KM and KA in International Coooperation

Lesson from the K-Link project in Central Asia

Gianluca Colombo

, Alessio Vertemati

, Emanuele Panzeri

, Eva Grolíková

and Philipp Reichmut

Siris Academic S.L., Av. Francesc Cambó 17, 9th floor B, 08003 Barcelona, Spain

Palacký University Olomouc, Křížkovského 8 771 47 Olomouc, Czech Republic

Keywords: Knowledge Management, Central Asia, International Cooperation, Organizational Development, Search

Engine, Knowledge Artifact.

Abstract: The K-Link initiative concerns a Knowledge Management (KM) solution for Central Asian institutions com-

mitted to Sustainable Land Management (SLM) and Natural Resource Management (NRM) projects imple-

mentation in the framework of international development cooperation. It supplies a technical solution based

on the use of a distributed network in which all data and information remain stored on each institution’s server

and can be accessed from each institution’s website without the need to create a central hub. One of the main

issue regarding the sharing of documents in the Central Asian context actually regards motivations for sharing.

In fact the trigger for information sharing in Sustainable Land Management in Central Asia is given by the

management of information. This principle verified and tested with the participant institutions led to the res-

olution to design and implement a specific Document Management System (K-DMS) in order to allow end

users to get to know the sharing features of K-Link as special cases of more general information management

ones. The K-DMS represents computer-based environment for the production and sharing of knowledge arti-

facts.

1 INTRODUCTION

The K-Link initiative concerns a Knowledge Man-

agement (KM) solution for Central Asian institutions

committed to Sustainable Land Management (SLM)

and Natural Resource Management (NRM) projects

implementation in the framework of international de-

velopment. It has been financed by GIZ GmbH and

implemented by SIRIS Academic S.L; it started on

December 2013 and ended up with a running pilot on

June 2015. Before going through the K-Link project

in details and showing how it is promoting knowledge

artifacts production, a survey on both general and spe-

cific background conditions which motivate KM ini-

tiative in Central Asia in the framework of interna-

tional cooperation need to be taken into account.

All five Central Asian states along with Azerbai-

jan were one of the poorest in the former Soviet Un-

ion. And when they became unexpectedly independ-

ent in 1991, they faced three large negative shocks:

the end of central planning, the dissolution of the So-

viet Union system, and hyperinflation. At first the

shock in all Central Asian countries caused the fall of

output, and increased their poverty and inequality but

each country faced it differently. The newly inde-

pendent states emerged from the USSR with similar

economic systems and some similarity in economic

structure (Brill Olcott, 1998). However, by the

twenty-first century their economies differentiated

but not their political systems, for all of them had es-

tablished super-presidential systems with concen-

trated power and weak legislatures (Pomfret, 2006).

The level of economic growth in Central Asian coun-

tries also proportionally reflects the number of Inter-

net users (IWS, 2014). The Global Information Tech-

nology Report (Dutta, Geiger and Lanvin, 2015)

ranks Tajikistan 62

, Kazakhstan 69

and Kyrgyz-

stan 111

out of 143 countries in terms the capacity

for innovation (to what extent do companies/organi-

zations have the capacity to innovate). And the same

report ranks them 90

, 74

and 119

respectively in

terms of impact of ICTs on new services and products

(to what extent does ICT enable new business mod-

els). Unfortunately data for Uzbekistan and Turkmen-

istan were not included in the report. Based on this

information it could be stated that (1) although the In-

ternet penetration rate is not in all states as high as it

could be, it is growing steadily, (2) in all states there

is better capacity for innovation as well as for impact

Colombo, G., Vertemati, A., Panzeri, E., Grolíková, E. and Reichmut, P..

KM and KA in International Coooperation - Lesson from the K-Link project in Central Asia.

In Proceedings of the 7th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (IC3K 2015) - Volume 3: KMIS, pages 421-428

ISBN: 978-989-758-158-8

421

of ICTs on new services and products than in one

fourth of other countries in the world. Hence this sit-

uation can be used as a basis for usage of new tech-

nologies (ICTs etc.) for capacity building in Central

Asia. According to Rondelli (2014) all actors (i.e. re-

searchers, NGOs and institutions) in Central Asia

have difficulties accessing knowledge, for even after

years of study part of the existing data still escapes

their attention. And (reflected) knowledge (e.g. best

practices) itself contains the ability to change the

learning (to develop capacities) of individuals, groups

or society as a whole. Therefore a Knowledge Man-

agement approach to academic-practitioner and prac-

titioner collaboration is one of the most effective

ways in order to bridge the research-practice and

practice-practice gap that is hindering the develop-

ment of efficient problem-solving strategies.

These are the general motivations in which a KM

initiative in Central Asian is necessarily framed. Re-

garding the specific motivations, this leaded the K-

Link project implementation, some additional re-

marks on KM needs in Sustainable Land Manage-

ment (SLM) need to also be considered.

Especially as Kyrgyzstan is one of Asia’s most

sparsely wooded countries, with a forest cover of only

4-5%. Despite that fact, the world’s largest connected

natural fruit and walnut forests are to be found in the

south of Kyrgyzstan (Jalalabad and Osh Region).

They are considered the origin of many domesticated

fruit and nut varieties; the in-situ conservation of

these wild varieties is of global importance. Those

forests are located at an altitude of 1,100 – 1,800

m.a.s.l. and are inseparably connected to the adjacent

biodiversity rich highland pastures (1,800 – 3,400

m.a.s.l.), in terms of their ecology and usage. At the

same time, the forests form a significant basis for ru-

ral livelihoods and have important regulating func-

tions for the downstream regions, e.g. for the water

regime. The soils are important carbon sinks, thanks

to their thick humus layers.

Degradation processes, induced by unsustainable

resource use (firewood extraction, one-dimensional

concentration of the management on nut harvesting,

wood pasture), result already in a significant reduc-

tion of the water retention capacity of the forests. The

forest stand has changed in its structure and has been

reduced in its diversity. The harvests have decreased

considerably harming the natural resilience of the for-

ests. The negative impacts on the forests as well as

the surrounding ecosystems are additionally exacer-

bated by the already perceived adverse impacts of cli-

mate change. The pressure on the forest resources in-

creases steadily due to high levels of population

growth. Uncontrolled settlement development leads

to a fragmentation of the forests (Borchardt et al.,

2011; Swetnam et al., 1999).

Acknowledging the unique nature of the forests, a

high variety of research on biodiversity and the breed-

ing of new varieties of nuts and fruits was conducted

already in Soviet times. After the independence of

Kyrgyzstan namely Kyrgyz, German and Swiss re-

search institutes did intensive research on the region,

covering ecologic as well as socio-economic and

management aspects.

Simultaneously, a multitude of steps to align the

Kyrgyz forest sector towards multi-functional for-

estry and to include the local population in the man-

agement were undertaken within the scope of the

Swiss KIRFOR Programme from 1995 to 2009. The

walnut forests in the south of Kyrgyzstan were a pilot

area for the implementation of participatory forest

management approaches. This project has undoubted

accomplishments, however, fundamental reasons for

the degradation of the forests prevail and new have

occurred respectively. A very important reason for

the degradation of the walnut forests is the unsolved

issue of overgrazing of near to forest pastures and the

unfinished reform of the pasture sector in Kyrgyzstan.

NGOs, State Agencies and Academics with the sup-

port of the GIZ Regional Programme on Sustainable

Use of Natural Resources in Central Asia are cur-

rently testing appropriate approaches for a better

management of pastures in South Kyrgyzstan.

The paper is organized as follows. In section 2

Knowledge Management issues are framed with re-

spect to the central Asian international cooperation

needs. In section 3 the K-Link project is presented

and discussed. In section 4 the K-Link is discussed

taking into account the Knowledge Artifact design

principles. In section 5 Conclusions and future direc-

tions are presented.

2 KM NEEDS FOR CENTRAL

ASIAN SLM

Knowledge management platforms in its many forms

can be used both to improve the way organizations

communicate and to disseminate knowledge inter-

nally as well as externally. Generally, it could be said

that even after many years of KM use most of the or-

ganizations are not yet taking maximum advantage of

what new knowledge management systems and tech-

nologies have to offer (UNICEF, 2008). To do so the

organizations should focus on sustainability, it means

to decide on which areas their organizations need to

KITA 2015 - 1st International Workshop on the design, development and use of Knowledge IT Artifacts in professional communities and

aggregations

422

focus on, and how much effort, time and financial re-

sources they are willing to invest.

The field of international cooperation in Central Asia

is not so different from other regions it is just specific.

Its specific functioning and all involved actors are de-

scribed in detail bellow.

International Implementing Agencies. like GIZ is -

are normally requested by donors - in the case of GIZ

the German Federal Ministry of Economic Coopera-

tion and Development [BMZ] - to elaborate project

proposals on biodiversity preservation and poverty al-

leviation in Central Asia. Over the last decade large

amount of projects have been focused on the walnut

forests of Southern Kyrgyzstan.

In order to fulfil the request of Donors, implanting

Agencies usually send a mission to the rural areas

(e.g., walnut areas).

For their respective preparation as well as for compil-

ing the project proposal according to the donors for-

mat and standards the implanting agency mission nor-

mally uses manifold different documents (Donors

policy papers, implanting agencies country strategy,

scientific papers on the project area, documentation

and best practices of other projects in the region, etc.)

The NGOs involved into the project proposal and fu-

ture implementation are already approached during

the project finding mission by the implementing

agencies in order to develop a proposal for a training

concept for local land users to enhance the pasture

management in and around the area of interest (e.g.,

the walnut forests).

The selected NGOs have a long standing experience

in developing trainings in the field of pasture manage-

ment and sustainable land management in general. In

order to develop an approach NGOs usually retrieve

and study different kinds of documents.

Academics. e.g., University of Central Asia, which is

well connected to the international Agroforestry Cen-

ter (ICRAF) - are also involved into the process.

Many valuable data and publications originating from

Soviet times are actually stored in the institutions li-

brary. Many Universities in Central Asian have in-

deed established a wide national and international re-

search cooperation network.

Based on that wealth of knowledge and contacts, re-

searchers of the university in cooperation with re-

search fellows from the EU Universities normally

prepare a research project on monitoring the dynam-

ics of biodiversity and condition of the target rural ar-

eas. An analytical study and baseline of the currently

existing conditions is thus compared with historic

data and later with monitoring data at the mid and end

of the planned donors financed development project.

Normally also private companies take part into the

process. The companies are looking for two things:

firstly for new markets and export schemes (tax and

tariff arrangements; standards; laws; experiences of

nut producers in other countries). Secondly they are

also committed to find out innovative schemes to con-

nect them with huge importing companies on the in-

ternational markets.

The State Agencies have of course a huge interest in

an improvement of the current situation in SLM (e.g.,

walnut areas). This is due to the importance of the for-

ests and water resources for the global efforts to pre-

serve biodiversity. Accordingly to it, the State

Agency highly welcomes the initiative of Donors to

launch projects in that area. However, State Agencies,

which are the political implementation partner of the

planned projects, are aware of many other projects,

which have been implemented or are under imple-

mentation there. In order not to double and to coordi-

nate the new project with the others, the State Agen-

cies are trying to compile an overview of past and pre-

sent development projects for any specific area of in-

terest (K-Link, 2015).

3 THE K-LINK PROJECT

According the above reported surrounding conditions

and stakeholder needs, in the context of the K-Link

(2015) initiative by Knowledge Management is

meant the adoption of technologies to the support of

reliable resources detection, organization and sharing

among stakeholders working in SLM and NRM in the

specific area on international cooperation in post-so-

viet countries. By “reliable resources” is here meant

documentation regarding statistics, best practices,

valuable past experiences of projects implementation

(both positive and negative), scientific reports and de-

liverables. All these information sources mainly re-

gards projects proposal building, implementation,

monitoring and results that normally are spread into

unstructured documents owned by single institutions,

i.e. State Agencies, NGOs, Implementing Agencies

and Academics all over Central Asia. The main re-

quirement K-Link was expected to cover was, in other

words, the stakeholders’ need to speed up the process

of information access (Baeza-Yates and Ribeiro-

Neto, 1999) and to reduce the information entropy

provided by popular Web search engines. According

to the requirements analysis carried out with the client

and the stakeholders, three main constraints should

have been also be taken into account:

KM and KA in International Coooperation - Lesson from the K-Link project in Central Asia

423

- to enable sharing while respecting ownership of

data and information,

- to support information management without im-

posing a single model for organizing the re-

sources (documents and data) and

- without forcing end users to share.

The K-Link supplies a technical solution to this prob-

lem, based on the use of a distributed network in

which all data and information remain stored on each

institution’s server and can be accessed from each in-

stitution’s website without the need to create a central

hub. Plugged into pre-existing content management

platforms that run at the end-users institutions side,

K-Link extends those systems with information re-

trieval, management and sharing features. On the one

hand by applying state of the art content-based search

engine to the parsing of local documents repositories,

the K-Link offers a palette of services aimed at sup-

porting users in information discovering and access-

ing. On the other hand the peer-to-peer software ar-

chitecture enables end-users to share information

with the networked institutions (or members of the in-

stitutions) without forcing them to use a central hub

while respecting the freedom to decide what to share,

with whom and when. Six institutions have been in-

volved into the K-Link Pilot, namely one academic

placed in Bishkek-Kyrgyzstan – the University of

Central Asia (UCA), two NGOs one from Kyrgyzstan

and the other one from Kazakhstan – Camp Alatoo

and Ecomusem, the State Agency of Kyrgyzstan on

Forestry Management, an Implementing Agency

ICARDA and one private company from Kazakhstan

that is CAREC.

3.1 The Peer-to-Peer Architecture

The K-Link architecture has been engineered to be

modular and extendible to better fit the changing re-

quirements and the different environments where the

system would be integrated, while enforcing its dis-

tributed aspects.

The K-Link system is composed by two main

components, namely the Core and the Adapters com-

ponents: the K-Link Core (or in short KCore) compo-

nent represents the heart of the distributed network

and it is in charge of the document analysis and the

text search features. The Adapters components are

specific pieces of software that allow third-party Con-

tent Management Systems to be integrated with the

K-Link system.

The KCore can be identified as a node in the dis-

tributed K-Link network and represents the entry

point for the document textual analysis and the full-

text search features; most of such functionalities are

exposed via dedicated Application Programming In-

terface (API) by using the RESTful paradigm. A

KCore is responsible of the following aspects:

• Multi-format document indexing;

• Text and Geographical analysis and extraction;

• Document metadata extraction;

• Full-Text search (with document filtering and

grouping capabilities);

• Secure API invocation with authentication and

role-based permissions;

• Distributed network management, or-chestra-

tion, scalability and information sharing;

• K-Link network status and components health-

check;

Furthermore, the KCore has been designed to be able

to support both the local institution “private” and the

distributed (or “public”) document search. Particular

attention has been used to secure a complete separa-

tion of the two ‘private’ and ‘public’ collections: the

two set of document are, in the KCore, represented by

two different search engines where the functionalities

related to the ‘private’ documents collection is not ac-

cessible from the distributed network, this to avoid

any private document data leaking.

The K-Link Adapters are specific pieces of soft-

ware that allows a Content Management Systems

(CMS), or other pre-existing document management

software owned by the participant institutions, to ac-

cess and interact with the K-Link services, by using

the exposed KCore’s APIs. An adapter is specific to

a particular CMS: the K-Link project provides the

adapters for the most common Open Source CMSes

(Drupal, Wordpress, ModX). Additional adapters can

be subsequently implemented by following the freely

available KCore APIs specification, and the source

code of the currently available adapters.

3.2 From Information Retrieval to

Management

One of the main issues regarding the sharing of doc-

uments in the Central Asian context actually regards

the following motivations: why should someone in-

vest time for uploading and documents into an infor-

mation sharing system? What is the benefit of shar-

ing? After one year spent in discussing and negotiat-

ing with Central Asian stakeholders on these topics,

we realized that the trigger for information sharing in

Sustainable Land Management in Central Asia

(SLM) is given by the management of information.

In other words, the sharing of information is a

consequence of the management of information, not

vice versa. This turns the question of information

KITA 2015 - 1st International Workshop on the design, development and use of Knowledge IT Artifacts in professional communities and

aggregations

424

sharing into a problem of the organizational develop-

ment of the participating organizations, their internal

information workflows and knowledge management

processes. It is a consequence of the fact that the in-

formation to be shared is about something that is rel-

evant to the organization’s own practices, mission

and purpose, and that this purpose for most organiza-

tions includes generating, processing and distributing

knowledge as part of their mission.

Depending on the nature and degree of profes-

sionalism of the organization, the link between its

mission and its intrinsic motivation to engage in in-

formation sharing as part of its internal information

management strategies can take different forms. It is

important to see this as a question of organizational

development, as K-Link was commissioned by a de-

velopment organization (GIZ) that engages in sup-

porting the development of third-world countries. In

this setting, the organizational development and over-

all professionalism of NGOs as well as Government

institutions tend to be suboptimal, and there exist lit-

tle if any professional knowledge management ap-

proaches.

Introducing K-Link and providing capacity build-

ing for it therefore has direct impacts on the

knowledge management approaches in these organi-

zations. Most of the participating organizations are

supposed to engage in knowledge processes as part of

their mission; e.g. NGOs should engage in research,

participate in professional discourse with their peers

and/or inform the public, and government institutions

should be transparent and accountable and provide

clear policy information pertinent to its thematic sec-

tor.

The experience with introducing K-Link has

shown that the introduction of KM instruments even

in developing country settings has significant impacts

on how organizations approach questions of KM as

part of their general management and operating strat-

egy. Through embedding the management and shar-

ing of information into the everyday workflows of the

organization, the organization can do a significant

step in professionalizing its own practices. In the con-

text of K-Link this systemic change is now observa-

ble in several of the participating organizations.

The principle of sharing as a consequence of in-

formation management – verified and tested with

both the client and the participant institutions – led to

the resolution to design and implement a specific

Document Management System (K-DMS) so to allow

end users to get to know the sharing features of K-

Link as special cases of more general information

management ones (Kwasnik, 1991)..

The analysis of the basic features of the DMS (and

their requirement analysis) has been conducted by

taking into account a set of use cases discussed with

the NGOs (which represent the key providers of com-

petences for any project implementation in SLM do-

main) and then extended to all the other stakeholders.

The requirements analysis lead to the identifica-

tion of two main requirements:

− Req1: The need of exploring the available docu-

mentation on current projects so to gather the in-

formation needed for writing the NGO history

and reporting the on-going projects:

− Req2: Documents are potentially uploaded and

edited by multiple persons or working groups

and a version tracking is somehow required to

handle the process of document creation.

An important remark on the nature of the working

groups is reported here (quotes from the technical

documentation produced during the requirements

analysis and knowledge acquisition campaign carried

out with stakeholders.): “In case the project is trans-

disciplinary (contribution from two or more working

groups in a more or less balanced ratio – usually

there is one working group in the lead), or in case the

project manager is not a coordinator of a working

group, the project file is stored on the server on the

working group level”. This circumstance brought to

an additional requirement:

− Req3: Some documents and/or entire projects

need a way to be shared between working groups

or need to be categorized in multiple working

groups.

Furthermore, “[...] What is really relevant here is that

the technical proposal is normally realized as a re-

finement of the preliminary baseline that is on its turn

arranged by taking into account previous base lines

documentation and technical proposal regarding

similar projects. [...]“. This process points out that:

− Req4: the management must be able to access an

organized set of material and to search into them

to better find the documents they need.

Many NGOs in Kyrgyzstan and Kazakhstan use a

central server that can be accessed through the insti-

tution internal network to store documents. On the

server they have started to organize the reports on the

mitigation of conflicts among farmers. The next pas-

sage is extracted from the use case document: [...]

“They are now trying to migrate the reports into the

server, but reports are in Kyrgyz and at normally only

a subset of reports are translated into Russian and/or

English. Once the translation has been done, they are

KM and KA in International Coooperation - Lesson from the K-Link project in Central Asia

425

uploaded into the server. Mediators decide what re-

port is to be translated according to its relevancy. But

moderators often do not have access to the server be-

cause most are from outside (e.g. university). In ad-

dition these reports are stored on the personal lap-

top“.

Thus, the project implementation normally involves

external experts as key-contributors of the reports,

which don’t have the access to server.

− Req5: The main service here is to provide a con-

trolled access to the documents repository and

management, hopefully limited to specific shared

folders, so to allow mediators and moderators to

access the documents and the report writers to

upload documents and share or notify its availa-

bility to the other persons involved.

According to the above mentioned analysis the fol-

lowing main services have been decided for the K-

Link DMS:

1. Controlled Document Upload and Viewing:

the document upload, editing and management

experience must be controlled by a form of au-

thentication.

2. Collections of Documents: the way to overcome

the limitation of folders based approach for or-

ganizing the documents. Collections can be la-

belled (eventually importing a taxonomy) so to

enrich the documents descriptors and enhance

the documents retrieval. By using the label con-

cept it is possible to categorize documents in

multiple collections. Two main types of collec-

tion have been designed:

a. Personal collections: used by the user to fasten

the access of the documents;

b. Institutional collections: used to represents the

current working groups’ organization.

3. Groups: Users can be grouped into sets so to fa-

cilitate the share of documents to the stakeholders.

4. Share Documents and Collections: the mecha-

nism through which users are enabled to share

one single document (or a collection of docu-

ments) with other users of the DMS. Document

owners can further define which documents can

be accessed by other DMS users participating

into specific projects.

5. Document sharing is intrinsically performed in-

side the involved institutions by using email at-

tachments, with this feature we would like to re-

duce the need of the email communication and

the confusion created by the multiple versions of

that email communications.

6. Document Version Tracking: enables to upload

new versions of the same document.

7. Document Naming Policy check: the feature co-

vers the need of forcing all the produced docu-

ments (or files) to respect the naming convention

established by the institution.

8. Bulk Import of documents: allows users to mas-

sively upload and index existing documents from

the local institution document storage, such as

their hard-drives, into the DMS.

Figure 1: Hierarchy of the main page of the K-Link DMS.

KITA 2015 - 1st International Workshop on the design, development and use of Knowledge IT Artifacts in professional communities and

aggregations

426

9. Starred: The feature which allows to mark (and

subsequently filter) documents through a favour-

ite set.

10. Make Document Public: allows end-user to pub-

lic a document into the K-Link Network.

Furthermore, different types of users have been de-

fined to better manage the differences in roles and

permissions on which set of actions they can (or can-

not) perform on the documents and collections. This

feature has been widely discussed with both the client

(GIZ GmbH) and the stakeholders. Actually the dif-

ferent permissions over the documents (ability to up-

load, share and organize) reflect the typical organiza-

tional structure of the NGOs, implementing agencies

and research institutes we interviewed in Central

Asia.

As we work in Central Asia the Multilanguage topic

is important. The interface is automatically localized

based on the user’s language, both configured or re-

quested by the browser. The documents could contain

any language supported by the search engine compo-

nent. User generated content is actually considered in

the original language only.

A usability study is planned and will be initially

conducted over 10 users. Moreover, we will add an

A/B testing mechanism to the DMS implementation.

4 K-LINK AS KNOWLEDGE

ARTIFACTS PROMOTER

Initially, the K-DMS was conceived to provide insti-

tutions with an easy way to upload documents and

consequently to feed the K-Link network and trigger

its adoption. Later on, it evolved beyond the needs of

documents uploading in the direction of an infor-

mation management and project management system.

Even more, after the first release while tuning the K-

DMS features, it emerged from stakeholders the need

to be supported in the decision making process for

documents structure organization and conceptualiza-

tions. Thus, the K-DMS started to be re-designed to

cope with the need of mediating the process of infor-

mation conceptualization within SLM communities

of practice.

The established hierarchy of the elements in the

page is showed in Figure 1.

The first header row (1 in Figure 1) is reserved to

show the search form and to show the global naviga-

tion (3), which includes the user profile link. Some

users might not see other pages in the global naviga-

tion other than the current page; this is caused by the

fact that the global navigation is subject to the user

permissions. The second header row is half divided

into the navigation breadcrumbs (2) and the page re-

lated actions (4). The page content is divided into

80% for the main page area (5) and 20% for the doc-

uments navigation sidebar. The first part of the docu-

ments navigation sidebar (6 in Figure 1) is reserved

to show macro documents organization like the

shared, the starred and switch between K-Link pub-

licly available documents and institutional docu-

ments. The second part of the document navigation

sidebar (7) shows the documents Collections.

Collections have been introduced instead of fold-

ers to organize documents. Collections can be per-

sonal or institutional. Personal collections are visible

only to the user that has created it, while the institu-

tional collections are visible to all the users of the in-

stitution that can manage documents. To better distin-

guish between personal and institutional collection a

colour-coded band has been introduced in addition to

the full and empty tag icon, respectively for personal

and institutional collections.

To interact with the collections a contextual menu

was introduced. The concept of Collection is wider

than the folder concept.

A Collection is a logical group of documents that

might not be directly mapped to a single folder on the

file system. A document can be pertaining to more

than one Collection at the same time, while this is not

possible when using folders without duplicating the

document. To organize the documents in collections

the user interaction is based on the drag and drop of

the document, or the selected documents, to a collec-

tion.

Recalling Knowledge Artifact as discussed in

(Cabitza et al., 2013) the definition adheres to the K-

DMS through which different representations of same

sets of information may be allowed and orchestrated

(Kreiner, 2002).

The principle of representational locality (Cabitza

et al., 2013; Váncza and Márkus, 1993) responds to

the need of not imposing an exogenous language or

notation to a community of users. The introduction of

collections exactly responds to this requirement inas-

much as users were already adopting folders to organ-

ize and share documents.

The principle of representational openness, that

means that “a computer-based system should allow

for the introduction of new symbols and signs to

cover the need to represent either portions or aspects

of the reality of interest that were previously, possibly

rightly, disregarded.”. The supplying of labels and

symbols (e.g. stars) whose semantic is completely

user dependent has been done exactly to respect the

representational openness.

KM and KA in International Coooperation - Lesson from the K-Link project in Central Asia

427

Finally, the principle of autonomy in managing levels

of underspecification has been realized by the distinc-

tion of collections into personal and institutional, by

which different degree of specification and granular-

ity may be freely choose by different users collabora-

tive working on the organization of the same set of

documentation.

5 CONCLUSION

The paper has discussed the K-Link project and its

role as promoter of KAs. In the framework of Inter-

national Cooperation addressed by the project, KM

initiatives usually clashes against cultural and politi-

cal barriers, which hinder collaborative forms of

working from being implemented. According to the

requirements analysis carried out with the stakehold-

ers, the way to overcome these bottlenecks was to fol-

low the prescription of conceiving the K-Link solu-

tion according to the general requirements for the de-

sign of a computer-based system acting as a KA. The

decision to implement the approach by integrating the

services provided by the application a distributed con-

tent-based search engine with information manage-

ment features provided to each end-user institution

proved to work well. In addition, the framework of-

fered by the International Cooperation gives to KM

initiatives focused on KA promotion a new applica-

tion domain for experimenting KA as trigger of or-

ganizational development services.

In this respect the K-DMS has been mainly envi-

sioned to create a platform powerful enough to enable

new services.

Actually the design of Collections feature in the K-

DMS is expected to be the starting point for supplying

end users with bottom-up and collaborative concep-

tual modelling services based on the original folders

structure enhanced with a logical layer.

The main actions planned to consolidate K-Link as

knowledge artifact promoter are here:

− To test topic models techniques (TP) for identi-

fying clusters of “conceptually connected” docu-

ments inside the search results so to detect a) the

evolution and distribution of interests along the

Communities of Practice that take part into the

Klink network b) the detection of core indicators

in the field on Sustainable Land Management for

these Communities c) the arrangement of new

facets to support end users in filtering the search

results.

− To use the results of the previous test to support

the modelling of decisional processes. In this di-

rection the integration of tools like Moki is under

study (MK);

− To extend the collaborative features with chat

and annotation systems.

REFERENCES

Baeza-Yates, R. and Ribeiro-Neto, B., 1999. Modern In-

formation Retrieval. Harlow: Addison-Wesley.

Borchardt, P., Schmidt, M. and Schickhoff, U., 2010. Veg-

etation Patterns in Kyrgyzstan’s Walnut-Fruit Forests.

Under the Impact of Changing Forest Use in Post-So-

viet Transformation. In: Die Erde 141 (3). 255-275.

Brill Olcott, M., 1998. Central Asia: Confronting Inde-

pendence. [pdf] Barker Institute: Rice University.

Available at: http://bakerinstitute.org/media/files/Rese

arch/133113ba/central-asia-confronting-independence.

pdf [Accessed 8 September 2015].

Cabitza, F., Colombo, G., Simone, C., 2013. Leveraging

underspecification in knowledge artifacts to foster col-

laborative activities in professional communities. In In-

ternational Journal of Human-Computer Stud-

ies, 71(1), 24-45.

Dutta, S., Geiger, T. and Lanvin, B., 2015. The Global In-

formation Technology Report 2015 ICTs for Inclusive

Growth. Geneva: World Economic Forum.

IWS, 2014. Internet World Stats. [online]. Available at:

http://www.internetworldstats.com/stats.htm/,[Ac-

cessed 7 September 2015].

K-Link, 2015. Introduction. [online] Available at:

http://klink.asia/, [Accessed 7 September 2015].

Kreiner, K., 2002. Tacit knowledge management: the role

of artifacts. Journal of Knowledge Management, 6 (2),

pp.112 – 123.

Kwasnik, B., 1991. The Importance of Factors that are not

Document Attributes in the Organization of Personal Doc-

uments. Journal of Documentation, 47(4), pp.389–398

Pomfret, R., 2006. The Central Asian Economies since In-

dependence. Princeton, NJ. Princeton University Press.

Swetnam, T.W., Allen, C. D. and Betancourt, J., 1999. Ap-

plied Historical Ecology: Using the past to manage for

the future. In: Ecological Applications (4). 1189-1206.

Rondelli, B., 2014. Technical Offer: Regional NRM ZA –

K-LINK,11.2238.1-006.00. Barcelona. SIRIS Academic.

UNICEF, 2008. Learning from KM Experiences. [online]

Available at: http://www.unicef.org/rosa/Learning_fro

m_KM_Experiences.pdf, [Accessed 7 Sept,2015].

Váncza, J. and Márkus, A., 1993. Features and the principle

of locality in process planning. International Journal of

Computer Integrated Manufacturing, 6 (1-2).pp.126-136

TP, http://www.cs.princeton.edu/~blei/topicmodeling.html

MK, https://moki.fbk.eu/website/index.php>

KITA 2015 - 1st International Workshop on the design, development and use of Knowledge IT Artifacts in professional communities and

aggregations

428