Repute

From eg

Jump to: navigation, search

See Google page rank, eBay rank, whuffie and Wikipedia edit rank for a list of well-known examples of repute affecting transactions on the Internet. A more robust form of repute is in academic credential and credit ratings.

The founder of the whuffie project claims that while "attention is traded just fine with Google AdSense, there's full market for attention of people watching your web page. But for the page itself, the rank of the page itself, it is not the market that decides if it will be first in the search results. It is the reputation of the page, and it is computed with a very strange and top secret formula in google's case, but the general idea is known." The search engine optimization methods, part of the web trade, track these formulae.

Repute and attention are closely related but they aren't identical. Listening to one piece of information over another, or listening to one MP3 not another, is more likely with higher repute, and obviously displaces the time/attention of the user - the real scarce commodity - the only thing that's truly irreplaceable. In other words, attention or free time displaced by attention is the actual resource, while repute merely helps determine its allocation, as do other factors like where/when/who the repute is assessed at/on/by.

Repute acts to speed some transactions or make some possible, but it's hard to say how. A capital asset model is required to identify potential substitutions and capital rivalries. For instance, better design or process usually results in less waste and use of resources, so instructional capital substitutes for natural capital. And, people do favours for each other that other people would charge cash for, so social capital substitutes for financial. Repute affects whether the design is actually tested or whether people risk doing favours for each other. Where good designs go untested or a general lack of cooperation prevents a group from succeeding, repute is likely involved. See rival good for an explanation.

Contents

[edit] Bad idea

Online systems that attempt to design reputation systems inevitably fall to gaming. Arguments against them have been advanced by trolls, many of whom define themselves by their defiance of repute, considering it factionally-defined.

[edit] Clay Shirky

"As David Weinberger noted in his talk The Unspoken of Groups, clarity is violence in social settings. You don’t get 1789 without living through 1788; successful constitutions, which necessarily create clarity, are typically ratified only after a group has come to a degree of informal cohesion, and is thus able to absorb some of the violence of clarity, in order to get its benefits. The desire to participate in a system that constrains freedom of action in support of group goals typically requires that the participants have at least seen, and possibly lived through, the difficulties of unfettered systems, while at the same time building up their sense of membership or shared goals in the group as a whole. Otherwise, adoption of a system whose goal is precisely to constrain its participants can seem too onerous to be worthwhile. (Again, contrast the US Constitution with the Articles of Confederation.)

Most current reputation systems have been fit to their situation only after that situation has moved from theoretical to actual; both eBay and Slashdot moved from a high degree of uncertainty to largely stable systems after a period of early experimentation. Perhaps surprisingly, this has not committed them to continual redesign. In those cases, systems designed after launch, but early in the process of user adoption, have survived to this day with only relatively minor subsequent adjustments.

Digg is the important counter-example, the most successful service to date to design a reputation system in advance. Digg differs from the community patent review process in that the designers of Digg had an enormous amount of prior art directly in its domain (Slashdot, Kuro5hin, Metafilter, et al), and still ended up with serious re-design issues. More speculatively, Digg seems to have suffered more from both system gaming and public concern over its methods, possibly because the lack of organic growth of its methods prevented it from becoming legitimized over time in the eyes of its users. Instead, they were asked to take it or leave it (never a choice users have been know to relish.

Though more reputation design work may become Digg-like over time, in that designers can launch with systems more complete than eBay or Slashdot did, the ability to survey significantly similar prior art, and the ability to adopt a fairly high-handed attitude towards users who dislike the service, are not luxuries the community patent review process currently enjoys."- [1]

"Consider the costs of designing a reputation system in advance. In addition to the well-known problems of feature-creep (“Let’s make it possible to rank reputation rankings!”) and Theory of Everything technologies (“Let’s make it Semantic Web-compliant!”), reputation systems create an astonishing perimeter defense problem. The number of possible threats you can imagine in advance is typically much larger than the number that manifest themselves in functioning communities. Even worse, however large the list of imagined threats, it will not be complete. Social systems are degenerate, which is to say that there are multiple alternate paths to similar goals — someone who wants to act out and is thwarted along one path can readily find others.

As you will not know which of these ills you will face, the perimeter you will end up defending will be very large and, critically, hard to maintain. The likeliest outcome from such an a priori design effort is inertness; a system designed in advance to prevent all negative behavior will typically have as a side effect deflecting almost all behavior, period, as users simply turn away from adoption."

"Working social systems are both complex and homeostatic; as a result, any given strategy for mediating social relations can only be analyzed in the context of the other strategies in use, including strategies adopted by the users themselves. Since the user strategies cannot, by definition, be perfectly predicted in advance, and since the only ungameable social system is the one that doesn’t ship, every social system will have some weakness. A system designed in advance is likely to be overdefended while still having a serious weaknesses unknown the designer, because the discovery and exploitation of that class of weakness can only occur in working, which is to say user-populated, systems. (As with many observations about the design of social systems, these are precedents first illustrated in Lessons from Lucasfilm’s Habitat, in the sections “Don’t Trust Anybody” and “Detailed Central Planning Is Impossible, Don’t Even Try”.)

The worst outcome of such a system would be collapse (the Communitree scenario), but even the best outcome would still require post hoc design to fix the system with regard to observed user behavior. You could save effort while improving the possibility of success by letting yourself not know what you don’t know, and then learning as you go."

[edit] Craig Hubley

"Repute is social, and necessarily resides in the hearts and minds of many people, who do not necessarily cooperate on much. While one could imagine and propose voluntary factions of people who explicitly agreed on some algorithm, and worked to improve it, it's quite difficult to imagine them making any but the most trivial decisions with it if there were substantial technical barriers to improving it or any credentials were required to critique and repair it. And, by, definition, such barriers do exist in any system that attempts to reduce repute to rank and certainly any made of software. Paul Alder observes that social capital both creates and follows hierarchy, seemingly out of necessity. When feudalism was succeeded by capitalism, financial capital became more important, too imporant, but social status and attention choices are and always were regulated by factors that are poorly understood. Accordingly, any attempt to build systems where formal ranking matters is an attempt at mind control. If it worked, it would compete with the dominant mind control system called money. There's only room for one system of bits that tries to tell people what to bond to." See body/bond/bits and individual/social/instructional for details.

"So it will never work. People who try to make it work are fantasizing about socialisms so perfect that they correctly anticipate the attention and desire of very many people, and rarely or never show anything undesirable or hide anything desirable. It is easy to understand why private online services want such algorithms, but it's not easy at all to see why users would want them, how they serve users, or how they can be used to do anything other than personalized recommendations, in which case it's not the repute, but the match quality that matters. Go look at dating services instead."

"The arguments to avoid imposing some measure that purports to represent repute is much like those to avoid access control: a focus on gaining repute perverts the social interactions towards pleasing the system, not another user. A GEVAD ensues in which the Panopticon of the system biases actions, creating a groupthink. If you think this is necessary and desirable, consider that no corporate intranet maintains such a single measure: corporate position creates rank and within that position, you may be accorded more stock options, but you aren't accorded more attention in meetings. At least, none that you don't earn every day." - Craig Hubley

[edit] Non-revert as repute

[edit] Revert currency

Using an indicator of repute, such as wiki edit reversion, as an indicator of repute may help "identify trouble-makers who revert material that is valid, or write material that is invalid. It may limit ad hominem revert, as the reversion of valid text would show up as negative" and thus affect the repute of the person reverting, i.e. censoring a useful/true edit.

However this only happens "as soon as it was restored" and it may not be, if the sysop vandal point of view prevails, it may "reinforce a cabal, clique or sysop power structure... who will ensure that the text does not return, valid or not. This is analogous to central bank currency controls in a global economy, which can disempower local cliques but strongly empower a global one" as George Soros and the ICLEI's 2005 resolution calling for change to banking reserve rules to deal with climate change and global debt suggest.

All that said, a global clique is at least difficult to keep invisible and its dealings may ultimately be transparent. If it can remain so, any attempt to quantify repute may be "much preferable to hierarchy and authority schemes with no accountability, and no disincentive for sysop vandalism whatsoever. .. it is not clear how a currency would encourage good edits, or if sysop vandalism would reduce. For instance, usurper sysops can continue to do ad hominem revert and cause the currency of their enemies to go down, using their sysop power structure to reinforce itself but making it look as if they are censoring low-credibility people... Like real currency and credit systems, which are constantly manipulated in this way, by real power structure, lots of checks and balances are required." In other words, a reflective process might develop out of the initially reactive model.

There are many references to problems and opportunities using non-revert as repute in a large public wiki. The revert is at least objective, as you can say flatly whether some text appears in the current version of the article or not. However, this is also what makes it worth gaming the system. So transparency itself may be an issue.

[edit] Alternatives to formal repute

[edit] Shirky: metadata, instrumentation, attention

Shirky recommended, for Community Patent's launch only "metadata", implicit rank "instrumentation" and design "attention"

[edit] Hubley: in/as/to

Hubley generalizes Shirky's model to frame/match/mediate. He points out that "objective context information about where/when/who events take place doesn't just fit in normative namespaces, but are always mentioned informally and incidentally on many pages. They're indices, not just facts. Fixed factual information presented in any structured data frame tends to strongly bias the presentation of all other data and create a mindset; this restricts goal statements, and deliberately so, in order to properly support a narrow range of activities, each of which can be measured - see activity-based costing. Repute, like organization protocols or schedules, serves to coordinate, but it does so in a crude in/out exclusion model: either you're worthy of full attention, or you're worthy of none at all. The slashdot-style thresholds have done nothing at all to exclude GNAA and so on, so this demonstrably doesn't work, no matter how many geeks slave away trying to "fix it"."

"For all these reasons, you're best to abandon GEVAD and let user groups themselves design voluntary metrics, if they can, and create factions. These can decide who is up/down but they don't get to decide who's in or out from anyone else's perspective. If you fall far enough down in some faction's eyes, you'd be invisible, but then you'd be even freer to organize a revolt than you were when they were watching you. By no means is your participation at stake in any but a clique, which you don't have to join. It should also be impossible for global repute to over-ride local matching. If an algorithm says two people should talk or work together or play certain roles, that must over-ride both persons' notion of repute, it's only those designed by the faction that would be cognizant of it. In very extreme cases you'd have to mediate or adjudicate." - Craig Hubley

Because things are done "in" frames, people cast in roles play "as" those roles, and it requires explicit referral "to" a third party who mediates, he calls this in/as/to.

[edit] ECG distrust

The ECG distrust metric also avoids positive repute, but it embraces negative repute.

Retrieved from "http://let.sysops.be/wiki/Repute"
Personal tools