Extending BPM(N) to Support Face-to-Virtual (F2V) Process Modeling

Sasha Mile Rudan

1,2 a

, Sinisa Rudan

and Birger Møller-Pedersen

University of Oslo, Oslo, Norway

ChaOS, Belgrade, Serbia

Keywords:

BPM, Face-to-Virtual Process, Hybrid Process, BPMN Extension, COVID-19 Digital Disruption, Creative

Processes Digitalization.

Abstract:

In this paper, we present research on CoPI4P (Community of Practice & Interest for Purpose) communities that

consist of knowledge and creative workers in domains such as education, environmentalism, inter-disciplinary

research, engaged art, and practice F2V (face-to-virtual) processes, i.e. processes that consist of both face-to-

face and virtual (online) activities. Standards and methods for modeling business processes have successfully

solved various crucial problems, such as providing boundary objects (or a common language) between busi-

ness analysts, domain experts, process performers and business solutions developers. Performers can execute

business processes with standards that support the execution of such processes. However, for CoPI4P com-

munities, BPM standards 1) require additional domain specialization and 2) are either too imperative, or fail

to provide enough guidance. This paper identiﬁes these challenges and provides partial solutions applicable

to modeling and supporting CoPI4P communities and the corresponding F2V business processes. In addition,

results of focus groups interviews and surveys are provided that shed light, notably, on the way CoPI4P com-

munities have been impacted by COVID-19 and how they have coped with digital disruptions to their working

models. Our contribution to business processes management lies in researching and identifying the peculiar-

ities of such communities and their workﬂows, followed by interventions in the BPM, and particularly the

BPMN, domain. We then propose a toolset of BPMN extensions required to match the observed F2V work-

ﬂows and thus to digitalize CoPI4P business models. In this context, we introduce the notion of sub-process

palettes as a means to reduce the rigidity, and introduce personalization, of processes.

1 INTRODUCTION

With the rise of knowledge and creative workers and

creative economy, we face new businesses and the

challenges of modeling them (Loo, 2016). In contrast

to conventional business models and the correspond-

ing business processes, these new business models

are difﬁcult to model in a deterministic and prescrip-

tive way (Swenson, 2010). For knowledge and cre-

ative workers, working procedures become increas-

ingly blurry, business performers are no longer per-

formers that merely execute (human) tasks; in such

business scenarios, they expect freedom of choice re-

garding, notably, their decisions, methodologies and

time handling, to name just some aspects of their

work (Loo, 2016).

Our work focuses on the aforementioned business

https://orcid.org/0000-0002-9815-5996

∗

Face-to-Virtual is a Combination of Face-to-Face and

Virtual (Online) Activities.

scenarios and systems that help communities geared

to solve complex interdisciplinary problems in face-

to-virtual (F2V) events. Such events take the form

of rapidly changing activities happening in real-time

with various frictions as each activity can involve dif-

ferent media (face, virtual, hybrid), methodologies,

and/or participants. They involve participants who

work in a non-coordinated manner and with full cre-

ative freedom. The participants often work ofﬂine

with results eventually provided (digitized) online.

This is why we use the term face-to-virtual (F2V) for

such events and processes.

At the same time, the communities these partic-

ipants belong to (i) design, (ii) evaluate, (iii) evolve,

and (iv) practice various methodologies that are inher-

ently well structured and imperative in their intention

toward the participants and the communities. This

digitalization dissonance between the non-imperative

behavior of the performers and the imperative expec-

tations of the community conducted processes pro-

vides a great challenge in terms of the digitization of

350

Rudan, S., Rudan, S. and Møller-Pedersen, B.

Extending BPM(N) to Support Face-to-Virtual (F2V) Process Modeling.

DOI: 10.5220/0010349203500361

In Proceedings of the 9th International Conference on Model-Driven Engineering and Software Development (MODELSWARD 2021), pages 350-361

ISBN: 978-989-758-487-9

such community workﬂows.

In this paper, we present our work on evaluating

a visual workﬂow language (particularly BPMN) as

a way of (BPM) modeling of the working practices

of such communities, evaluating their beneﬁts over

non-formal and non-semantic encoding of their work-

ﬂows. We research and explore community practices

and challenges with non-digitalized workﬂow prac-

tices. Afterwards, we evaluate the attitudes of the

communities stakeholders’ toward the introduction of

visual workﬂows and toward the extension of BPMN

we proposed as a way to properly digitalize their busi-

ness processes. We discuss both approaches for busi-

ness digitalization; (i) top-down, conducted by busi-

ness and domain experts, and (ii) bottom-up, con-

ducted either by community members through par-

ticipatory design, or through ecosystem augmentation

(e.g. process-mining).

This paper is structured as follows. First, we in-

troduce the necessary terminology (sec. 2) followed

by presenting the research challenges involved (sec.

3). After that, we present our contribution with rel-

evant research and ﬁndings on the types of commu-

nities and processes discussed in this study (sec. 5).

We then present the proposed BPMN extensions (sec.

7) and introduce new BPM(N) mechanisms - sub-

process palettes to reduce the rigidity, and introduce

personalization, of processes (sec. 8) and conclude

the paper (sec. 9).

2 RELEVANT TERMINOLOGY

Here we introduce the relevant terminology including

standard BPM terms, terms from other relevant dis-

ciplines and new terms that we introduced to explain

new concepts.

On using the ”business” Denominator: Our re-

search focuses on exploring and supporting work-

ﬂows that prominently depend on participants’ cre-

ative and free forms of work. However, the business

”tone” does not fully suit either of these kinds of

work, or the communities conducting it. Therefore,

instead of contextualizing each aspect of such work

and corresponding processes with the ”business” de-

nominator, we prefer to refer to them either without

the business denominator or with a more speciﬁc de-

nominator, e.g. social, creative or innovative.

Newly Introduced Terms:

Trans-domain Ecosystems: we refer to open so-

ciotechnical systems partially encoded through the

digital world, partially through the external physical

world, and partially through the social world of the

community members.

(Eco)system Values: ecosystem artifacts that

ecosystem domain experts have recognized as impor-

tant for the speciﬁc purpose of the ecosystem and the

community or business it stands for.

CoPI4P (Community of Practice & Interest for

Purpose) (pronounced as ”Copy for P”): a com-

plex and heterogeneous community of practitioners

(hence CoP

) and participants interested (hence CoI

)

in the topic of the community and its purpose (hence

CoP

) ((Chatzinotas, 2017)).

Face-to-Virtual (F2V) Process: a process that

combines face-to-face and virtual (online) activities.

The events consisting of both kind of activities are

also called hybrid (Bonakdarian et al., 2009), but the

F2V stresses the interaction across different ”worlds”

and the fact that the F2V activities affect both virtual

and physical worlds and artifacts in them. The work-

shops and Participatory action research (PAR) ses-

sions conducted and presented in the course of this

work consist of F2V events orchestrated with such

face-to-virtual processes.

Sub-process Palette: a collection of sub-

processes that are functionally semi-equivalent, but

with different behavioral footprint and matching each

of community persona. They are functionally semi-

equivalent in the sense that the differences in their

performance outcomes (mostly in the regard of the

ecosystem values) are negligible from the perspec-

tive of domain-experts or the community performing

them, and as such they are perceived as equivalent and

used in an interchangeable manner.

3 CoPI4P CHALLENGES

Modeling business processes is an inherent part of

any business analysis work. In the recent decades,

tools supporting this work expanded (Indulska et al.,

2009), with the possibility of digitization of business

processes and describing business processes through

digital business models. The robust and widely

used business process modeling language is BPMN

(Recker, 2008). BPMN has proven itself as a solid

language for modeling business processes, describing

For general info please check on the Community

of Practice: https://en.wikipedia.org/wiki/Community of

practice

For general info please check on the Community

of Interest: https://en.wikipedia.org/wiki/Community of

interest

For general info please check on the Community

of Purpose: https://en.wikipedia.org/wiki/Community of

purpose

Extending BPM(N) to Support Face-to-Virtual (F2V) Process Modeling

351

each particular role in multi-actor business processes,

activities and points of collaboration and messages

exchange.

This precision in process description has helped

business analysts to design accurate processes that

business participants can follow so as to contribute

their results and progress. The inherent association

and alignment of the executed activities with the re-

sults of activities execution enables business analysts

to evaluate the work being done, notice outstanding

and problematic parts of work and compare it with the

past work of the same or other relevant business ac-

tor. This, in short, explains the feedback loop in mod-

ern business models, and the beneﬁts of using struc-

tured models like BPMN and argues the beneﬁt for

CoPI4P communities embracing BPM digitalization

practices.

However, the strict (and generic) nature of the

BPMN language is not adequate for domain-speciﬁc

businesses; the language lacks the vocabulary to ex-

press speciﬁc domain needs. This is the reason for

business communities of practice (CoPs) to extend the

BPMN standard to better ﬁt their speciﬁc domains

(Bocciarelli and D’Ambrogio, 2011). In the same

way as UML provides extension mechanisms through

stereotypes, tags and constraints (Koch and Kraus,

2002), the BPMN 2.0 metamodel provides a set of ex-

tension elements, which allows to add new attributes

to existing BPMN elements and new elements types

to the standard. This allows BPMN adopters to ex-

pand the original BPMN metamodel with necessary

extensions to ﬁt a particular business model.

Such BPMN extensions allow for domain-speciﬁc

constructs while still keeping the digitized domain-

speciﬁc business processes BPMN-compliant. Social

BPM (Social Business Process Management) (Kems-

ley, 2011) demonstrates another relevant example of

a need for BPMN extensions: an approach to encap-

sulate important social aspects of business processes.

The main focus of the Social BPM is on extend-

ing BPM to support social activities, actors, events

(as specializations to the corresponding BPMN en-

tities) and semantically integrate them into conven-

tional business processes (Brambilla et al., 2011) (but

also keeping them executable).

In a similar way, we need to describe and encode

unique aspects of CoPI4P business workﬂows. We

need to encode the dynamic nature of their work, and

explicitly support their creative freedom (which we

will address in the sec. 8) to chose their optimal way

of performing the expected work.

We need to describe the ”soft-aspects” of each

business process activity (like motivation sources,

space for creative freedom, frictions, etc) to increase

efﬁcacy of multiple stakeholders; facilitators need

to understand the context of the performing activity,

participants (performers) need to understand work,

available freedom and social capacity of the activity,

ecosystem has to be able to properly coordinate and

visualize the performing activity.

4 RESEARCH METHODOLOGY

Our investigation of visual conceptual modeling lan-

guage notations (Bork et al., 2018), (Proper et al.,

2018), (Bork et al., 2020), (Pankowska, 2019) is be-

yond the scope of this study, so we will limit our-

selves to the most important reasons for choosing

BPMN over other standards. We compared BPMN,

CMMN, ArchiMate, UML, and IFML. BPMN came

as the winner as (i) it is aligned with other OMG

standards (as CMMN, DMN, VDML), enabling sup-

port for declarative case business management, (ii) it

seems to be more widely supported in frontend open

space, (iii) it is executable.

In our work with CoPI4P (Community of Prac-

tice & Interest for Purpose) communities, we have fo-

cused on 3 types of stakeholders: (i) domain experts

(organizer, leaders, facilitators), (ii) regular commu-

nity members (members, participants, followers), and

(iii) (eco)system architects (designers, developers).

The fundamental research methodologies prac-

ticed across our work with CoPI4P communities are

Participatory action research (PAR) ((Chevalier and

Buckles, 2019)) and Participatory Design ((Brat-

teteig and Wagner, 2014), (Simonsen and Robertson,

2012)). Due to the nature of these communities and

the fact that their work is concerned with group dy-

namics, our study of the majority of the communities

involved started, in its ﬁrst phase, with unobtrusive

participation and observation. This gave us an initial

understanding of their work, goals and eventually the

challenges they faced.

In the second phase, we started with more en-

gaged Participatory action research (PAR), whereby

we practiced CoPI4P communities’ methodologies

and contributed to their ecosystem values. This devel-

oped mutual trust, familiarity with community mem-

bers and their practice, and a chance to co-design.

In the third phase, we engaged with the ﬁrst

group of stakeholders - domain experts. We practiced

focus group interviews with domain experts. This

gave us an additional understanding of our research

scope through both formative evaluation (before the

research was conducted) and summative evaluation of

our research contributions.

The participatory design work happened at two

MODELSWARD 2021 - 9th International Conference on Model-Driven Engineering and Software Development

352

levels. The ﬁrst is the community level, where

we co-designed community methodologies, practices

and particular workshops and events with commu-

nity members. The second one is the augmentation

level where we were augmenting community prac-

tices through digitized business processes.

Finally, to gain more precise and uniﬁed data from

the ﬁrst group of stakeholders - facilitators - we con-

ducted surveys evaluating their needs (e.g. what type

of BPMN extensions they needed) and their satisfac-

tion with our contributions (e.g. introducing BPM

into their work practices).

5 RESEARCH AND FINDINGS

We base our work on our action research enrolment

with CoPI4P (Community of Practice & Interest for

Purpose) communities. These are communities that

practice with new forms of research and education,

engaged artists and poets, youth organizations, cli-

mate changes and ecology related organizations. To

give a quick understanding of their scope we orga-

nized them in the following 5 categories:

i. Eco & Sustain. Dev.: sevaral intl. events, in-

cluding Climathon hackathons for climate chal-

lenges, happening each year in 55 countries.

ii. Engaged Art (Multiple): communities that

use various methodologies, media and trans-

disciplinary initiatives combining art, science and

knowledge to achieve social change.

iii. Digital Humanities (Multiple): a ﬁeld compris-

ing communities of researchers that are inher-

ently interdisciplinary, bridging two different dis-

ciplines. Authors design their tools and work-

ﬂows.

iv. Educational (Multiple): communities that work

on innovative models of education, such as

model-based and paradox-based education, edu-

cational games and lifelong learning.

v. Knowledge Federation (KF): international com-

munity gathered around the challenges of knowl-

edge federation across interdisciplinary commu-

nities.

Through PAR sessions and focus group interviews,

we have captured the following challenges posed by

non-digitalized workﬂow practices (expressed as ”It

is hard to ...”): (i) communicate workﬂows with

other stakeholders, (ii) monitor the evolutionary de-

velopment of workﬂows, (iii) evaluate workﬂows, (iv)

claim improvements of either community or ecosys-

tem values achieved through conducted workshops

(critical for commercial workshops), (v) ensure the

coordination of other facilitators by top level facili-

tators, understand and communicate which group dy-

namic results have already been achieved with the

help of the previous activities in the workﬂow (for

example, to determine if participants already share a

common vision).

There is no real application of BPM practice in

these business scenarios and communities, and open

source communities and industries do not seem to be

interested in using the BPM standards and modeling

to augment such communities and processes.

We identify the characteristics of CoPI4P commu-

nities that constitute a challenge in implementing their

workﬂows, conceptualize them in terms of the associ-

ated process models and summarize them later in the

section (table 1).

The challenges are largely due the fact that knowl-

edge and creative workers are not performers that

merely ”execute” precisely prescribed (human) tasks

designed for them and given to them to execute. Such

workers want to organize and handle their own time

as they usually work with no time limitation (in or-

der to foster the process of creativity). They often

practice open innovation, discover their own business

peers (B2B), and ﬁnd innovative ways to reach cus-

tomers (B2C). Such workers rarely work in isolation.

Rather, we can often model them as part of a CoPI4P

(Community of Practice & Interest for Purpose), a

wise-crowd actively interacting and continuously con-

tributing to a shared corpus of knowledge, ideas, prac-

tices, or whatever the ecosystem-values of the com-

munity they belong to happen to be. Therefore, we

need to enable business scenarios that provide such

communities to contribute their work in F2V (face-

to-virtual) events (either real-time or asynchronous),

which additionally complicates business modeling in

this context.

The above-mentioned business scenarios argue

against the rigid and imperative business process

modeling. However, such workﬂows are mostly gov-

erned with speciﬁc creative work methodologies (like

Rapid Foresight methodology, model based educa-

tion, paradox problem resolution, structured demo-

cratic dialog, among others) that rely on the group dy-

namic and coordinated work. Additionally, they con-

sist of articulated work for each process activity with

the outcome of particular activity (e.g. ”Write Needs”

and ”Write Offers”) used for later activities (e.g. ”1.

Match Offers ...”) as presented on the example of such

a workﬂow (ﬁg. 1).

Most of them use specially designed tools and

artifacts for facilitating and capturing information,

knowledge, and to foster various aspects of their

Extending BPM(N) to Support Face-to-Virtual (F2V) Process Modeling

353

Figure 1: Workﬂow process describing CoPI4P’s demo F2V workshop.

work, such as ideation, innovation and creativity. We

have also found that they very often ﬁnd digital tech-

nologies disruptive for the group dynamic and are po-

tentially reluctant to use them, mostly due to previ-

ous negative experience or conservative policies or

habits which, based on our survey ﬁndings, signiﬁ-

cantly changed during the COVID-19 period (Tesar,

2020), (Khalili, 2020). Providing declarative expe-

rience instead of more solid imperative IT solution

would increase the discomfort (the feeling of being

lost) of both facilitators and participants.

They use non-digital memoization artifacts whose

contents then becomes necessary for the further dig-

ital phases of the community’s business workﬂow.

Therefore, we have to capture and respond promptly

and bridge and transfer ecosystem values across the

real-world and virtual-world in face-to-virtual pro-

cesses in order to preserve the group dynamic, and

provide rapid feedback, satisfaction and ”ROI”

for

both participants and facilitators.

Apart from deeper discussions through individ-

ual interviews, co-design and focus groups interviews

with high-level facilitators, we have conducted a sur-

vey (provided in both English and Russian) interview-

ing international facilitator and educators (over 30 of

them, the countries represented include Serbia, Rus-

sia, Italy, India, Norway, USA, Germany, UK, Malta,

Portugal, Latvia, Croatia, Canada, and Belgium). The

survey included questions on the impact of COVID-

19 on their work and its relationship with IT technol-

ogy, workﬂow and BPM augmentation, as well as de-

mographic and professional data.

The ﬁgure 2 presents the ﬁrst set of our ﬁndings

that demonstrates, notably, that (a) facilitators’ work

overall did suffer due to lockdowns and isolation re-

lated to COVID-19. Some of them (b) successfully

transformed their working practices into the digital

realm and beneﬁted from IT tools. However, focus

groups interviews have shown that facilitators sacri-

ﬁced their pre-COVID working practices and key val-

ues for the sake of digitizing their work and commu-

nities, and lost control over and understanding of their

community dynamic. This probably explains why the

Return on investment.

majority among them would (c) beneﬁt from visual

workﬂows and a BPM approach, (d) ﬁnd beneﬁt in the

proposed BPMN extensions (decorations), and (e) are

interested in using ColaboFlow framework for aug-

menting their working practices with IT. The last two

diagrams of the ﬁgure 3 show that (b) the majority of

facilitators acknowledge IT technology as augmenta-

tive, and (c) the over 45

In the following table 1, we present CoPI4P com-

munities and the correlated F2V processes more sys-

tematically through the features recognized and ana-

lyzed here. We will especially focus on the features

that have impact on the business modeling of their

workﬂows.

6 NEW DESIGN PARADIGM

In this section, we argue for a new design and devel-

opment paradigm tailored for trans-domain ecosys-

tems. We approach it by considering the problem

of extending an IT system with intangible, ”soft”-

parameters of business processes (such as performers’

motivation, developmental transformations, social

and psychological aspects and satisfaction, among

others) into already designed and developed IT sys-

tem. We argue that with the proposed paradigm shift

such an extension

We will focus our argumentation for the proposed

paradigm shift on introducing the notion of motiva-

tion in systems implementation and standards. When

it comes to work motivation, existing studies address

motivation in an organizational context (Gebauer and

Fleisch, 2007), (Lepper and Greene, 2015), (Ilgen

and Pulakos, 1999), as well as the ”psychology of”

motivation within strictly business organisms - the

how and the why of business motivation mechan-

ics, notably the WHY in Zachman Framework, Busi-

ness Motivation Model (BMM) (an OMG’s standard)

and ArchiMate motivation extension (Berkem, 2008),

(Quartel et al., 2010), (Bhattacharya, 2017).

However, there is no clear work on (and moti-

vation, no matter how word-pun that sounds) under-

standing and encoding motivational aspects in either

BPM tools or standards. This is true on any as-

MODELSWARD 2021 - 9th International Conference on Model-Driven Engineering and Software Development

354

Table 1: The list of the features of CoPI4P communities and F2V (face-to-virtual) workshops relevant to the process modeling

and execution.

CoPI4P features

term description

ad-hoc consist of well-established core members and ad-hoc ever-changing participants that con-

tribute to the community core eventually

experiment-

ation

the modus-operandi is evolution through continuous experimentation and incremental im-

provement practices

group-

dynamic

their values strongly rely on the inner community dynamics, motivation, and energy

tech-

concerned

due to the group-dynamic they are very careful with introducing IT tech tools in their

workshops and events

F2V features

term description

group-

dynamic

the group-dynamic, number and structure of teams often change across the event span

frictions frictions along face vs. virtual dimensions, single vs, group, among others

evaluation real time evaluation of individual, teams and community’s contributions

facilitators multiple facilitators are responsible for each event, moderating both external outputs as

well as internal user transformations, groups development and changes in its dynamics

continuity facilitators and methodologies can switch with no breaks between them

process-

engagement

engagement with the event process might be at multiple levels: (i) descriptive and col-

laborative model used at the workﬂow design phase, (ii) a blueprint for conducting the

workﬂow, (iii) run the process evaluation, (iv) participants perform (some of) the process

activities through, among others

f2v-bridging the non-digital output of one activity is the input of a later activity. We need to ensure

rapid, crowd-sourced digitization, digitalization or interpretation of non-digital outputs

Figure 2: Facilitator survey ﬁndings; (a) work suffered (due to COVID-19), (b) beneﬁt of using IT tools, (c) beneﬁts of

presenting workshops through workﬂows, (d) beneﬁts of ColaboFlow Decorations, and (e) interest in using ColaboFlow.

Figure 3: Facilitator survey ﬁndings; (a) ColaboFlow decorators selections (AVG), (b) I see IT tech (regarding my processes)

as, (c) Status of a supporting IT-tool.

Extending BPM(N) to Support Face-to-Virtual (F2V) Process Modeling

355

pect of business model; for business performers, ac-

tivities or BPM process context. Clearly, such in-

formation, whether machine readable or not, would

bring business beneﬁts, especially when CoPI4P busi-

ness scenarios are considered. It is similar to other

”soft-”parameters of business processes, like per-

formers’ social and psychological aspects or satisfac-

tion, among others.

The approach of business modeling of a whole

ecosystem as an open sociotechnical system rather

than just a digital system brings additional challenges.

The whole ecosystem is partially encoded through

(i) the digital world (online platforms, visual rep-

resentations of workﬂows, mobile apps), partially

through (ii) the external physical world (books, note-

books, methodological cards and tools), and partially

through the (iii) social world of community members

(facilitators, moderators, observers, participants) and

tacit knowledge (methodologies, experience, emo-

tions, individual and collective intelligence). On the

one hand, the boundaries between the parts of the

ecosystem introduce sociotechnical gaps (Ackerman,

2000), while the non-digital parts of the ecosystem

make it hard to capture and digitalize the ecosystem’s

values. On the other hand, this (holistic) approach

and problems of digitalization makes it even harder to

use pre-designed business processes as a prescriptive

guide for community behavior outside of the digital

borders.

Additionally, the ecosystem values in such

CoPI4P communities can be hard to digitalize. Apart

from pure information and knowledge, they can com-

prise social content and social connections (social

capital) in a social network, as well as tacit (inter/intra

human) knowledge (Duguid, 2012), skills, collective

intelligence (Woolley et al., 2015), and human emo-

tions.

To support CoPI4P communities and the business

modeling that is emerging from their distinctive prac-

tice, we need a shift in the system design itself. The

motivation comes from the core difference in what we

are designing, as instead of designing systems, we de-

sign ecosystems. Ecosystems stress the importance of

the holistic and non-technical aspect of the solution.

We build on top of the sociotechnical systems (Whit-

worth et al., 2006) (Trist, 1981) and theory that under-

stands the importance of the duality in the design of

such ecosystems. Apart from the already mentioned

ecosystem worlds - digital, physical, and social - that

constitute such ecosystems, these ecosystems encom-

pass multiple ecosystem domains as well. There are

social, technical, democratic, geographical, psycho-

logical, artistic, racial, religious, scientiﬁc, and many

other domains or ﬁelds of inquiry. We are not propos-

ing to integrate (support for) the entire variety of these

domains as that strategy would be both unrealistic and

counter-effective. The aim is rather to avoid ”exclud-

ing” certain kinds of ecosystems from potential study

merely due to the limitations of the current design and

development principles and paradigms.

The problem is much larger than implementing

particular sets of standards or polices by the system

designer and developers. Rather, it concerns the chal-

lenge of the inherent incompetence of system design-

ers and domain experts to design a whole ecosystem.

It is valid for both the initial period of engagement

with the designed ecosystem, and even more when it

comes to an unpredictable future. If we take an exam-

ple of the democratic domain, we can see that current

social-media platforms lack mechanisms that would

monitor and protect platforms’ democratic values and

principles (Rudan and Rudan, 2014). Additionally,

it is hard to imagine today’s social media platforms

embracing such mechanisms, especially ones origi-

nating from a 3rd party independent institution such

as are usually the most competent for particular do-

mains. Therefore, we need to redesign the very way

we design and develop ecosystems.

We thus propose to expand ecosystem design

based on the sociotechnical and socio-material the-

ories that inherently recognize the limited set of

ecosystem aspects (i.e. sociotechnical or socio-

material ecosystems (enlisting) =¿ socio, technical,

material, ...) to apply a term that can potentially en-

compass an unlimited number of domains; therefore

we introduce a new term: trans-domain ecosystems.

This term recognizes ecosystems as liberal and ag-

nostic in their nature, but at the same time it calls

for a paradigm shift in ecosystem design and devel-

opment. Such a paradigm shift requires a design that

would embrace an open world rather than a closed

world, and provide mechanisms for introducing new

ecosystem domains. It calls for declarative descrip-

tion, evaluation, and evolution of an ecosystem, done

by all the key stakeholders of the ecosystem rather

than solely by developers; including the community

and experts in the given domain. It calls for a com-

mon language and a continuous ever-evolving bridge

between ecosystem designers, developers and CoPI4P

communities.

Without going into details, as it is out of the scope

of this paper, the candidate for such a language is

clearly BPMN, a language that can potentially be spo-

ken by all the key stakeholders of an ecosystem. The

additional, but often overseen

, values of ecosystem

This understanding comes from our preliminary explo-

ration of the state-of-the-art research, focus-group discus-

sions with system architects of CoPI4P communities sys-

MODELSWARD 2021 - 9th International Conference on Model-Driven Engineering and Software Development

356

design through BPM languages are ecosystem trans-

parency to its stakeholders and ecosystem external-

ization of its business logic that makes it possible to

extend and support ecosystem co-evolution through

its practice with the help of all its stakeholders.

The proposed paradigm shift and trans-domain

ecosystems to be implemented require (i) an exten-

sion of BPM(N) to support the additional needs of

the CoPI4P communities as presented in the section

7 and (ii) support for (among others) functionally-

equivalent sub-processes that we present in the sec-

tion 8.

Our R&D implementation and support for trans-

domain ecosystems is called Colabo.Space

. It is

an Open Source solution with a modular architec-

ture. It provides a knowledge repository for practic-

ing knowledge federation, and an extendable toolset

of collaborative methodologies for various aspects of

community collaboration such as knowledge feder-

ation, structured dialogue facilitation, creativity and

innovation, and mechanisms for bridging the bar-

rier between virtual and physical worlds. The men-

tioned components of the Colabo.Space ecosystem

are orchestrated through the ColaboFlow compo-

nent, a BPM component responsible for business pro-

cesses execution and BPMN visualization (integrat-

ing the open source and modular Camunda solution

for BPMN: https://bpmn.io/ and proprietary solutions

for performing BPMN processes by human partic-

ipants on the one side, and micro-services, on the

other).

7 EXTENDING BPMN

We propose a set of extensions of the BPMN standard,

based on the BPMN 2.0 native extension mechanism

, in order to support the discussed paradigm shift, im-

plementation of trans-domain ecosystems, and to in-

clude intangible, ”soft”-parameters of business pro-

cesses (sect. 6=. The notion of motivation is ad-

dressed by CI genome extension.

Based on the research explained in the section 4,

and aligned with the CoPI4P and F2V features identi-

ﬁed and listed in the table 1, the original set of ex-

tensions was provided: 1) through focus group in-

terviews with the top level facilitators (from various

types of CoPI4Ps), we have identiﬁed potentially nec-

essary extensions, 2) through surveys with a larger

group of facilitators, we have evaluated the impor-

tance of each of the extensions (ﬁgure 3 (a)) and 3)

tems, and evaluating open source Github projects.

Please check the ofﬁcial website of the system: https:

//colabo.space

through open questions in the survey, we have con-

ﬁrmed their universal applicability. The list of exten-

sions is presented in the table 2.

An excerpt from a demo process, covering repre-

sentative aspects of the studied processes and prac-

tices, as well as scenarios of the interviewed practi-

tioners/facilitators, is illustrated at the ﬁgure 4.

Several workshops conducted by authors and de-

scribed in the section 5 demonstrates digital disrup-

tion in the context of the communities that don’t prac-

tice digital augmentation of their face-to-face events.

As such, authors had the very sensitive task of ”dig-

itizing” their practice from face-to-face to face-to-

virtual, and ”patching” it at particular spots with digi-

tal augmentation. Having business modeling notation

such as BPMN, we could digitize the workﬂow. .

The digitizing procedure consists of (i) identify-

ing (together with domain and community experts)

known activities the given process should consist of

and describing them as best as we can, followed by

(ii) proposing necessary extensions for the process we

want to contribute to it, (iii) understanding what the

potential frictions may be, and (iv) evaluating digital

disruptions in the proposed workﬂow.

8 SUB-PROCESS PALETTES

In the BPMN standard, processes are presented on

one (conceptual) ”plane”, which is usually realized

with a pool of business roles (realized as set of lanes)

and the interaction between them. BPMN does sup-

port the encapsulation of process activities through

sub-processes, which enable the isolation of a set of

activities into a separate (conceptual) plane and the

possibility of reusing the part of business logic as a

sub-process. Modularity, reuse, and readability are

the main incentives for decomposing processes into

sub-processes.

The most critical issue with using regular BPMN

standard is the lack of ﬂexibility to choose, adjust

and personalize business processes from the perspec-

tive of performers. Very often rational is to go

with declarative (case based) BPM (Goedertier et al.,

2015) like OMG’s CMMN standard instead. Unfor-

tunately, CoPI4P communities ﬁnd it too confusing

and without capacity of quick (necessary at the real-

time events) capturing the ”ﬂow” of the F2V process

with declarative BPM approach. Instead the prefer-

able scenario, that we experienced (participatory de-

sign and participatory action research approach) prac-

ticing with CoPI4P, is to provide the process conﬁ-

dence with imperative structure that BPMN provides

but additionally loosened to give freedom and ﬂe-

Extending BPM(N) to Support Face-to-Virtual (F2V) Process Modeling

357

xibility to describe behavioral and personal varieties.

This is our motivation to propose sub-process

palettes, or meaningful chunks of BPMN processes.

A sub-process palette is a collection of similar, func-

tionally semi-equivalent, sub-processes. The sub-

processes in a sub-process palette are functionally

semi-equivalent in the sense that the differences in

their performance outcomes are negligible from the

perspective of domain experts or the community per-

forming them, and as such they are perceived as

equivalent and used in an interchangeable manner.

The beneﬁts coming with the proposed notion

of the sub-process palette are: (i) non-rigid (re-

duced) imperativeness, (ii) personalization (iii) evolv-

ing sub-process palettes into design patterns recog-

nized as best practices in process design, (iv) vi-

sually communicating top-level functional represen-

tation of the whole sub-process palette, (v) provid-

ing a container (mostly for the sake of human com-

prehension) for the community participants’ activi-

ties as they are process-mined into functionally semi-

equivalent sub-processes, a process that is crucial for

capturing tacit knowledge (HOW) and not just the re-

sult of an activity (WHAT), (vi) the ability to interact

with and conﬁgure, in an aggregated (clustered) man-

ner, the exposed top-level aspects of the underlying

sub-processes.

The ﬁgure 5 presents an example of the sub-

process palette Proposing a CONTEST consisting of

3 sub-processes. It is an example of knowledge dis-

semination processes in the ISSS community ((Rudan

et al., 2015)).

All three diagrams in the ﬁgure are functionally

semi-equivalent from the perspective of performing

the activity of proposing a contest. The diagram 5 (a)

is an initial diagram, which assumes that the contest

appears healthily now and then, whenever it makes

sense to oppose an inappropriate INSIGHT.

9 CONCLUSION AND FUTURE

WORK

In this paper, we have presented a paradigm shift

in system design toward trans-domain ecosystems

that better match CoPI4P (Community of Practice

& Interest for Purpose) communities and F2V (face-

to-virtual) processes. We have, further, proposed

to extend BPMN with a new concept, sub-process

palettes, and domain-speciﬁc support for F2V pro-

cesses. Working closely with a variety of COPI4P

communities, we have gained conﬁdence in the con-

ceptual and practical beneﬁt of the proposed solu-

tion. With exemplar F2V processes, the wider group

of facilitators showed no identiﬁable cognitive over-

load. However, the effect on more complex work-

ﬂows communicated across various stakeholders of a

community (namely, regular participants) remains to

be understood. Given ColaboFlow’s support for dif-

ferent views (perspectives) on visual workﬂows, we

expect to address any of the potential problems. There

are still open questions, with a positive note, when it

comes to the framework as a whole and its capability

to share common community practices, attract com-

munity members and engage them with the core com-

munity practices.

With the support of sub-process palettes, we

are capable of ecosystem augmentation through a

process-mining approach. It opens various possibil-

ities, like detecting a community’s best practices and

behavioral patterns, as well as helps with monitoring

and recognizing anti-patterns in regard of a commu-

nity’s social health or ecosystem values production.

Apart from doing initial ABM (Agent Based Model-

ing) and evaluating the ratio between the community

or ecosystems gains vs. the level of loss of user free-

dom (through following system suggestions for activ-

ities), this is still an open space for future research.

Another key question and potentially challenge con-

cerns fusing community crafted processes with the

potentially vast number of sub-processes mined by

the system.

We are also interested in the aspects of commu-

nity engagement with visual workﬂows and process-

driven IT technology overall. We have conducted an

initial survey on that topic and recorded considerable

interest in the whole spectrum of possible aspects, but

as space is limited, this must be left for future research

and reports.

Finally, we want to further understand the long-

term impact of COVID-19 on the transformation of

the CoPI4P practices and their toolsets. We are

very interested in the potential friction in the techni-

cal integration of the proposed design paradigm by

(eco)system architects, although the initial interviews

with this focus group indicated a positive interest,

mainly due to the possibility of semi-transparent sys-

tem integration and expected overall beneﬁts of such

a paradigm shift.

ACKNOWLEDGEMENT

We want to thank to ChaOS NGO and University of

Oslo for providing the support for conducting the re-

search, to Dino Karabeg (University of Oslo), Eu-

genia Kelbert (UEA, UK and HSE, Moscow), Uri

Noy Meir (ImaginAction and Bahir Consultancy),

MODELSWARD 2021 - 9th International Conference on Model-Driven Engineering and Software Development

358

Table 2: Elements of the BPMN extension to support CoPI4P and face-to-virtual processes.

COPI4P and F2V (Face-to-virtual) BPMN Extension

Representation Description

collective manual task - An activity with this decoration is performed manually

(without IT support) and collectively (by the community/team/group). Supported

features (table 1): group-dynamic.

collective task - An activity with this decoration is collectively performed (by the

community/team/group) in a virtual world (by digital tools). Supported features:

group-dynamic.

A value with this decoration is an evaluation of related activity by participants

(about content, satisfaction) using different mechanisms (voting, comments, scales).

It provides facilitators with in-vivo insights on the group dynamics and enables

decision-making on which compensation actions to take. Supported features: fric-

tion, process-engagement, group-dynamic, and evaluation.

CI genome - Decorates an activity with CI (COLLECTIVE INTELLIGENCE)

aspects. This way, facilitators can design crowd-sourced processes (Malone et al.,

2010), addressing participants’ motivation for activity. Supported features: facilita-

tors, process-engagement.

individual-transformation - e.g. embodiments, attitudes changes, that are expected

to happen as an activity outcome. Supported features: group-dynamic, evaluation,

and facilitators features, and developmental and educational workshops, taking in

mind internal, experiential, and intangible ones.

collective-transformation - (e.g. community building, a team’s shared vision) ex-

pected to happen as an activity outcome. It focuses on intangible outputs and sup-

ports group evolution and collective intelligence development. Supported features:

group-dynamic, evaluation, and facilitators.

individual-skill - skills (e.g. visualisation, negotiation) that a participant acquires by

performing the activity. Supported features: group-dynamic, evaluation, and facili-

tators.

collective-skill - (e.g brainstorming) that community/team/group acquire by per-

forming the activity. Supported features: group-dynamic, evaluation, and facilita-

tors.

AI-assistant - the activity is augmented with AI/IT-ASSISTANCE provided by Co-

laboFlow (e.g. team creation, ideas or interests matching). Supported features: f2v-

bridging, evaluation, facilitators. It beneﬁts to the most of other features.

Extending BPM(N) to Support Face-to-Virtual (F2V) Process Modeling

359

Figure 4: Business process behind a demo face-to-virtual workshop.

Figure 5: ”Proposing the CONTEST” sub-process palette - (a) 0th approximation, (b) 1st approximation (alternative 0), (c)

2nd approximation (alternative 1).

Maria Rodinova and Mikhail Kozarinov (Soling),

Oleg Muromtsev and Alyona Surikova (Esher Para-

dox school), Irina Antonova and Tomour Shchoukine

(NakedMinds Lab), Dimitrije Bukvic (independent

researcher), Lazar Kovacevic (Inverudio), Jack Park

(TopicQuests Foundation), Matthew Reynolds (Uni-

versity of Oxford), George Fazekas (Queen Mary

University of London), Ivanka Radmanovic (Associ-

ation of Writers of Serbia), Elvio Ceci from CeAS

- Lupt, Mick Mengucci from Lab.I.O., Dr Jasmina

Madzgalj and Vesna Sabanovic (City of Belgrade,

Secretariat for Environmental Protection), MSc Sin-

isa Mitrovic (Chamber of Commerce of Serbia),

Marc-Antoine Parent (Conversence), Dragana Tri-

funovic (DigitalNorway), Myrto-Helena Pertsinidi

(Jugend- & Kulturprojekt e.V.) and all other facilita-

tors and communities involved in our research.

REFERENCES

Ackerman, M. S. (2000). The intellectual challenge of

cscw: the gap between social requirements and tech-

nical feasibility. Human–Computer Interaction, 15(2-

3):179–203.

Berkem, B. (2008). From the business motivation model

(bmm) to service oriented architecture (soa). J. Object

Technol., 7(8):57–70.

Bhattacharya, P. (2017). Modelling strategic alignment of

business and it through enterprise architecture: Aug-

menting archimate with bmm. Procedia computer sci-

ence, 121:80–88.

Bocciarelli, P. and D’Ambrogio, A. (2011). A bpmn exten-

sion for modeling non functional properties of busi-

ness processes. In Proceedings of the 2011 Sympo-

sium on Theory of Modeling & Simulation: DEVS In-

tegrative M&S Symposium, pages 160–168. Society

for Computer Simulation International.

Bonakdarian, E., Whittaker, T., and Bell, D. (2009). Merg-

ing worlds: when virtual meets physical: an experi-

ment with hybrid learning. Journal of Computing Sci-

ences in Colleges, 25(1):61–67.

Bork, D., Karagiannis, D., and Pittl, B. (2018). Systematic

analysis and evaluation of visual conceptual modeling

language notations. In 2018 12th International Con-

ference on Research Challenges in Information Sci-

ence (RCIS), pages 1–11. IEEE.

Bork, D., Karagiannis, D., and Pittl, B. (2020). A survey of

modeling language speciﬁcation techniques. Informa-

tion Systems, 87:101425.

Brambilla, M., Fraternali, P., and Vaca, C. (2011). Bpmn

and design patterns for engineering social bpm solu-

tions. In International Conference on Business Pro-

cess Management, pages 219–230. Springer.

Bratteteig, T. and Wagner, I. (2014). Disentangling partic-

ipation: power and decision-making in participatory

design. Springer.

Chatzinotas, G. (2017). Community management. PhD the-

sis, University of Thessaly.

Chevalier, J. M. and Buckles, D. J. (2019). Participatory

MODELSWARD 2021 - 9th International Conference on Model-Driven Engineering and Software Development

360

action research: Theory and methods for engaged in-

quiry. Routledge.

Duguid, P. (2012). ‘the art of knowing’: social and tacit

dimensions of knowledge and the limits of the com-

munity of practice. In The Knowledge Economy and

Lifelong Learning, pages 147–162. Brill Sense.

Gebauer, H. and Fleisch, E. (2007). An investigation of

the relationship between behavioral processes, moti-

vation, investments in the service business and ser-

vice revenue. Industrial Marketing Management,

36(3):337–348.

Goedertier, S., Vanthienen, J., and Caron, F. (2015). Declar-

ative business process modelling: principles and mod-

elling languages. Enterprise Information Systems,

9(2):161–185.

Ilgen, D. R. and Pulakos, E. D. (1999). The Changing Na-

ture of Performance: Implications for Stafﬁng, Moti-

vation, and Development. Frontiers of Industrial and

Organizational Psychology. ERIC.

Indulska, M., Green, P., Recker, J., and Rosemann, M.

(2009). Business process modeling: Perceived bene-

ﬁts. In International Conference on Conceptual Mod-

eling, pages 458–471. Springer.

Kemsley, S. (2011). Leveraging social bpm for enterprise

transformation. Social BPM Work, Planning and So-

cial Collaboration Under the Impact of Social Tech-

nology. BPM and Workﬂow Handbook Series, pages

77–83.

Khalili, H. (2020). Online interprofessional education dur-

ing and post the covid-19 pandemic: a commentary.

Journal of Interprofessional Care, 34(5):687–690.

Koch, N. and Kraus, A. (2002). The expressive power of

uml-based web engineering. In Second International

Workshop on Web-oriented Software Technology (IW-

WOST02), volume 16. CYTED.

Lepper, M. R. and Greene, D. (2015). The hidden costs of

reward: New perspectives on the psychology of human

motivation. Psychology Press.

Loo, S. (2016). Creative Working in the Knowledge Econ-

omy, volume 3. Taylor & Francis.

Malone, T. W., Laubacher, R., and Dellarocas, C. (2010).

The collective intelligence genome. MIT Sloan Man-

agement Review, 51(3):21.

Pankowska, M. (2019). Business models in cmmn, dmn

and archimate language. Procedia Computer Science,

164:11–18.

Proper, H. A., Bjekovi

c, M., van Gils, B., and de Kinderen,

S. (2018). Enterprise architecture modelling: Purpose,

requirements and language. In 2018 IEEE 22nd In-

ternational Enterprise Distributed Object Computing

Workshop (EDOCW), pages 162–169. IEEE.

Quartel, D., Engelsman, W., and Jonkers, H. (2010). Archi-

mate extension for modeling and managing motiva-

tion, principles and requirements in togaf. Reading,

Berkshire: Whitepaper, The Open Group.

Recker, J. C. (2008). Bpmn modeling–who, where, how and

why. BPTrends, 5(3):1–8.

Rudan, S. M. and Rudan, S. (2014). Democracy frame-

work politics & leadership in online communities. In

2014 First International Conference on eDemocracy

& eGovernment (ICEDEG), pages 67–72. IEEE.

Rudan, S. M., Rudan, S., and Karabeg, D. (2015). Repro-

gramming anthropocene-crowdsourced governance of

trans-technical systems. In Proceedings of the 59th

Annual Meeting of the ISSS-2015 Berlin, Germany,

volume 1.

Simonsen, J. and Robertson, T. (2012). Routledge interna-

tional handbook of participatory design. Routledge.

Swenson, K. (2010). The quantum organization: How so-

cial technology will displace the newtonian view. So-

cial BPM, pages 19–34.

Tesar, M. (2020). Towards a post-covid-19 ‘new normal-

ity?’: Physical and social distancing, the move to on-

line and higher education.

Trist, E. (1981). The evolution of socio-technical systems.

Occasional paper, 2(1981):1981.

Whitworth, B., Ahmad, A., Soegaard, M., and Dam, R.

(2006). Encyclopedia of human computer interaction.

von C. Ghaoui. Hershey: Idea Group Reference. Kap.

Socio-technical systems, pages 533–541.

Woolley, A. W., Aggarwal, I., and Malone, T. W. (2015).

Collective intelligence and group performance. Cur-

rent Directions in Psychological Science, 24(6):420–

424.

Extending BPM(N) to Support Face-to-Virtual (F2V) Process Modeling

361