How can contributors to open-source communities be trusted? On the assumption, inference, and substitution of trust

From WikiLit
Jump to: navigation, search
Publication (help)
How can contributors to open-source communities be trusted? On the assumption, inference, and substitution of trust
Authors: Paul B. de Laat [edit item]
Citation: Ethics and Information Technology 12 (4): . 2010.
Publication type: Journal article
Peer-reviewed: Yes
Database(s):
DOI: 10.1007/s10676-010-9230-x.
Google Scholar cites: Citations
Link(s): Paper link
Added by Wikilit team: Added on initial load
Search
Article: Google Scholar BASE PubMed
Other scholarly wikis: AcaWiki Brede Wiki WikiPapers
Web search: Bing Google Yahoo!Google PDF
Other:
Services
Format: BibTeX
How can contributors to open-source communities be trusted? On the assumption, inference, and substitution of trust is a publication by Paul B. de Laat.


[edit] Abstract

Open-source communities that focus on content rely squarely on the contributions of invisible strangers in cyberspace. How do such communities handle the problem of trusting that strangers have good intentions and adequate competence? This question is explored in relation to communities in which such trust is a vital issue: peer production of software {(FreeBSD} and Mozilla in particular) and encyclopaedia entries {(Wikipedia} in particular). In the context of open-source software, it is argued that trust was inferred from an underlying ‘hacker ethic’, which already existed. The Wikipedian project, by contrast, had to create an appropriate ethic along the way. In the interim, the assumption simply had to be that potential contributors were trustworthy; they were granted ‘substantial trust’. Subsequently, projects from both communities introduced rules and regulations which partly substituted for the need to perceive contributors as trustworthy. They faced a design choice in the continuum between a high-discretion design (granting a large amount of trust to contributors) and a low-discretion design (leaving only a small amount of trust to contributors). It is found that open-source designs for software and encyclopaedias are likely to converge in the future towards a mid-level of discretion. In such a design the anonymous user is no longer invested with unquestioning trust.

[edit] Research questions

"In all a three-fold dynamics of trust emerges: of assumption, inference, and substitution of trust. After a brief survey of open-source communities in general, arguments are presented, explaining why trust is a central issue in the functioning of two particular examples: software and encyclopaedias. The remainder of the paper is then devoted to examining how this dynamics of trust unfolds within the two communities."

Research details

Topics: Ethics, Other collaboration topics [edit item]
Domains: Philosophy and ethics, Information systems [edit item]
Theory type: Analysis [edit item]
Wikipedia coverage: Case [edit item]
Theories: "To me, her theorizing about the assumption of trust

seems to be the more plausible avenue in the case of Wikipedia. By opening up their entries to immediate modification, contributors appeal to the encyclopaedic capabilities of unknown others ‘out there’. This theorizing about the ‘normative pressure’ emanating from opening up content would seem to confirm the conjecture that, especially in cyberspace, assuming trust is an important mechanism for creating trust in the first place (cf. de Laat 2005). In this vein both Wikipedia and diaristic blogs revealing personal intimacies (as analyzed in de Laat 2008) seem to rely boldly on the mechanism that trust can be produced ex post." [edit item]

Research design: Conceptual [edit item]
Data source: N/A [edit item]
Collected data time dimension: N/A [edit item]
Unit of analysis: N/A [edit item]
Wikipedia data extraction: N/A [edit item]
Wikipedia page type: N/A [edit item]
Wikipedia language: English, German [edit item]

[edit] Conclusion

"Open-source communities rely squarely on the contributions of mostly anonymous strangers in cyberspace, so a central concern, whenever these focus on ever-evolving content, such as software modules or encyclopaedic entries, is the problem of whether and to what extent such volunteers can be trusted to contribute in good faith and in a competent fashion. It has been argued that such communities do indeed have a whole array of mechanisms at their disposal to handle this matter of trust: Table 1 lists the mechanisms involved. When rules and regulations are still few and far between, the full weight falls on the processes of inferring or assuming the presence of trust. Some assurances may be generated by an underlying shared culture. When hackers started to use the Internet to develop OSS, they had been cooperating with each other for decades in ‘real life’ and developing a shared ‘hacker ethic’. On the Internet they now bumped into a younger generation of new hackers. If Mizrach (1997) is right, their ‘new hacker ethic’ shared a considerable overlap with the old one. As a result the potential open source audience as a whole was tied to some kind of hacker ethic and their trustworthiness would appear to be guaranteed. The Wikipedia experiment was not so fortunate: no ‘encyclopaedic ethic’ existed at the outset, so when vandalism and harassment started to emerge— revealing that not all anonymous contributors could merely be assumed to be trustworthy—the community hurriedly embarked on a campaign to educate actual and potential Wikipedians. In so far as contributors come to feel bound by this emerging ‘Wikiquette’, they can surely be trusted to further the encyclopaedic cause. In general, inference of trustworthiness preferably relies on solid signs from the particular trustee involved. Opensource communities, however, depend on anonymous strangers who usually do not reliably signal anything and are merely represented by the IP addresses assigned to them. Such communities can therefore only try to gauge the existence of a climate of trust among their potential contributors in general and infer in a weak sense that they probably can be trusted. Due to the veil cast by the Internet, open-source communities are condemned to forming probability estimates of trustworthiness in place of certainties. Note, though, that another, more robust option for inferring trust is becoming available. Lately, within OSS circles the hacker ethic is seen to be eroding. In response some projects subject role occupants to stricter screening, especially when granting commit privileges is at stake. Solid proof of one’s allegiance to the hacker ethic is required. ‘Strong’ inference of trust, therefore, may replace ‘weak’ inference of trust. As long as an appropriate kind of ethic seems to be lacking within the community, assuming trust is the prominent mechanism, by default. Wikipedia has been shown to be a case in point. Posting encyclopaedic entries on the Internet as a wiki is an appeal to fellow Wikipedians to show their editing capabilities. It is a sign of ‘substantial’ trust (in the sense of McGeer 2008). This theorizing about the ‘normative pressure’ emanating from opening up content would seem to confirm the conjecture that, especially in cyberspace, assuming trust is an important mechanism for creating trust in the first place (cf. de Laat 2005). In this vein both Wikipedia and diaristic blogs revealing personal intimacies (as analyzed in de Laat 2008) seem to rely boldly on the mechanism that trust can be produced ex post. Sooner or later open-source communities start to introduce rules and regulations to manage the complexities involved. The governance tools distinguished above are modularization, formalization, division of roles and decision making. While the first two parameters would seem to meet with universal approval among participants, the last two need to be introduced with care. An important choice has to do with the amount of discretion granted to collaborators. In a high-discretion design, role division is minimal and decision making decentralized; as a result, the trust granted remains high. A low-discretion design, with elaborate division of roles and centralized decision making, leaves little discretion to collaborators; as a result, trust is to a large extent substituted. This third mechanism of substitution of trust was first explored with reference to OSS projects. A typical division of roles consists of contributors, committers and maintainers (or module owners). I have shown how FreeBSD tends towards a higher discretion design, while Mozilla tends towards a design of lower discretion. Similarly, the encyclopaedic Wikipedia project has become the subject of design. The fact is that their contributors retain a broad measure of discretion: anonymous users may still edit entries of their choice (and may create new entries after registration). Rising vandalism has not been combated by reducing the discretion of ordinary users, but by introducing ‘administrators’ who are granted powers to block particular users and freeze articles involved in edit wars. As a result, users stay at the helm in Wikipedia. Nevertheless, rampant vandalism has led to mounting pressures within Wikipedia for another measure: review of all changes (edits) with regard to vandalism. This system is undoubtedly a step that will reduce the discretion of ordinary users. It leads to the introduction of levels of trustworthiness, measured by one’s past performance within Wikipedia. Interestingly, this is not uniformly appreciated across language versions. It has already been introduced in the German Wikipedia (and some others), where the majority applauds the rules as a contribution to the fight against vandalism. Their English-speaking fellow Wikipedians, however, vehemently resist its introduction as an encroachment on their editing rights. To them, this is the onset of bureaucratic control. This finding suggests that intercultural perceptions of open-source design may differ considerably. This would seem to be an interesting research field, ripe for exploration: how essentially the same virtual organization is perceived, evaluated and designed across different cultural domains. As a logical extension, plans are afoot within Wikipedia to appoint ‘super-reviewers’ who are to check articles for their quality. If and when such a system of review plus super-review may become standard practice, the user will no longer reign supreme in crafting Wikipedian entries. Such a development would imply two remarkable processes of ‘structural convergence’: on the one hand, Wales’ encyclopaedia would then be governed in a similar fashion to comparable—but smaller—virtual encyclopaedic undertakings like Citizendium, h2g2 and Knol. These projects also apply review procedures to guarantee quality. On the other hand, generally speaking, roles within online open encyclopaedic projects would then resemble those that are usually discerned in OSS. The open-source communities for producing encyclopaedias on the one hand and software on the other would be managed in similar ways. The communities involved would seem to be concurring on the verdict that open-content production cannot do without a process of moderation. For all of them, unquestioning trust in users has proved to be an unworkable assumption"

[edit] Comments


Further notes[edit]

Facts about "How can contributors to open-source communities be trusted? On the assumption, inference, and substitution of trust"RDF feed
AbstractOpen-source communities that focus on contOpen-source communities that focus on content rely squarely on the contributions of invisible strangers in cyberspace. How do such communities handle the problem of trusting that strangers have good intentions and adequate competence? This question is explored in relation to communities in which such trust is a vital issue: peer production of software {(FreeBSD} and Mozilla in particular) and encyclopaedia entries {(Wikipedia} in particular). In the context of open-source software, it is argued that trust was inferred from an underlying ‘hacker ethic’, which already existed. The Wikipedian project, by contrast, had to create an appropriate ethic along the way. In the interim, the assumption simply had to be that potential contributors were trustworthy; they were granted ‘substantial trust’. Subsequently, projects from both communities introduced rules and regulations which partly substituted for the need to perceive contributors as trustworthy. They faced a design choice in the continuum between a high-discretion design (granting a large amount of trust to contributors) and a low-discretion design (leaving only a small amount of trust to contributors). It is found that open-source designs for software and encyclopaedias are likely to converge in the future towards a mid-level of discretion. In such a design the anonymous user is no longer invested with unquestioning trust. longer invested with unquestioning trust.
Added by wikilit teamAdded on initial load +
Collected data time dimensionN/A +
ConclusionOpen-source communities rely squarely on tOpen-source communities rely squarely on the contributions

of mostly anonymous strangers in cyberspace, so a central concern, whenever these focus on ever-evolving content, such as software modules or encyclopaedic entries, is the problem of whether and to what extent such volunteers can be trusted to contribute in good faith and in a competent fashion. It has been argued that such communities do indeed have a whole array of mechanisms at their disposal to handle this matter of trust: Table 1 lists the mechanisms involved. When rules and regulations are still few and far between, the full weight falls on the processes of inferring or assuming the presence of trust. Some assurances may be generated by an underlying shared culture. When hackers started to use the Internet to develop OSS, they had been cooperating with each other for decades in ‘real life’ and developing a shared ‘hacker ethic’. On the Internet they now bumped into a younger generation of new hackers. If Mizrach (1997) is right, their ‘new hacker ethic’ shared a considerable overlap with the old one. As a result the potential open source audience as a whole was tied to some kind of hacker ethic and their trustworthiness would appear to be guaranteed. The Wikipedia experiment was not so fortunate: no ‘encyclopaedic ethic’ existed at the outset, so when vandalism and harassment started to emerge— revealing that not all anonymous contributors could merely be assumed to be trustworthy—the community hurriedly embarked on a campaign to educate actual and potential Wikipedians. In so far as contributors come to feel bound by this emerging ‘Wikiquette’, they can surely be trusted to further the encyclopaedic cause. In general, inference of trustworthiness preferably relies on solid signs from the particular trustee involved. Opensource communities, however, depend on anonymous strangers who usually do not reliably signal anything and are merely represented by the IP addresses assigned to them. Such communities can therefore only try to gauge the existence of a climate of trust among their potential contributors in general and infer in a weak sense that they probably can be trusted. Due to the veil cast by the Internet, open-source communities are condemned to forming probability estimates of trustworthiness in place of certainties. Note, though, that another, more robust option for inferring trust is becoming available. Lately, within OSS circles the hacker ethic is seen to be eroding. In response some projects subject role occupants to stricter screening, especially when granting commit privileges is at stake. Solid proof of one’s allegiance to the hacker ethic is required. ‘Strong’ inference of trust, therefore, may replace ‘weak’ inference of trust. As long as an appropriate kind of ethic seems to be lacking within the community, assuming trust is the prominent mechanism, by default. Wikipedia has been shown to be a case in point. Posting encyclopaedic entries on the Internet as a wiki is an appeal to fellow Wikipedians to show their editing capabilities. It is a sign of ‘substantial’ trust (in the sense of McGeer 2008). This theorizing about the ‘normative pressure’ emanating from opening up content would seem to confirm the conjecture that, especially in cyberspace, assuming trust is an important mechanism for creating trust in the first place (cf. de Laat 2005). In this vein both Wikipedia and diaristic blogs revealing personal intimacies (as analyzed in de Laat 2008) seem to rely boldly on the mechanism that trust can be produced ex post. Sooner or later open-source communities start to introduce rules and regulations to manage the complexities involved. The governance tools distinguished above are modularization, formalization, division of roles and decision making. While the first two parameters would seem to meet with universal approval among participants, the last two need to be introduced with care. An important choice has to do with the amount of discretion granted to collaborators. In a high-discretion design, role division is minimal and decision making decentralized; as a result, the trust granted remains high. A low-discretion design, with elaborate division of roles and centralized decision making, leaves little discretion to collaborators; as a result, trust is to a large extent substituted. This third mechanism of substitution of trust was first explored with reference to OSS projects. A typical division of roles consists of contributors, committers and maintainers (or module owners). I have shown how FreeBSD tends towards a higher discretion design, while Mozilla tends towards a design of lower discretion. Similarly, the encyclopaedic Wikipedia project has become the subject of design. The fact is that their contributors retain a broad measure of discretion: anonymous users may still edit entries of their choice (and may create new entries after registration). Rising vandalism has not been combated by reducing the discretion of ordinary users, but by introducing ‘administrators’ who are granted powers to block particular users and freeze articles involved in edit wars. As a result, users stay at the helm in Wikipedia. Nevertheless, rampant vandalism has led to mounting pressures within Wikipedia for another measure: review of all changes (edits) with regard to vandalism. This system is undoubtedly a step that will reduce the discretion of ordinary users. It leads to the introduction of levels of trustworthiness, measured by one’s past performance within Wikipedia. Interestingly, this is not uniformly appreciated across language versions. It has already been introduced in the German Wikipedia (and some others), where the majority applauds the rules as a contribution to the fight against vandalism. Their English-speaking fellow Wikipedians, however, vehemently resist its introduction as an encroachment on their editing rights. To them, this is the onset of bureaucratic control. This finding suggests that intercultural perceptions of open-source design may differ considerably. This would seem to be an interesting research field, ripe for exploration: how essentially the same virtual organization is perceived, evaluated and designed across different cultural domains. As a logical extension, plans are afoot within Wikipedia to appoint ‘super-reviewers’ who are to check articles for their quality. If and when such a system of review plus super-review may become standard practice, the user will no longer reign supreme in crafting Wikipedian entries. Such a development would imply two remarkable processes of ‘structural convergence’: on the one hand, Wales’ encyclopaedia would then be governed in a similar fashion to comparable—but smaller—virtual encyclopaedic undertakings like Citizendium, h2g2 and Knol. These projects also apply review procedures to guarantee quality. On the other hand, generally speaking, roles within online open encyclopaedic projects would then resemble those that are usually discerned in OSS. The open-source communities for producing encyclopaedias on the one hand and software on the other would be managed in similar ways. The communities involved would seem to be concurring on the verdict that open-content production cannot do without a process of moderation. For all of them, unquestioning

trust in users has proved to be an unworkable assumption
has proved to be an unworkable assumption
Data sourceN/A +
Doi10.1007/s10676-010-9230-x +
Google scholar urlhttp://scholar.google.com/scholar?ie=UTF-8&q=%22How%2Bcan%2Bcontributors%2Bto%2Bopen-source%2Bcommunities%2Bbe%2Btrusted%3F%2BOn%2Bthe%2Bassumption%2C%2Binference%2C%2Band%2Bsubstitution%2Bof%2Btrust%22 +
Has authorPaul B. de Laat +
Has domainPhilosophy and ethics + and Information systems +
Has topicEthics + and Other collaboration topics +
Issue4 +
Peer reviewedYes +
Publication typeJournal article +
Published inEthics and Information Technology +
Research designConceptual +
Research questionsIn all a three-fold dynamics of trust emerIn all a three-fold dynamics of trust emerges: of

assumption, inference, and substitution of trust. After a brief survey of open-source communities in general, arguments are presented, explaining why trust is a central issue in the functioning of two particular examples: software and encyclopaedias. The remainder of the paper is then devoted to examining how this dynamics of trust unfolds within the two communities. trust

unfolds within the two communities.
Revid10,806 +
TheoriesTo me, her theorizing about the assumptionTo me, her theorizing about the assumption of trust

seems to be the more plausible avenue in the case of Wikipedia. By opening up their entries to immediate modification, contributors appeal to the encyclopaedic capabilities of unknown others ‘out there’. This theorizing about the ‘normative pressure’ emanating from opening up content would seem to confirm the conjecture that, especially in cyberspace, assuming trust is an important mechanism for creating trust in the first place (cf. de Laat 2005). In this vein both Wikipedia and diaristic blogs revealing personal intimacies (as analyzed in de Laat 2008) seem to rely boldly on the mechanism that trust can be produced ex post.hanism that trust can be

produced ex post.
Theory typeAnalysis +
TitleHow can contributors to open-source communities be trusted? On the assumption, inference, and substitution of trust
Unit of analysisN/A +
Urlhttp://www.springerlink.com/content/352t844j94386413/ +
Volume12 +
Wikipedia coverageCase +
Wikipedia data extractionN/A +
Wikipedia languageEnglish + and German +
Wikipedia page typeN/A +
Year2010 +