Using Open-Source Software to Support Network Management Transformation in Indonesia's Digital Telecommunications Industry

2.1 Research design

This study employed an explanatory sequential mixed-methods design with three integrated components, combining quantitative and qualitative approaches to comprehensively examine OSS adoption in Indonesia's telecommunications sector [1, 7, 11-14, 16].

Quantitative phase

The respondent distribution by region and demographic strata is presented in Table 2, ensuring a balanced representation across Java and the Outer Islands.

• Table 2 surveyed 400 telecommunications professionals, selected through stratified random sampling across Java (62%) and Outer Islands (38%)
• Measured thirteen variables across four domains:

1. Individual characteristics: Gender, age, and education level
2. Organizational factors: Position (e.g., technical staff, managers) and work experience
3. Technical aspects:
   • OSS usage frequency
   • Types of OSS tools deployed (e.g., Docker, Kubernetes)
4. Environmental influences:
   • Regulatory support for OSS adoption
   • Policy alignment with digital transformation goals

Rationale for design choices

Table 1 provides a structured synthesis of the methodological framework employed in this study, detailing theoretical justifications, contextual alignment, and measurement references that underpin the research design.

1. The mixed-methods approach was selected to:
   • Triangulate findings through quantitative data (survey) and qualitative insights (interviews)
   • Address limitations of prior single-method studies in OSS research
2. Sample size (n=400) was determined via power analysis using G*Power 3.1, ensuring:
   • 80% power to detect small-to-medium effects (f²=0.15)
   • Coverage of key demographic strata (urban/rural, company size)

Table 2. Respondent demographics by age group

2.2 Sampling strategy

Quantitative Sampling

This study employed a stratified random sampling approach to ensure representative coverage of Indonesia's telecommunications sector [7].

Target Population

Indonesian telecom professionals involved in OSS projects (N≈5,000 based on APTSI 2022 data [14]).

Sampling Frame

• 85% corporate employees (Telkomsel, XL Axiata, Indosat) [3]
• 15% government/regulatory staff

Stratification Criteria

Key methodological references are compared in Table 3, highlighting the stratified design's role in ensuring representation [7].

Table 3. Methodological references for sampling components

Inclusion Criteria

1. Minimum 1 year OSS experience
2. Currently employed in telecom sector
3. Direct involvement in network management

Qualitative Sampling

Purposive selection was conducted based on survey performance metrics [9, 11, 13]:

• 10 highest-scoring OSS implementers (top 10% by adoption score)
• 5 mid-range performers (40th-60th percentile)
• 5 struggling adopters (bottom 10%)
• Theoretical saturation was achieved at 18 interviews, consistent with established qualitative research standards.

2.3 Hypotheses

The following hypotheses were proposed to guide the analysis of factors influencing OSS adoption and quality in Indonesia's telecommunications industry, focusing on gender and marital status as key demographic variables [17, 18].

Hypothesis 1 (H1)

Gender has a significant effect on the quality of OSS used in network management.

This hypothesis examines gender-based variations in OSS perception and utilization, building on prior findings about technology adoption disparities in Southeast Asia.

Rationale:

• Supported by literature on gender differences in technology acceptance
• Aligns with Indonesia's workforce demographics showing 68.3% female participation in tech roles

Hypothesis 2 (H2)

Marital status affects OSS adoption and use in Indonesia's telecommunications industry.

This hypothesis tests whether marital responsibilities influence time allocation and technology engagement, extending prior work-life balance studies in emerging markets [14, 16, 18].

Rationale:

• Based on stress-adaptation theories in technology adoption [19]
• Accounts for Indonesia's cultural context of familial obligations

Testing Methodology

Both hypotheses were evaluated through:

1. Quantitative analysis:

• Independent samples t-tests (gender)
• ANCOVA controlling for age/experience (marital status)

2. Qualitative validation:

• Thematic analysis of interview transcripts
• Table 4 shows the quality levels across working domains

3. Key: NS = Not Significant (α=0.05); κ = Cohen's kappa inter-coder reliability

Interpretation of Findings

1. H1 (Gender):

• While statistically insignificant (p=0.12), qualitative data revealed:
  • 68.3% of female respondents showed slightly higher OSS quality scores.
  • Stronger peer-learning tendencies among female professionals (κ=0.81)

2. H2 (Marital Status):

• No direct effect (p=0.75), but qualitative insights identified:
  • Workplace flexibility as a compensatory factor (18/20 interviewees)
  • Organizational policies mitigated marital status impacts (κ=0.79)

To ensure methodological rigor in interpreting both the quantitative and qualitative findings, this study draws upon established frameworks for hypothesis formulation, statistical testing, and mixed-methods integration. These methodological underpinnings, including the application of ANCOVA and Cohen’s kappa reliability assessment, provide a robust foundation for validating demographic influences on OSS quality. The full list of methodological references supporting this phase of the research is summarized in Table 5.

Table 4. Quality levels across working domains

Table 5.  Methodological references supporting hypothesis testing and mixed-methods validation

Key Takeaways:

1. Results align with global OSS studies showing minimal demographic effects
2. Highlights the importance of organizational factors over demographics
3. Demonstrates value of mixed-methods approaches in contextualizing null findings

2.4 Integration protocol

This study implemented a three-phase mixed methods integration framework to connect quantitative and qualitative findings:

1. Connected analysis

• Outlier Follow-up: Conducted in-depth interviews with 5% of survey outliers (n=20) identified via:
  
  • Mahala Nobis distance (p<0.01)
  • Residual analysis in SEM (|z|>2.58)
• Protocol:
  
  • Semi-structured interviews (45-60 mins)
  • Member checking with participants
  • Triangulated with HR records

2. Embedded validation

• Qualitative-SEM Mapping:

3. Meta-inference generation

The following are four evidence-based principles through joint displays:

Principle 1: Prioritize competency-based hiring

• Supported by:

• Quantitative: Certification had β=-0.18 vs. skills β=0.42
• Qualitative: "We test GitHub contributions, not diplomas" (14/20 interviewees)

Principle 2: Optimal team experience mix (30%<30yr, 50% 6-10yr, 20%>15yr)

• Validation:

• Peak performance at 6-10 years (M=68.7)
• "Cross-generational mentoring works" (κ=0.77)

Principle 3: 3-stage OSS adoption roadmap

1. Pilot (0-6mo): Limited-scope POCs
2. Scale (6-18mo): Departmental integration
3. Optimize (18-36mo): Full automation

Principle 4: Hybrid training programs

• Blended:
  
  • 70% hands-on labs (Docker/K8s)
• 30% policy/compliance training Efficacy: 23% higher retention vs. traditional (p<0.05)

Methodological Rigor

• Trustworthiness Criteria:
  
  • Credibility: Prolonged engagement (12mo fieldwork)
  • Transferability: Thick description of Indonesia context
  • Dependability: Audit trail maintained in NVivo [11]
  • Confirmability: 80% inter-coder agreement (κ>0.75)

To enhance construct validity, qualitative themes derived from open-ended responses were systematically aligned with relevant constructs in the Structural Equation Model (SEM). This alignment was guided by existing OSS adoption studies emphasizing skills-based readiness over formal qualifications. Each code was mapped to its theoretical counterpart using standardized factor loadings (β), ensuring empirical consistency. The full code-to-construct mapping and corresponding references are presented in Table 6.

Table 6. Code-to-construct alignment

***p<0.001, **p<0.01 (2-tailed) * [12]

2.5 Sample size justification

Power analysis calculation

In Table 7, we determined the minimum sample size required using GPower 3.1 [12] with the following parameters:

Table 8 shows sample size benchmarking calculation steps:

1. Base Requirement:
   
   • 129 participants (for f²=0.15, power=0.80, α=0.05)
2. Non-Response Adjustment:
   
   • +20% buffer → 155 (accounts for Indonesia's survey response rates)
3. Design Effect:

• ×2.5 for stratified sampling (intraclass correlation ρ=0.06)

4. Final Target:

• 388 → Rounded to 400 for equal stratification across:

• Java (n=248) and Outer Islands (n=152)
• Urban (n=292) and Rural (n=108)

To validate the appropriateness of the sample size (n=400), a benchmarking analysis was conducted against established OSS studies in similar domains. The comparative overview is presented in Table 8, illustrating consistency with regional and global research standards.

Table 7. Power analysis calculation

Table 8. Comparative benchmarking of OSS sample sizes from prior studies

Table 9. Sampling precision metrics

Table 10. Methodological references

Sensitivity analysis

1. Post-Hoc Power:
   
   • Actual power = 82.4% for smallest effect (f²=0.12)
2. Monte Carlo Simulation (1,000 replications):
   
   • 95% CI coverage achieved at n=372
   • Type I error rate = 4.9-5.3% (within expected range)
3. Thematic Saturation:
   
   • In Table 9, no new qualitative themes emerged after the 350 surveys.
   • Achieved κ>0.80 inter-coder reliability

In Table 10, methodological references on key points:

1. Sample size exceeds minimum requirements for all planned analyses
2. Stratification ensures representativeness of Indonesia's telecom workforce
3. Sensitivity analyses confirm robust statistical power

2.6 SEM analysis

Model specification

In Figure 1, we develop a confirmatory model consisting of three latent constructs, which build on the Technology-Organization-Environment (TOE) theoretical framework [20]:

1. Technical Readiness (α=0.87)
   
   • Age (β=0.32***)
   • Education (β=0.41***)
   • OSS skills (β=0.68***)
   • Tool proficiency (β=0.65***)
2. Organizational Support (α=0.83)
   
   • Training access (β=0.72***)
   • Leadership commitment (β=0.69***)
   • Resource allocation (β=0.63***)
3. Regulatory Environment (α=0.79)
   
   • Policy alignment (β=0.81***)
   • Incentives (β=0.77***)

1.png

Figure 1. SEM path diagram of OSS adoption factors

Estimation Method [21]

• Estimator: Robust Maximum Likelihood (MLR) to handle non-normal data

• Missing Data: Full Information Maximum Likelihood (FIML) with 4.2% missingness

• Software: Mplus 8.6 with 1,000 bootstrap samples

Model Fit Indices

To assess structural model adequacy, key fit indices were benchmarked against recommended thresholds. The results, as shown in Table 11, indicate satisfactory model performance with acceptable values across all levels.

1. Initial Model: Poor fit (CFI=0.872, RMSEA=0.112)
2. Modifications:

• Added error covariances:

• OSS skills ↔ Tool proficiency (MI=12.7, β=0.38***)
• Leadership ↔ Resources (MI=9.3, β=0.42***)

• Removed non-significant path:

• Regulatory → Technical Readiness (β=0.08, p=.412)

Table 11. Model fit indices across SEM specifications and thresholds

Robustness Checks

• Measurement Invariance [22]:

To validate the stability of the model across measurement conditions, invariance testing was conducted across configural, metric, and scalar levels. The resulting CFI changes support partial invariance, as shown in Table 12.

Table 12. Results of multi-level invariance testing across measurement models

3. Predictive validity [22]:

• 82.3% classification accuracy
• Stone-Geisser Q²=0.78 (large effect)

4. Effect sizes:

• Largest: Technical→OSS Quality (β=0.42, f²=0.28)
• Smallest significant: Regulatory→Organizational (β=0.29, f²=0.11) [23]

• Key Findings
• Standardized path coefficients:

• Technical → OSS Quality: 0.42*** (p<0.001)
• Organizational → OSS Quality: 0.38*** (p<0.001)
• Regulatory → Organizational: 0.29* (p=0.023)

• Variance Explained:

• OSS Quality: R²=0.63
• Organizational Support: R²=0.41

2.6.3 Sensitivity analysis

(1) Threshold Variations [12]:

1) Confirmed stability under both:

• ΔCFI<0.01

• ΔCFI<0.002

2) Monte Carlo Validation:

• 500 replications showed 91% correct model specification

Mplus Syntax Excerpt

TITLE: Final SEM Model

DATA: FILE = OSS_Adoption.dat

VARIABLE: NAMES = age educ skills...;

MODEL:

  Tech_Read BY skills toolprof educ age;

  Org_Supp BY train lead res;

  Reg_Env BY policy incent;

  OSS_Qual ON Tech_Read Org_Supp;

  Org_Supp ON Reg_Env;

OUTPUT: STDYX MODINDICES(7.0);

Key Takeaways:

1. Model demonstrates strong psychometric properties
2. Technical readiness is the strongest OSS adoption predictor
3. Results are robust across demographic subgroups

Indices Analysis

Figure 2 shows the indices heatmap. This visualization displays the modification indices between all pairs of observed variables in SEM model. Higher values (shown in warmer colors) indicate pairs where model fit could be improved by allowing a direct relationship.

2.png

Figure 2. Modification indices heatmap for SEM model

Post-Hoc Analysis and Model Validation

Modification Indices Analysis

The heatmap revealed two key areas for improvement:

1. Highest MI:

• OSS Skills ↔ Tool Proficiency (MI=12.7)

• Justification: Shared method variance from self-reported competency scales
• Solution: Added correlated errors (ΔCFI=+0.032)

2. Secondary MI:

• Leadership ↔ Resources (MI=9.3)

• Rationale: Organizational factors naturally covary in practice
• Solution: Added covariance path (ΔRMSEA=-0.014)

Measurement Invariance Deep Dive

We conducted multi-group analysis following the guidelines below [12]:

1. Age Groups Comparison:

• Metric invariance achieved (ΔCFI=0.005)
• Scalar non-invariance detected in:

• Tool proficiency intercepts (Δ=0.32, p<.01)
• Incentives intercepts (Δ=0.28, p<.05)

• Interpretation: Younger workers undervalue certifications

2. Company Size Comparison:

• Full metric invariance (ΔCFI=0.003)
• Partial scalar invariance with 3/9 varying parameters:

• Training access (Δ=0.41, p<.001)
• Policy awareness (Δ=0.38, p<.01)

• Implication: SMEs need tailored training approaches

Effect Size Benchmarking

Table 13 benchmarks our SEM coefficients against global studies, revealing that technical readiness (β=0.42) exerts a stronger influence in Indonesia than in Hauge et al. [9] sample (β=0.38). This suggests that Indonesia’s rapid 5G rollout [3] heightens the urgency of technical upskilling, while organizational support (β=0.38) aligns with Linux Foundation’s [1] findings, emphasizing universal best practices. Policymakers should note the weaker regulatory impact (β=0.29) versus technical factors, advocating for sandbox environments to mitigate risk [14].

Table 13. Compared to prior studies

Practical Significance Assessment

1. Technical Readiness Impact:

• 1 SD increase → 23% higher OSS adoption (p<.001)
• Accounts for 28% of quality variance

2. (9) Organizational Support:

• Companies with strong support had:

• 40% faster implementation (p<.01)
• 31% lower staff resistance (p<.05)

Limitations and mitigations

1. Common Method Bias:

• Addressed through:

• Procedural separation of surveys/interviews
• Harman's single factor test (18.3% variance)

2. Sample Characteristics:

• Overrepresentation of Java (62%) mitigated by:

• Stratified sampling weights
• Post-hoc subgroup analysis

Implementation Recommendations

Based on SEM findings, we propose:

1. Technical Upskilling:

• Focus on hands-on OSS training (β=0.68***)
• Certification programs with practical components

2. (4) Organizational Changes:

• Leadership training programs (β=0.69***)
• Resource allocation guidelines

3. (5) Policy Enhancements:

• Clear incentive structures (β=0.77***)
• Regulatory sandbox for OSS experimentation [24]

Advanced SEM Diagnostics and Robustness Verification

1. Nonlinearity and Interaction Effects

We tested for potential nonlinear relationships and moderating effects using:

A. Latent Moderated Structural Equations (LMS)

• Evaluated age × experience interaction on OSS adoption
• Significant interaction (β = 0.18*, p = 0.022) suggests:

• Younger workers benefit more from experience
• Practical implication: Targeted mentoring for <30 cohort

B. Quadratic Terms

• Added squared terms for continuous predictors:

• Organizational support showed diminishing returns (β = -0.12*)
• Technical readiness maintained linear relationship

2. Alternative Model Comparison

To evaluate theoretical robustness and parsimony, several alternative SEM specifications were compared to the proposed baseline model. The results of this comparative analysis are summarized in Table 14.

Table 14. Competing model fit comparison

Key Finding: Baseline model demonstrates superior parsimony while maintaining theoretical coherence

3. Cross-Validation Procedures

A. k-Fold Validation (k=5)

• Average prediction error:

• Training set: 12.3%
• Test set: 14.7% (Δ=2.4%, acceptable)

B. Holdout Validation

• 30% random holdout sample (n=120)
• Maintained 80.1% classification accuracy vs. full sample

4. Effect Decomposition

To examine the indirect effects among constructs, mediation analyses were conducted using bootstrapped path estimates. The results of this analysis are detailed in Table 15.

Table 15. Results of bootstrapped mediation analysis on OSS adoption pathways

Interpretation: Organizational support fully mediates regulatory impacts while partially mediating technical effects [16]

5. Reliability Generalization

A. Composite Reliability

• Technical Readiness: ρ = 0.86
• Organizational Support: ρ = 0.84
• Regulatory Environment: ρ = 0.81

B. Hierarchical Omega

• Second-order construct reliability: ω = 0.89

6. Predictive Validity Extension

12-Month Longitudinal Follow-up (n=87 companies):

• Model predicted actual OSS adoption with:

• Accuracy: 78.4%
• Sensitivity: 81.2%
• Specificity: 75.6%

7. Bayesian SEM Verification

Markov Chain Monte Carlo (MCMC) Estimation:

• Convergence achieved (PSRF ≈ 1.01)
• 95% Credible Intervals aligned with ML estimates:

• TECH→OSS: [0.37, 0.47]
• ORG→OSS: [0.31, 0.45]

8. Multicollinearity Diagnostics

• Maximum VIF: 2.8 (below 5.0 threshold) [25]
• Condition Number: 19.3 (acceptable <30)

9. Effect Heterogeneity Analysis

Latent Class Analysis Revealed:

• Class 1 (65%): Strong TECH effects (β=0.51***)
• Class 2 (35%): Strong ORG effects (β=0.47***)
• Segment characteristics:

• Class 1: Large enterprises, legacy systems
• Class 2: Startups, cloud-native

10. Final Model Recommendations

Based on comprehensive diagnostics:

1. For Large Enterprises:

• Prioritize technical infrastructure upgrades (TECH β=0.51)

2. For SMEs:

• Focus on leadership buy-in and training (ORG β=0.47)

3. Policy Implications:

• Tiered regulatory approaches based on company maturity

This study identified the factors influencing the adoption of open-source software in the Indonesian telecommunications industry, namely, technology readiness, management support, and regulation. This chapter discusses how these findings can be applied in practice and provides directions for further research that can deepen our understanding of technology adoption in the telecommunications industry. This study aimed to evaluate the quality of open-source software (OSS) within the context of Indonesia's telecommunications industry, focusing specifically on its role in network management transformation. To achieve this, comprehensive data were gathered from 400 participants engaged in OSS development across various regions, including Jakarta and Surabaya, Indonesia.

Utilizing a structured interview schedule, the research collected first-hand information that was subsequently organized into suitable tabular formats for analysis. A variety of statistical tools were employed, including percentages, averages, standard deviations, chi-square tests, multiple regression analysis, and factor analysis, to scrutinize the data thoroughly. Based on the outcomes of these analyses, key findings were identified and summarized, with several recommendations proposed to enhance the quality and efficacy of OSS in the telecommunications sector

This chapter is structured as follows:

• Implications of Research Findings:
• Explain the practical contribution of research findings and how these findings can be applied by telecommunications companies and policymakers in the future.
• The theoretical implications of the research    findings are discussed in the context of the existing literature.
• Recommendations for Future
• Research We provide suggestions for further research that can deepen and broaden our understanding of open-source software adoption.
• Identify areas that require additional research based on the findings and limitations of this study.
• Chapter Conclusion: This chapter concludes the main points and leads the reader to the overall conclusion of the study.

By discussing the implications of the findings and providing recommendations for future research, this chapter aims to provide practical and theoretical guidance for practitioners and researchers in developing effective strategies and policies for adopting open-source software in the telecommunications industry. Furthermore, the first part of this chapter will discuss the implications of research findings.

4.1 Resolving the age-experience paradox

Key Findings Requiring Explanation

• Under-30 cohort outperformed older groups (68.3 vs. 66.4, *p*<.01)
• 6–10-year experience group showed peak performance (68.7)
• Traditional telecom studies typically favor >15yr experience (Linux Foundation, 2020)

Digital Native Advantages (Technical Factors)

1. Learning Adaptability
   
   • 73% of under-30 respondents reported self-taught OSS skills via:
     • Interactive platforms (GitHub: 82% usage)
     • Microlearning (55% used Codecademy/Duolingo-style apps)
   • Only 29% of over-40 respondents utilized these resources (*p*<.001)
2. Tool Familiarity
   
   • Native comfort with:
     • Containerization tools (Docker/Kubernetes adoption: Under 30=74%, Over40=31%)
     • CI/CD pipelines (Under30=68%, Over40=27%)

Table 29 further illustrates these digital skill gaps by age, showing that younger respondents (<30 years old) significantly outperformed their older counterparts in cloud-native development (81% vs. 39%), AI-assisted coding (63% vs. 18%), and multi-platform debugging (72% vs. 45%). These differences are statistically significant (p < .001).

Table 29. Digital native skill advantages

Indonesia-Specific Dynamics (Contextual Factors)

1. Industry Transformation
   
   • Rapid 5G rollout (2020-2024) prioritized new skills over legacy knowledge
   • The Interview data revealed the following:

"The shift to Open RAN made 20-year radio expertise less relevant than Python scripting skills" (Telkomsel Architect, 35)

2. Education System Shifts
   
   • Vocational schools (diploma programs) emphasized as follows:
     • Practical OSS contributions (67% included in curricula)
     • Hackathon participation (42% of under-30 sample)
   • Traditional university programs lagged (only 19% included OSS modules)
3. Labor Market Patterns
   
   • Salary data showed:
     • 23% premium for OSS skills vs. legacy telecom certifications
     • 15% faster promotion track for under-30 high performers

Experience Hybridization Effect

Figure 7 illustrates the ‘sweet spot’ (6–10 years of experience) where professionals blend foundational telecom knowledge with OSS skills. This hybrid competency aligns with Indonesia’s 5G rollout demands [3], suggesting targeted training for mid-career workers (30–40 years) to bridge legacy and modern systems The 6–10-year experience group succeeded by blending:

• Foundational telecom knowledge (core network protocols)
• Acquired digital skills (self-reported 3.2x more OSS upskilling than >15yr group)

In Figure 7, the X-axis shows years of experience, while the Y-axis represents the OSS performance score. Notice the clear peak in performance between 6 and 10 years of experience—this is highlighted as the "Optimal hybrid competency window." This suggests that professionals with 6–10 years of experience tend to achieve the highest OSS performance, likely because of a blend of foundational knowledge and practical expertise.

7.png

Figure 7. The optimal hybrid competency window (6–10 years experience) balances foundational knowledge and digital skills [7]

This insight has important policy implications.

1. Workforce Development
   
   • Targeted reskilling for mid-career (30-40yr) professionals
   • "Reverse mentoring" programs pairing junior OSS experts with senior domain specialists
2. Education Reform
   
   • Accelerate OSS integration in university curricula
   • Industry-recognized microcredentials for critical skills (e.g., Kubernetes for Telecom)

4.2 ASEAN comparative analysis of age-performance dynamics

In Figure 8 Indonesia’s exceptional youth-driven adoption contrasts with regional peers like Singapore and Vietnam, where mid-career professionals dominate OSS deployment [15]. According to GSMA (2024), Indonesia’s ‘Kubernetes for 5G’ policy (2022) accelerated youth uptake, while Singapore’s certification requirements favor experienced professionals [15]. This divergence underscores the need for age-agnostic skill assessments, as exemplified by Vietnam’s ‘FOSS First’ mandate.

Key Contrasts with Indonesia

1. Education Systems
   
   • Vietnam's Advantage:
     • 92% of IT curricula include mandatory OSS contributions vs. Indonesia's 67%
     • Earlier coding exposure (Grade 6 vs Indonesia's Grade 10)
   • Thailand's Challenge:
     • Only 19% of universities teach Git vs 53% in Indonesia
2. Industry Adoption Rates

Correlation and scatterplot displayed for OSS adoption by age group.

As shown in Table 30, there is a strong negative correlation (r = –0.72) in OSS adoption rates between younger (<30) and older (>40) professionals across ASEAN countries in 2023. The scatterplot visualizes this relationship, highlighting that countries with high youth adoption (such as Indonesia) tend to have lower adoption among older professionals, and vice versa (such as Singapore). This suggests generational divides in OSS uptake, which could inform targeted policies or training interventions.

8.png

Figure 8. ASEAN comparative analysis

Table 30. Highlights Indonesia’s exceptional youth-driven adoption versus regional peers

3. Regulatory Environments
   
   • Singapore: Strict certification requirements favor experienced professionals
   • Philippines: BPO-driven demand for young, low-cost OSS talent
   • Indonesia: Unique "Kubernetes for 5G" policy (2022) accelerated youth uptake

Case Study: Vietnam's Hybrid Model

• Success Factors:
  
  • Government-led "FOSS First" policy in state projects
  • 30% salary bonus for OSS certifications
• Replicable Insights:
  
  • Standardized OSS competency frameworks
  • Age-blind skill assessments

Policy Cross-Learning Opportunities

1. From Singapore:
   
   • Stackable micro credentials for mid-career professionals
2. From Philippines:
   
   • Youth apprenticeship programs with telcos
3. From Vietnam:
   
   • Public-sector OSS adoption mandates

Line chart of OSS performance by age group and country displayed.

This line chart shows the normalized OSS performance across age groups (25–55) for Indonesia, Singapore, and Thailand. Indonesia peaks early (under 30), Singapore peaks in the mid-career range (35–40 years), and Thailand peaks later (40+ years). These trends highlight how each country's workforce develops OSS competency at different career stages, which can inform targeted talent development and policy.

1. Implications for Telecommunications Companies

2. Increased Technology Readiness:

•   Technology Infrastructure:

Telecommunications companies must invest in adequate technological infrastructure to support the adoption of open-source software. This includes ensuring system compatibility, security, and the availability of technical resource 30% in 440% Development

• Training and Development:

Investing in training and developing employees’ technical skills is essential. Companies may conduct specialized training programs to improve their understanding and technical skills necessary to implement and maintain open-source software.

Gradual Adoption:

Gradual adoption of open-source software can help companies manage risks and facilitate smoother transitions. For example, a company can start with a small project before expanding the use of open-source software to the entire organization or department.

• Strengthening Management Support:

In Figure 9, leadership commitment top management must actively champion open-source software (OSS) adoption through:

9.png

Figure 9. ASEAN age-performance curves

Resource allocation: Dedicated budgets for OSS training and infrastructure (e.g., 23% of IT budgets in high-adoption firms [25]). Policy alignment: Integrating OSS into corporate digital transformation roadmaps [26]. Visible advocacy: Regular communication of OSS benefits (e.g., cost savings up to 40% [27]). Organizational Culture Foster innovation by: Incentivizing OSS contributions (e.g., GitHub activity as a KPI [11]). Establishing cross-functional OSS committees to bridge technical and operational teams [28]. Change Management Address resistance through: Pilot projects demonstrating quick wins (e.g., 6-month POCs improved adoption by 31% [29]). Training programs tailored to managerial roles (e.g., "OSS for Decision Makers" workshops [30]).

• Top Management Commitment:

Support from top management is critical for the successful adoption of open-source software. Management must demonstrate a strong commitment through resource allocation, policy support, and effective communication regarding the benefits of open-source software.

• Culture of Innovation:

Creating an organizational culture that supports innovation and technological change can help reduce resistance to the adoption of open-source software. Management can encourage openness to new technologies and appreciate employees’ innovative initiatives.

• Dedicated Team Creation:

Establish a dedicated team of IT experts, project managers, and business leaders to oversee the adoption and implementation of open-source software. This team ensures that all technical and operational aspects are well managed.

3. Utilizing Supportive Regulations:

Collaboration with Policymakers:

Telecommunications companies can collaborate with governments and regulatory agencies to leverage the incentives and support available for open-source software adoption. This includes participation in fiscal incentive programs, research grants, and technical support.

Regulatory Compliance:

Ensuring regulatory compliance that supports the use of open-source software can help companies avoid legal barriers and increase stakeholders’ trust.

Policy Advocacy:

Companies can engage in policy advocacy to encourage the establishment of regulations that better support the adoption of open-source software. This includes providing input to policymakers on the needs and challenges facing the industry.

The analysis yielded several significant conclusions regarding the quality of OSS implementations and the factors that contribute to this quality.

1. Quality Among Genders: The findings indicate that female respondents maintained a higher level of quality in OSS than their male counterparts, suggesting a need for organizations to encourage male participation in OSS development.

2. Age Factor: Young respondents (under 30 years) demonstrated a notable commitment to maintaining high-quality OSS, with results validated by chi-square tests at a 1% significance level.

3. Marital Status: Unmarried respondents exhibited higher OSS quality levels than married respondents, indicating that marital responsibilities may impact engagement in OSS activities.

4. Family Structure: Respondents from joint families reported higher quality levels in OSS, implying that the support structure of joint families may foster better collaboration and resource sharing in OSS development.

5. Educational Background: Individuals with diploma- or certificate-level education showed a higher level of OSS quality than those with graduate or postgraduate degrees, suggesting that practical technical training might be more beneficial for OSS quality than formal education alone.

6. Industry Experience: Respondents working in sectors outside traditional IT roles, such as education and training, also maintained higher OSS quality levels, indicating that diverse professional backgrounds can contribute positively to OSS outcomes.

7. Role of Software Programmers: Those engaged as software programmers in development exhibited the highest OSS quality levels, as confirmed by chi-square tests at a 1% significance level.

8. Experience Duration: Respondents with 6-10 years of experience in the IT sector maintained higher OSS quality levels, supported by chi-square tests at a 1% significance level, suggesting that moderate experience is particularly beneficial.

9. Participation Motivation: Individuals who participated in OSS development primarily for their organization's needs exhibited a strong commitment to maintaining OSS quality, validated at the 1% significance level.

10. Tool Usage: Respondents utilizing communication tools in their OSS activities reported higher quality levels compared to those using other types of software tools.

11. Income Levels: Those earning above 10 million IDR annually maintained superior OSS quality, as confirmed by chi-square tests at a 1% significance level.

12. Supplemental Income from OSS: Respondents with additional income from OSS development below 3 million IDR also demonstrated high-quality maintenance, indicating the importance of supplementary financial motivation in OSS involvement.

13. Correlation with Quality Factors: Multiple regression analysis indicated a positive correlation between OSS quality and factors such as gender, age group, educational status, work status, experience, need for participation, extent of OSS usage, and income from OSS development activities.

14. Discriminant Function Analysis: This analysis revealed that factors such as gender, educational status, work status, and income significantly distinguish between high- and low-quality OSS maintainers.

15. Factor Analysis Outcomes: This study highlighted the significance of OSS through four categories: the impact of OSS, the operative function of development, the efficiency of development, and the reliability essential for volunteers.

16. Cluster Analysis: Respondents were categorized into four groups: Quality Assurance Testers, Defect Handlers, Project Managers, and Communicators, each playing distinct roles in OSS quality maintenance.

17. Major Issues Identified: The Henry Garrett Ranking Technique revealed that significant challenges included the necessity for software to define hardware accessibility and the difficulty in identifying stable software releases, which are critical to maintaining OSS quality.