in e ace o he use s include:
• Inc eased Use Engagemen : By p o iding mo e in ui i e and in e ac i e ways o engage
wi h da a, use s a e mo e likely o explo e and u ilize open da a esou ces, leading o mo e
in o med decision-making and inno a ion.
• B oadened Accessibili y: The inclusion o accessibili y ea u es and mul i-language suppo
ensu es ha a wide ange o use s, ega dless o hei echnical expe ise o physical abili ies,
can e ec i ely use he pla o m.
• Enhanced Da a U iliza ion: Wi h ools ha simpli y da a disco e y, analysis, and
isualiza ion, use s can mo e easily de i e insigh s and alue om open da a, leading o
g ea e use and impac o hese esou ces.
• Facili a ed Collabo a ion: The collabo a i e ea u es os e a communi y a ound da a
explo a ion and analysis, encou aging knowledge sha ing and join p oblem-sol ing. By
implemen ing an In elligen Da a In e ac ion Pla o m (IDIP), open da a becomes mo e
accessible, engaging, and use ul o a b oade audience, ul ima ely enhancing he impac and
alue o open da a ini ia i es.
De elopmen p ocess pa ame e s
Rega ding he de elopmen p ocess o his p oposed solu ion, i will need o include he ollowing:
• Use T aining and Onboa ding: P o ide u o ials o onboa ding sessions o help use s
unde s and how o e ec i ely use he eedback sys em. This could include demons a ions o
how eedback leads o angible imp o emen s.
• P i acy and Anonymi y: Ensu e ha use s can p o ide eedback anonymously i hey choose,
especially when dealing wi h sensi i e da a o issues. P i acy conside a ions should be
pa amoun , pa icula ly in handling pe sonally iden i iable in o ma ion (PII).
• Open Sou ce and Communi y Collabo a ion: Conside making he pla o m open sou ce o
encou age communi y-d i en de elopmen and inno a ion. This also allows o b oade
collabo a ion and in eg a ion wi h o he open da a ini ia i es.
Fo Na u al Language P ocessing (NLP) In e aces:
• Con e sa ional Que ying: De elop an NLP-based in e ace ha allows use s o in e ac wi h
da a using na u al language (i.e. using BERT).
• Con ex ual Sugges ions: The in e ace could o e sugges ions as use s ype, p edic ing hei
needs based on he con ex o hei que ies. This ea u e helps guide use s, especially hose
who may no be amilia wi h complex da a s uc u es o que ying languages.
Fo Augmen ed Reali y (AR) and Vi ual Reali y (VR) In eg a ion:
• 3D Da a Explo a ion: De elop AR/VR en i onmen s whe e use s can explo e complex
da ase s in 3D space. Fo example, in a VR en i onmen , use s could "walk h ough" a ci y
map ha displays popula ion densi y, pollu ion le els, and anspo a ion ne wo ks in a
isually imme si e way.

D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
54
Implemen a ion and Deploymen :
• Use -Cen ic Design P ocess: De elop he pla o m wi h a use -cen ic app oach, in ol ing
po en ial use s in he design and es ing phases o ensu e he in e ace mee s hei needs and
expec a ions.
• Modula A chi ec u e: Build he pla o m using a modula a chi ec u e ha allows o easy
upda es and he in eg a ion o new ea u es. This ensu es he pla o m can e ol e wi h use
needs and echnological ad ancemen s: (
open da a -> LLM model (BERT) –> RAG echnique
-> use in e ace
)
• Cloud-Based In as uc u e: Deploy he pla o m on a cloud-based in as uc u e o ensu e
scalabili y, accessibili y, and e icien handling o la ge da ase s.
Sus ainabili y:
• Ensu ing ha he igh s will belong o he local go e nmen .
• A company will ake ca e o he main enance o he p oduc , making necessa y moni o ing
and adap a ions using u u e science inno a ions and de elopmen s (a e a public ende ),
ensu ing he con inua ion o he unding).
• Use in e ace and echnical specs o o be kep o a ce i ied eposi o y (i.e. ZENODO) – Open
Science EU.
6.2.3.3. A mul i-channel inclusi e eedback mechanism o di e se communi ies
The abili y o p o ide eedback and use pa icipa ion is c i ical o c ea ing a ai digi al u u e in
a da a-d i en socie y. So, we p opose he idea abou a mul i-channel eedback mechanism o
os e inclusi i y and engagemen in open da a sha ing. I aims o empowe bo h non- echnical
ci izens and people wi h echnical skills, o p o ide aluable eedback h ough open da a po als.
The e a e wo main channels o his idea:
• Fi s , we wan o empowe hose who al eady ha e da a li e acy skills o p o ide hei eedback
h ough open da a po als.
• Second, we wan o ind new ways o connec and include ci izens wi h he open da a po als
h ough in e media ion by public lib a ies. Public lib a ies o e a weal h o se ices o ci izens
and a e, by design, inclusi e places.
This idea p oposes a ans o med eedback loop ha no only add esses he echnical needs o
open da a po als bu also s eng hens communi y bonds, educes digi al inequi ies, and
empowe s ci izens ac oss all le els o pa icipa e in shaping he da a landscape. I also has he
po en ial o enhance anspa ency, ele a e public us , and ensu e ha da a-d i en go e nance
is uly e lec i e o all oices.
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
55
Figu e 23 and Figu e 24 show he wo k low o he desc ibed wo channels.
Figu e 23: Channel 1 o a mul i-channel inclusi e eedback mechanism o di e se
communi ies
Figu e 24: Channel 2 o a mul i-channel inclusi e eedback mechanism o di e se
communi ies
6.2.3.4. A Plug-in o Feedback Mechanisms o Exis ing Open Da a Po als
The mo i a ion behind his idea is i old. Fi s , om he da a use s’ side, he di icul y o p o iding
eedback on he po al (lack o commen ing ea u es, lack o epo ing ools), as well as he lack
o awa eness o how o why o in e ac wi h he po al, as well as wha he e ec o his in e ac ion
migh be. In addi ion, use s migh include uno icial da a sou ces, o w ong da a. F om he da a
holde s’ side, he e is also li le o no in e ac ion wi h he da a, and he isk o uno icial da a
sou ces being used is also high. The da a also uns he isk o becoming easily ou da ed. F om he
da a po al owne s’ side, he men ioned issues esul in low a ic, less in e ac ion wi h he
websi e, low e enue om possible ad e isemen and coope a ions, as well as p oblems wi h
impac epo ing (i ele an ).
Figu e 25 shows an example o a da ase in e ac ion case a he Eu opean Da a Po al. The cu en
a ailable in e ac ion wi h he da a is: download, me ada a, and alida ion. This igu e shows an
example o he limi ed op ions o in e ac i e p ocesses ega ding he in-wo d da ase each ime.
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
56
The e is no easy way o add eedback o his da ase , gi e ecommenda ions on wha is needed,
o see o he , exis ing commen a ies om o he use s o his da ase .
Figu e 25: Example case o in e ac ion wi h a da ase
The p oposed solu ion equi es ha issues a e in eg a ed in he da a po als, ha use s can add
issues in a use - iendly way, and hey can p opose changes (in he o m o pull eques s), and
ha use s a e gene ally awa e o hese possible in e ac ions. Fo ins ance, he F ench da a po al,
al eady implemen s a commen ing sys em in o he da ase s, as shown in Figu e 26.
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
57
Figu e 26: An example o a use in e ac ion wi h a da ase on da a.gou .
He e we can see ha unde he Tab “Discussions” use s can commen o s a a eedback h ead
o his da ase . Technologies which can be used o his pu pose include he W3C Web
Accessibili y Ini ia i e (WAI), which p o ides ools o disabled people (e.g., people wi h hea ing
p oblems, impai ed ision and o he ), Uda a (pypi.o g, 2024), a cus omisable social pla o m o
open da a, VCS so wa e, o da ase e sion con ol, gi , and mo e.
The solu ion p oposed is a plugin o open da a po als, imp o ing use expe ience in an easy and
e icien way o communica ion and in e ac ion wi h da a and he po al owne s, as well as da a
quali y o speci ic use cases. In addi ion, da a holde s (owne s) can p o ide highe quali y da a
o imp o ed e iciency and/o in e nal ope a ions and gain new ideas o da a p o ision. Da a
po als will inc ease hei use - iendliness, and he highe quali y o da ase s will esul in mo e
impac o he pla o m, some hing i al o imp o ing he open da a ecosys em.
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
64
Idea hon P og amme
Day 1
• In oduc ion and challenges p esen a ion
• Teams’ o ma ion and acili a ion
• Teamwo k
Day 2
• Men o ing
• Teamwo k
• Final p esen a ion

D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
65
Appendix B – Use -cen ed Design
Use -cen ed design (UCD) is an i e a i e design p ocess in which designe s p io i ize use s and
hei needs a e e y s age. In UCD, design eams engage use s h oughou he design p ocess
using a ious esea ch and design echniques o c ea e highly usable and accessible p oduc s.
Acco ding o ISO 9241-210:2010 (See Figu e 27), once a plan o he design p ocess is es ablished,
ou in e connec ed human-cen ed design ac i i ies a e unde aken:
Figu e 27: The use -cen ed design p ocess based on ISO 9241-210:2010
Unde s and and speci y he con ex o use: This ac i i y in ol es iden i ying and desc ibing he
cha ac e is ics o he a ge use s and o he ele an s akeholde s. I includes ou lining he use s'
goals and asks, as well as unde s anding he o ganiza ional, echnical, and physical en i onmen s
ha de ine he sys em's con ex o use.
Speci y use equi emen s: In his phase, use needs a e iden i ied, and unc ional and o he
equi emen s o he p oduc a e speci ied. This p ocess conside s he con ex o use and aligns
wi h he sys em's business objec i es.
P oduce design solu ions o mee use equi emen s: Design solu ions a e de eloped based
on he con ex o use desc ip ion. This begins wi h designing use asks, use -sys em in e ac ion,
and he use in e ace o mee use equi emen s. Design solu ions become mo e angible
h ough scena ios, simula ions, p o o ypes, o mock-ups.
E alua e he designs agains equi emen s: Use -cen ed e alua ion is essen ial h oughou he
p ojec 's a ious s ages, s a ing om an ea ly s age. While he pa icipa ion o end-use s in he
e alua ion is highly desi able, al e na i e me hods ha do no equi e di ec use in ol emen can
be used when p ac ical cons ain s p e en hei pa icipa ion.
These ac i i ies i e a e un il he designed solu ion mee s use equi emen s.
Hyb id in elligence
Hyb id in elligence is he esul o he in eg a ion be ween human and machine in elligence. This
app oach enhances human capabili ies in he use in e ace design p ocess. In pa icula ,
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
66
gene a i e a i icial in elligence (GenAI) is gaining impo ance. This ype o a i icial in elligence
can c ea e p oduc s in o ms such as ex , images and ideos by using gene a i e models, o en
in esponse o humanly in e p e able commands ha desc ibe he ask o be pe o med o he
desi ed ou pu .
In 2023, a Spanish de elope named Ja i López c ea ed "Ang y Pumpkins" (Ma , 2024), a pa ody
o he popula game Ang y Bi ds, en i ely using a i icial in elligence ools like Midjou ney, DALL-
E 3, and GPT-4. The de elopmen p ocess in ol ed c ea ing g aphical elemen s such as
backg ounds, cha ac e s, and objec s, as well as gene a ing he sou ce code o he game. López
employed an i e a i e p ocess, c a ing a se ies o p omp s o achie e he desi ed ea u es wi hou
w i ing a single line o code. This example demons a es he po en ial o gene a i e AI o educe
ime and e o in de eloping complex digi al p oduc s. This is jus one example o how GenAI is
dis up ing p ac i ione s. Simila ly, he academic communi y has esponded wi h an eme ging bu
expanding body o esea ch on he implica ions o his echnology o he design p ocess.
The a ailable esea ch has explo ed bo h he posi i e and nega i e implica ions o in oducing
his new echnology (Li e al., 2024). On he posi i e side, GenAI o e s he po en ial o enhance
he capabili ies o design eams by educing solu ion de elopmen ime and enabling humans o
ocus on ac i i ies ha add g ea e alue and usabili y. Howe e , i au oma ion is no implemen ed
co ec ly, i may lead o he loss o he human ouch in inal design p oduc s. In esponse,
p oposals such as ha o Weisz e al. (2024) ha e eme ged, emphasizing he need o es ablish
design p inciples o GenAI applica ions. These au ho s p opose six p inciples:
Responsible Design: Ensu e ha he AI sys em add esses eal use p oblems and minimizes
po en ial ha m o use s.
Design o Men al Models: Communica e e ec i ely wi h use s on how o in e ac wi h he AI
sys em, conside ing hei backg ound and goals.
Design o App op ia e Con idence and Reliabili y: Help use s de e mine when o us o
ques ion he AI sys em’s esul s by encou aging scep icism ega ding quali y issues, inaccu acies,
biases, unde ep esen a ion, and o he po en ial p oblems.
Design o Gene a i e Va iabili y: Assis use s in managing he capabili y o gene a i e models
o p oduce mul iple dis inc and a ied esul s.
Design o Co-C ea ion: Enable use s o in luence he gene a i e p ocess and collabo a e
e ec i ely wi h he AI sys em.
Design o Impe ec ion: Help use s unde s and and wo k wi h esul s ha may no always mee
hei expec a ions.
S ige e al. (2023) ha e explo ed he impac o a i icial in elligence on he human-cen ed design
ac i i y cycle. In he con ex o he use unde s anding phase, AI is p ima ily used o he au oma ic
summa iza ion o esea ch indings in o a i ac s commonly used in subsequen phases, such as
use pe sonas and use scena ios. Du ing he equi emen s speci ica ion phase, in e ac ing wi h a
GenAI can assis he de elopmen eam in e ining he inal se o sys em equi emen s. Howe e ,
he applica ion o GenAI is mos concen a ed in he p oduc ion o design solu ions o mee use
equi emen s. In his s age, GenAI can gene a e p o o ypes ha a e i e a i ely modi ied o de elop
de ailed solu ions. GenAI can assis designe s in p og essing om low- ideli y solu ions (e.g.,
ske ches) o high- ideli y solu ions (e.g., GUI code) o op imize hei designs. Finally, in he
e alua ion phase, whe e designs a e assessed agains equi emen s, GenAI can au oma e he
de ec ion o common pi alls, allowing he es eam o ocus on iden i ying majo laws.
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
67
Bleichne and He mansson (2023) and Ho e al. (2024) ha e examined he applica ion o GenAI in
use in e ace design and iden i ied he main comme cial gene a i e assis ance ools: Galileo-AI,
Uiza d, and Midjou ney. Table 13 highligh s hei main ea u es. Howe e , i is impo an o no e
ha his lis is no exhaus i e, gi en he apid eme gence o new ools and he cons an e olu ion
o exis ing ones.
Table 13 - GenAI ools o use in e ace design
Tool
Desc ip ion
Uiza d (uiza d.io)
Uiza d is an AI-powe ed se ice designed o
assis in c ea ing use in e aces and in e ac i e p o o ypes.
In addi ion o
gene a ing in e aces om ex p omp s,
Uiza d can scan sc eensho s and wi e ames, con e ing
hem in o edi able mock-ups wi h cus omizable
componen s.
Galileo AI (usegalileo.ai)
Galileo AI, launched in 2023, is a ool o designing bo h
web and mobile use in e aces based on ex o image
p omp s. I gene a es high- ideli y mock-ups ha can be
expo ed as HTML code and edi ed in o he in e ace
design ools.
Midjou ney (midjou ney.com)
Midjou ney is a gene al-pu pose ex - o- image model
accessible ia Disco d. While i is p ima ily used o
gene a ing images om
ex p omp s, i s ou pu can also se e as high-quali y
p o o ypes o use in e ace design when p o ided wi h
well-c a ed p omp s.
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
68
Re e ences
Ahmed, U., Alexopoulos, C., Piange elli, M., & Polini, A. (2024). BRYT: Au oma ed keywo d
ex ac ion o open da ase s. In elligen Sys ems wi h Applica ions, 23, 200421.
h ps://doi.o g/10.1016/j.iswa.2024.200421
Ali, M., Papageo giou, G., Aziz, A., Loukis, E., Cha alabidis, Y., & Lopez Pellice , F., J., (2024).
Towa ds he De elopmen o In e ope able Open Da a Ecosys ems: Ha nessing he Technical,
Seman ic, Legal, and O ganiza ional (TSLO) In e ope abili y F amewo k. In P oceedings o he
25 h Annual In e na ional Con e ence on Digi al Go e nmen Resea ch (dg.o '24). Associa ion
o Compu ing Machine y, New Yo k, NY, USA, 909–919.
h ps://doi.o g/10.1145/3657054.3657160
Ande son, J., Sa ka , D., & Palen, L. (2019). Co po a e Edi o s in he E ol ing Landscape o
OpenS ee Map.
ISPRS In e na ional Jou nal o Geo-In o ma ion
,
8
(5), A icle 5.
h ps://doi.o g/10.3390/ijgi8050232
A enas, J., Ha emann, L., & P iego, E. (2015). Open Da a as Open Educa ional Resou ces:
Towa ds T ans e sal Skills and Global Ci izenship.
Open P axis
,
7
(4).
h ps://doi.o g/10.5944/openp axis.7.4.233
Baack, S. (2015). Da a ica ion and empowe men : How he open da a mo emen e-a icula es
no ions o democ acy, pa icipa ion, and jou nalism.
Big Da a & Socie y
,
2
(2),
2053951715594634. h ps://doi.o g/10.1177/2053951715594634
Bleichne , A., & He mansson, N. (2023). In es iga ing he use ulness o a gene a i e AI when
designing use in e aces (Disse a ion). h ps://u n.kb.se/ esol e?u n=u n:nbn:se:uu:di a-
504251
Boyles, J. L. (2020). Labo a o ies o news? Expe imen ing wi h jou nalism hacka hons.
Jou nalism
,
21
(9), 1338–1354. h ps://doi.o g/10.1177/1464884917737213
BRYT Keywo d Ex ac ion Gi Hub Reposi o y: Gi Hub - umai ahmedq/BRYT-AKE
Camb idge Dic iona y. (2024).
Mo i a ion
.
h ps://dic iona y.camb idge.o g/dic iona y/english/mo i a ion
Ca a a, W., Chan, Wae-San., Fische , Sande ., & Van-S eenbe gen, E. C ea ing alue h ough
open da a: S udy on he impac o e-use o public da a esou ces. Eu opean Commission,
2015.h ps://da a.eu opa.eu/si es/de aul / iles/edp_c ea ing_ alue_ h ough_open_da a_0.pd
Celis Va gas, A., Magnussen, R., Mulde , I., & La sen, B. (2023). Towa ds a amewo k o Open
Da a li e acy in educa ion: A sys ema ic mapping e iew o Open Da a skills and lea ning
app oaches.
In e ac ion Design and A chi ec u e(s)
,
57
, 133–151. h ps://doi.o g/10.55612/s-
5002-057-008
Cha alabidis, Yannis & Zuide wijk, Anneke & Alexopoulos, Cha alampos & Janssen, Ma ijn &
Lampol shamme , Thomas & Fe o, En ico. (2018). The Wo ld o Open Da a - Concep s,
Me hods, Tools and Expe iences.
Cha apadhyay, S. (2014). Access and use o go e nmen da a by esea ch and ad ocacy
o ganisa ions in India: A su ey o (po en ial) open da a ecosys em.
P oceedings o he 8 h
In e na ional Con e ence on Theo y and P ac ice o Elec onic Go e nance
, 361–364.
h ps://doi.o g/10.1145/2691195.2691262
Chicaiza, J., Pied a, N., Lopez-Va gas, J., & To a -Ca o, E. (2017). Recommenda ion o open
educa ional esou ces. An app oach based on linked open da a.
2017 IEEE Global Enginee ing
Educa ion Con e ence (EDUCON)
, 1316–1321. h ps://doi.o g/10.1109/EDUCON.2017.7943018
Concilio, G., & Mulde , I. (2018).
Open4Ci izens Policy b ie
. OPEN4CITIZENS - Empowe ing
ci izens o make meaning ul use o open da a. h p://open4ci izens.eu/wp-
con en /uploads/2016/01/O4C_D4.8_30.06.2018_Final.pd
Cook, K., Caki la , C., Godda d, T., Demu h, R. C., & Wells, J. (2018). Teaching Open Science:
Published Da a and Digi al Li e acy in A chaeology Class ooms.
Ad ances in A chaeological
P ac ice
,
6
(2), 144–156. h ps://doi.o g/10.1017/aap.2018.5
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
69
Co be , J., Templie , M., & Takeda, H. (2018, Janua y 3). The Making o a “Top” Open Da a Ci y:
A Case S udy o Edmon on’s Open Da a Ini ia i e.
P oceedings o he 51s Hawaii In e na ional
Con e ence on Sys em Sciences
. h ps://doi.o g/10.24251/HICSS.2018.308
Co cho, O., Jiménez Blanco, J., Mo o e, C., and Simpe l, E. (2022). Da a.eu opa.eu and Ci izen-
gene a ed Da aOppo uni ies and challenges associa ed o he inclusion o ci izen-gene a ed
da a in da a.eu opa.eu. Luxembou g: Publica ions O ice o he Eu opean Union.
h ps://da a.eu opa.eu/si es/de aul / iles/ epo /da a.eu opa.eu_Repo _Ci izen-
gene a edda aonda a_eu opa_eu.pd
Coughlan, T. (2020). The use o open da a as a ma e ial o lea ning.
Educa ional Technology
Resea ch and De elopmen
,
68
(1), 383–411. h ps://doi.o g/10.1007/s11423-019-09706-y
C ea i e Commons. (2023).
CC BY-NC-SA 4.0 Legal Code | A ibu ion-NonComme cial-
Sha eAlike 4.0 In e na ional | C ea i e Commons
. h ps://c ea i ecommons.o g/licenses/by-nc-
sa/4.0/legalcode
Da ase e
. (n.d.). Da ase e: An open sou ce mul i- ool o explo ing and publishing
da a. h ps://da ase e.io
Da ies, T. (2011). Open Da a: In as uc u es and ecosys ems.
ACM Web Science Con e ence
.
Da is, K., & Shneye , E. (2020). Compu e science as a ool o de eloping u u e ci ic change-
make s.
SIGCSE 2020 - P oceedings o he 51s ACM Technical Symposium on Compu e Science
Educa ion
, 1344. h ps://doi.o g/10.1145/3328778.3372636
Enaholo, P. (2017). Beyond me e ad ocacy: CSOs and he ole o in e media ies in Nige ia’s open
da a ecosys em.
The Social Dynamics o Open Da a
. h ps://zenodo.o g/ eco ds/1117797
Enaholo, P., & Dina, D. (2020). Jou nalis s and he in e media ion o open da a: A Nige ian
pe spec i e. In
Si ua ing Open Da a: Global T ends in Local Con ex s
(pp. 31–49). A ican Minds.
h ps://www.scienceopen.com/hos ed-documen ?doi=10.47622/978-1-928502-12-8_2
Wha is Epicollec 5 | Epicollec 5 Use Guide
. (n.d.). Wha is Epicollec 5 | Epicollec 5 Use
Guide. h ps://docs.epicollec .ne
Falk, J. (2022).
How Game Jams and Hacka hons Accele a e Design P ocesses
(a Xi :2205.04966).
a Xi . h ps://doi.o g/10.48550/a Xi .2205.04966
GO FAIR ini ia i e. (n.d.). FAIR p inciples - GO FAIR. GO FAIR. h ps://www.go- ai .o g/ ai -
p inciples/
Goodchild, M.F. Ci izens as senso s: he wo ld o olun ee ed geog aphy. (2007). GeoJou nal 69,
211–221. h ps://doi.o g/10.1007/s10708-007-9111-y
González-Zapa a, F., & Heeks, R. (2015, May). Unde s anding mul iple oles o in e media ies in
Open Go e nmen Da a.
P oceedings o he 13 h In e na ional Con e ence on Social
Implica ions o Compu e s in De eloping Coun ies
.
G ay, J., Bouneg u, L., & Chambe s, L. (2012).
The Da a Jou nalism Handbook
(1s ed.). O’Reilly
Media, Inc.
He e a-Mu illo, D. J., Aziz, A., Nogue as-Iso, J., & Lopez-Pellice , F. J. Analysing use in ol emen
in open go e nmen da a ini ia i es. (2022). in
P oc. In . Con . Theo y P ac . Digi .
Lib a ies, pp.
175–186. h ps://doi.o g/10.1007/978-3-031-16802-4_14
He e a-Mu illo, D. J., Ochoa-O iz, H., Ahmed, U., López-Pellice , F. J., Re, B., Polini, A., and
Nogue as-Iso, J. (2024). P ocess Analysis in Humani a ian Volun a y Geog aphic In o ma ion: he
case o he HOT Tasking Manage ,
AGILE GIScience Se .
, 5, 5. h ps://doi.o g/10.5194/agile-giss-
5-5-2024
Ho, C., Liu, X., Qiu, Y., and Yang, S. (2024). Resea ch on Inno a i e Applica ions and Impac s o
Using Gene a i e AI o Use In e ace Design in P og amming Cou ses. In P oceedings o he
2024 In e na ional Con e ence on In o ma ion Technology, Da a Science, and Op imiza ion (I-
DO '24). Associa ion o Compu ing Machine y, New Yo k, NY, USA, 68–72.
h ps://doi.o g/10.1145/3658549.3658566
Implemen ing egula ion - 2023/138 - EN - EUR-Lex
. (n.d.). EUR-Lex — Access o Eu opean
Union law — choose you language. h ps://eu -lex.eu opa.eu/legal-
con en /EN/TXT/?u i=u ise :OJ.L_.2023.019.01.0043.01.ENG

D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
70
ISO 9241-210:2010, (2010). E gonomics o human-sys em in e ac ion - Pa 210: Human-cen ed
design o in e ac i e sys ems. S anda d. In e na ional O ganiza ion o S anda diza ion. Gene a,
CH. h ps://www.iso.o g/s anda d/77520.h ml
Hou, Y., & Wang, D. (2017). Hacking wi h NPOs: Collabo a i e Analy ics and B oke Roles in Ci ic
Da a Hacka hons.
P oceedings o he ACM on Human-Compu e In e ac ion
,
1
(CSCW), 53:1-
53:16. h ps://doi.o g/10.1145/3134688
Jaskiewicz, T., Mulde , I., Mo elli, N., & Pede sen, J. (2019). Hacking he hacka hon o ma o
empowe ci izens in ou sma ing “sma ” ci ies.
In e ac ion Design and A chi ec u e(s)
,
43
, 8–29.
h ps://doi.o g/10.55612/s-5002-043-001
Kuble , S., Robe , J., Neumaie , S., Umb ich, J., & Le T aon, Y. (2018). Compa ison o me ada a
quali y in open da a po als using he Analy ic Hie a chy P ocess.
Go e nmen In o ma ion
Qua e ly
, 35(1), 13 - 29. h ps://doi.o g/10.1016/j.giq.2017.11.003
Ki sios, F., & Kama io ou, M. (2018). Open da a hacka hons: An inno a i e s a egy o enhance
en ep eneu ial in en ion.
In e na ional Jou nal o Inno a ion Science
,
10
(4), 519–538.
h ps://doi.o g/10.1108/IJIS-06-2017-0055
Li, J., Cao, H., Lin, L., Hou, Y., Zhu, R., & El Ali, A. (2023). Use Expe ience Design P o essionals’
Pe cep ions o Gene a i e A i icial In elligence. P oceedings o he CHI Con e ence on Human
Fac o s in Compu ing Sys ems. h ps://doi.o g/10.1145/3613904.3642114
Loda o, T. J., & DiSal o, C. (2016). Issue-o ien ed hacka hons as ma e ial pa icipa ion.
New
Media & Socie y
,
18
(4), 539–557. h ps://doi.o g/10.1177/1461444816629467
Meije , A., & Po je , S. (2018). Ci izen-gene a ed open da a: An explo a i e analysis o 25 cases.
Go e nmen In o ma ion Qua e ly
,
35
(4), 613–621. h ps://doi.o g/10.1016/j.giq.2018.10.004
Linked Open Vocabula ies (LOV).
(n.d.). Linked Open
Vocabula ies. h ps://lo .linkedda a.es/da ase /lo /
Loukis, E. & Saxena, S u i & Rizun, Nina & Ma a si, Ma ia & Ali, Mohsan & Alexopoulos,
Cha alampos. (2023). Cha GPT Applica ion is-a- is Open Go e nmen Da a (OGD): Capabili ies,
Public Values, Issues and a Resea ch Agenda. 10.1007/978-3-031-41138-0_7.
Lou sa is, Michalis A ge inos & Alexopoulos, Cha alampos & Ma a si, Ma ia & Cha alabidis,
Yannis. (2023). Seman ic In e ope abili y o Legal In o ma ion: Mapping he Eu opean
Legisla ion Iden i ie (ELI) and Akoma N oso (AKN) On ologies. 41-53.
10.1145/3614321.3614327.
Ma a si, Ma ia & Ahmed, Umai & Alexopoulos, Cha alampos & Cha alabidis, Yannis & Polini,
And ea. (2024 b). Towa ds C oss-Domain Linking o Da a: A Seman ic Mapping o Cul u al
He i age On ologies. 165-176. 10.1145/3657054.3657077
Ma a si, Ma ia & Alexopoulos, Cha alampos & Cha alabidis, Yannis. (2024 d). A S uc u ed
Analysis o Domain-Speci ic Linked Open Vocabula ies (LOV): Indica o s o In e ope abili y and
Reusabili y. 10.1007/978-3-031-56478-9_10.
Ma a si, Ma ia, Alexopoulos, Cha alampos & Cha alabidis, Yannis. (2024 a). On he Seman ic
Analysis o Open (Go e nmen ) Da a Po als’ Me ada a P o ision and Schema. ( o be published
in a ew weeks in he ICEGOV’24 ACM p oceedings)
Ma a si, Ma ia, Gialoussi, Nina, Alexopoulos, Cha alampos & Cha alabidis, Yannis. (2024 c). A
P oposed Me hodology o Sub-On ology De elopmen in Comp ehensi e Scien i ic
In es iga ion Me hods and Tooling ( o be published in a ew weeks in he MTSR’24 Sp inge
p oceedings)
Ma a si, Ma ia & Lachana, Zoi & Alexopoulos, Cha alampos & Cha alabidis, Yannis. (2023). A
unc ional and seman ic analysis o a i ac ep esen a ion schema a in olklo e museum
websi es. In e na ional Jou nal o Me ada a, Seman ics and On ologies. 16. 105-117.
10.1504/IJMSO.2023.135339.
Ma , B. (2024). Gene a i e AI in P ac ice: 100+ Amazing Ways Gene a i e A i icial In elligence is
Changing Business and Socie y. John Wiley & Sons.
Meye , P. (2002).
P ecision Jou nalism: A Repo e ’s In oduc ion o Social Science Me hods
(4 h
ed.). Rowman & Li le ield Publishe s.
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
71
Na alkha, C. (2021).
Da a o Housing Jus ice: Examining Ac i is s’ Use o Open Go e nmen
Da a o Housing Jus ice in Bos on, MA and New Yo k, NY
[Thesis, Massachuse s Ins i u e o
Technology]. h ps://dspace.mi .edu/handle/1721.1/138973
ODK - Collec da a anywhe e
. (n.d.). ODK - Collec da a anywhe e. h ps://ge odk.o g
Opens ee Map Founda ion. (2024).
Co po a e Membe s—OpenS ee Map Founda ion
.
h ps://osm ounda ion.o g/wiki/Co po a e_Membe s
OpenS ee Map Wiki. (2024).
O ganised Edi ing/Ac i i ies—OpenS ee Map Wiki
.
h ps://wiki.opens ee map.o g/w/index.php? i le=O ganised_Edi ing/Ac i i ies&oldid=2656488
Palo a, D., & Vejacka, M. (2022). Expe ience wi h Open Da a in P ojec Based Lea ning.
2022
45 h Jubilee In e na ional Con en ion on In o ma ion, Communica ion and Elec onic
Technology, MIPRO 2022 - P oceedings
, 689–694.
h ps://doi.o g/10.23919/MIPRO55190.2022.9803595
Papageo giou, G., Eu ipides, L., Magnussen, R., & Cha alabidis, Y. (2023). Open da a jou nalism:
A domain mapping e iew.
P oceedings o he 16 h In e na ional Con e ence on Theo y and
P ac ice o Elec onic Go e nance
, 159–166. h ps://doi.o g/10.1145/3614321.3614340
Pa k, C. H., Longo, J., & Johns on, E. W. (2020). Explo ing Non-S a e S akeholde and
Communi y-Led Open Go e nance: Beyond he Th ee Pilla s o Open Go e nmen .
Public
Pe o mance & Managemen Re iew
,
43
(3), 587–612.
h ps://doi.o g/10.1080/15309576.2019.1677253
Pelleg ino, M., & An elmi, A. (2023). A School o Open Da a: A Li e a u e Re iew:
P oceedings o
he 15 h In e na ional Con e ence on Compu e Suppo ed Educa ion
, 172–183.
h ps://doi.o g/10.5220/0011747500003470
Pence, H. E., Williams, A. J., & Bel o d, R. E. (2015). New Tools and Challenges o Chemical
Educa ion: Mobile Lea ning, Augmen ed Reali y, and Dis ibu ed Cogni ion in he Dawn o he
Social and Seman ic Web. In
Chemis y Educa ion: Bes P ac ices, Oppo uni ies and T ends
.
h ps://doi.o g/10.1002/9783527679300.ch28
Poikola, A., Kola, P., & Hin ikka, K. A. (2010).
Public da a: An in oduc ion o opening in o ma ion
esou ces
. Minis y o T anspo and Communica ions.
Publica ions O ice o he Eu opean Union. Open da a ma u i y epo 2023. Publica ions O ice,
LU, 2023. h ps://da a.eu opa.eu/si es/de aul / iles/odm2023_ epo .pd
QField - E icien ield wo k buil o QGIS
. (n.d.). QField - E icien ield wo k buil o
QGIS. h ps://q ield.o g
Ricke , B., Cinnamon, J., & Die wech e , Y. (2020). When open da a and da a ac i ism mee : An
analysis o ci ic pa icipa ion in Cape Town, Sou h A ica.
Canadian Geog aphies / Géog aphies
Canadiennes
,
64
(3), 359–373. h ps://doi.o g/10.1111/cag.12608
Roge s, S. (2008, Decembe 18). Tu ning o icial igu es in o unde s andable g aphics, a he
p ess o a bu on.
The Gua dian
.
h ps://www. hegua dian.com/help/insidegua dian/2008/dec/18/unemploymen da a
Roge s, S. (2010, Augus 13). Flo ence Nigh ingale, da ajou nalis : In o ma ion has always been
beau i ul.
The Gua dian
. h ps://www. hegua dian.com/news/da ablog/2010/aug/13/ lo ence-
nigh ingale-g aphics
Roge s, S. (2011, Sep embe 26).
The i s Gua dian da a jou nalism: May 5, 1821
. The Gua dian.
h p://www. hegua dian.com/news/da ablog/2011/sep/26/da a-jou nalism-gua dian
Rome o, M., Usa , M., & O , M. (2015). Can Se ious Games Con ibu e o De eloping and
Sus aining 21s Cen u y Skills?
Games and Cul u e
,
10
(2), 148–177.
h ps://doi.o g/10.1177/1555412014548919
Runeson, P., Olsson, T., Linåke , J., (2021), Open Da a Ecosys ems — An empi ical in es iga ion
in o an eme ging indus y collabo a ion concep , Jou nal o Sys ems and So wa e, Volume 182,
111088, ISSN 0164-1212, h ps://doi.o g/10.1016/j.jss.2021.111088.
Saddiqa, M., La sen, B., Magnussen, R., Rasmussen, L. L., & Pede sen, J. M. (2019). Open da a
isualiza ion in danish schools: A case s udy.
Jou nal o WSCG
,
2019
(WSCG2019CS), 17–26.
h ps://doi.o g/10.24132/CSRN.2019.2902.2.3
D4.2 An app oach o s ee he beha iou o non-go e nmen da a holde s owa ds open da a
h ough a echnical s a egy
72
Saddiqa, M., Magnussen, R., La sen, B., & Pede sen, J. M. (2021). Digi al inno a ion in educa ion:
Pe spec i es, oppo uni ies and challenges o educa ional open da a and senso da a.
CEUR
Wo kshop P oceedings
,
2991
, 74–83.
Saddiqa, M., Rasmussen, L., Magnussen, R., La sen, B., & Pede sen, J. M. (2019). B inging open
da a in o Danish schools and i s po en ial impac on school pupils.
P oceedings o he 15 h
In e na ional Symposium on Open Collabo a ion, OpenSym 2019
.
h ps://doi.o g/10.1145/3306446.3340821
Salamon, L. M., & Anheie , H. K. (1992). In sea ch o he non-p o i sec o . I: The ques ion o
de ini ions.
Volun as: In e na ional Jou nal o Volun a y and Nonp o i O ganiza ions
,
3
(2), 125–
151. h ps://doi.o g/10.1007/BF01397770
Shaha udin, A., an Loenen, B., & Janssen, M. (2023). Towa ds a Common De ini ion o Open
Da a In e media ies.
Digi al Go e nmen : Resea ch and P ac ice
,
4
(2), 6:1-6:21.
h ps://doi.o g/10.1145/3585537
S ige, Å., Zamani, E.D., Mikale , P., & Zhu, Y. (2023). A i icial in elligence (AI) o use expe ience
(UX) design: a sys ema ic li e a u e e iew and u u e esea ch agenda. In o ma ion Technology
& People. h ps://doi.o g/10.1108/ITP-07-2022-0519
Thema ic Anno a ion o open Go e nmen da a, Gi Hub eposi o y c ea ed by IAAA lab,
Uni e si y o Za agoza, Spain, h ps://gi hub.com/IAAA-
Lab/Thema ic_Anno a ion_o _Go e nmen _Da a, accessed on 26/09/2024
uda a
. (n.d.). PyPI. h ps://pypi.o g/p ojec /uda a/
Vallejo-Figue oa, S., Rod iguez-A acho, M., Cas o-Gil, M., & C is obal, E. S. (2018). Using ex
mining and linked open da a o assis he mashup o educa ional esou ces.
2018 IEEE Global
Enginee ing Educa ion Con e ence (EDUCON)
, 1606–1611.
h ps://doi.o g/10.1109/EDUCON.2018.8363427
an Loenen, B., Vancauwenbe ghe, G., & C omp oe s, J. (Eds.). (2018).
Open Da a Exposed
.
T.M.C. Asse P ess. h ps://link.sp inge .com/book/10.1007/978-94-6265-261-3
an Loenen, B., Zuide wijk, A. ., Vancauwenbe ghe, G., Lopez-Pellice , F. J. ., Mulde , I.,
Alexopoulos, C. ., Magnussen, R., Saddiqa, M., Dulong de Rosnay, M., C omp oe s, J., Polini, A.,
Re, B., & Casiano Flo es, C. (2021). Towa ds alue-c ea ing and sus ainable open da a
ecosys ems: A compa a i e case s udy and a esea ch agenda. JeDEM - EJou nal o EDemoc acy
and Open Go e nmen , 13(2), 1–27. h ps://doi.o g/10.29379/jedem. 13i2.644 Weisz, J.D., He, J.,
Mulle , M., Hoe e , G., Miles, R., & Geye , W. (2024). Design P inciples o Gene a i e AI
Applica ions. P oceedings o he CHI Con e ence on Human Fac o s in Compu ing Sys ems.
h ps://doi.o g/10.48550/a Xi .2301.05578
Wol , A., Ca e o Mon ane , J. J., & Ko uem, G. (2016). U ban Da a in he p ima y class oom:
B inging da a li e acy o he UK cu iculum.
The Jou nal o Communi y In o ma ics
,
12
(3).
h ps://doi.o g/10.15353/joci. 12i3.3278
Zaman, H. B., Baha in, H., & Ahmad, A. (2021). Fusion Technology and Visualisa ion o Sha e
STEM Da a Using PETS Robo s (i-COMEL) o Open Da a Readiness Amongs P ima y School
Child en. In
Lec u e No es in Compu e Science (including subse ies Lec u e No es in A i icial
In elligence and Lec u e No es in Bioin o ma ics): Vol. 13051 LNCS
. h ps://doi.o g/10.1007/978-
3-030-90235-3_16