Is content Moderation a Dead End?
https://www.ben-evans.com/benedictevans/2021/4/13/is-content-moderation-a-dead-end
Facebook could stop 10B impressions from “repeat misinformers”, but doesn't
Thanks but I’d rather decide what’s misinformation for myself.
https://news.ycombinator.com/item?id=26794394
#disinformation #algorithms #propaganda #AVAAZ #AlgorithmicPropaganda #ComputationalPropaganda #ContentModeration #advertising #virality #trust #WeaponisedClickbait #FakeNews #ShallowDismissals
On content moderation, the metric that seems to be getting picked up is #Prevalence (Facebook) or #ViewRate (YouTube/Google), which looks not just at the number of items posted, but the times each are viewed. This is beginning to approach a useful metric, but still poses several problems.
It's easy to focus on simple numbers or characteristics. These almost always provide an oversimplified and incomplete view. Noting how many violative / disinformational pieces of content are posted without accounting for their presentation within members' streams or search results is incomplete.
Computing items times impressions ... is much better. It also provides a basis for determining the moderation load required vs. degree of access granted, given distributions of prevalence which tend strongly to follow a power law distribution. (More below.)
#facebook #google #youtube #moderation #contentModeration
2/
Adding in a risk component, a calculation of a potential for harm for a given piece, seems the next logical stage in the process. This prioritises content for moderation.
From this, and making a broad incursion into potentially Orwellian territory, is identifying risks by individuals or groups (detected clusters or networks) which might then affect moderation and distribution (prevalence/view rate) scores or targets.
Given that social media are, well, social, with groups tending to have a high level of coherence, behaviour towards standards might be determined and behaivours including failure-to-flag or amplifying offending content being further considered. Effectively a trust metric for posting and amplification.
#contentModeration #moderation #youtube #google #ViewRate #Prevalence
3/
Among the interesting suggestions I"ve seen recently (and have encountered / thought of myself previously before) is this "feudal vouching system":
highly engaged community members, vouch for others, who again vouch for others. If people in this "Dharma-tree" accumulate problematic behaviour points, the structure at first, bubbles better behaved members to the top. If the bad behaviour continues, by single members or sub-groups, the higher echelon of members will cut that branch loose, or loose their own social standing.
https://news.ycombinator.com/item?id=26797017
Rather than use high engagement as a basis for vouching, arbitrarily selected communities, perhaps of about 50-150 active participants (posting or moderating) might be better. Think Mastodon, but in order to be federated, good posts must be "voted off the island", in the good sense.
I've been applying a somewhat similar notion to collecting and managing reading material and suggestions, what I call #BOTI, or Best of the Interval. It's a round-robin style system, where I compile a list of references over a period of time (monthly to annually seems to be most appropriate for me, though hour / day / week / month / year / decade / ... could be applied). At the end of an interval, some limited number of items is carried forward.
This is one way of addressing the "firehose of content" nature of information, recognising that in any given time period, you only have so much personal bandwidth to dedicate.
With the federated model, content federating is itself subject to assessment, and is effectively re-vetted. Note that different communities might favour different content: cooking, kittens, Kabbalah, canoodling, cypherpunk, core meltdowns, classic cars, concerthalls. Vetted / re-vetted streams themselves might be of interest.
Both community- and time-based elements of this could get interesting.
4/
@woozle Miranda's Knitting & Tea Circles were my attempt.
Community formation is a major hurdle.
A major problem of growth-related risks is that low-scale risks / problems and high-scale risks / problems are all but entirely distinct sets.
Initiatives which launch with eyes on large-scale problems tend to fail before attaining that scale. (In large part because virtually all initiatives fail, but also because they're not focused on the initial problem of growing first.)
@woozle I'm not sure the problem is, strictly speaking, a technical one.
Or to be more specific, I'm pretty certain it is NOT a technical / software problem.
@dredmorbius how do you deal with actors like this, in communities that enable them: https://woof.group/@aphyr/106053271555436741
as in: what tools do people in a system get to not get swamped with stuff?
Attached: 1 image every time friendlysock posts ambiguous things like this on lobste.rs I stare at it *long* and hard and think about the last seven years of their "just asking questions about gender" posts. https://lobste.rs/s/ccseq4/activitypub_secret_weapon_fediverse#c_pntd6p
@meena What specifically is the issue here?
(I see several possible issues. I'm not deeply versed or vested in most. I am probably not perceiving this precisely as you do.)
I'd argue that code-of-conduct guidelines inherently address emergent phenomena and behaviours. If specific activities are consistently associated with problematic situations ... well, something's up.
(Be very careful as to how you interpret that as the assessment can cut many ways.)
@meena The simple and direct response to your question "what tools do people in a system get":
Keep in mind that these are exceedingly powerful and blunt tools, and (as my earlier response noted) cut in multiple directions and dimensions.
There is the question of incommensurate values, beliefs, or practices. One set I call "chirality", in which there's no prima faci basis for having a preference of one modality over another, but once that preference has emerged, honouring it is vital.
Leptose vs. dextrose, DNA and RNA direction, left- vs. right-handed traffic flow. None has an inherent advantage over the other. Once one is dominant, you're best off cooperating with that choice.
Harder is the expressed values question. Consider the question of beef or pork in a community feast featuring both muslims and hindus. Either choice represents an utterly unacceptable circumstance to at least one group. There is no immediate medical basis for exclusion (long-term consequences are discounted here), and cultures thrive on either or both food. But as a matter of cultural identity and belief these are irreconcilable options. Any joint activity would have to exclude both options.
These are relatively simple exemplars, I'm trying to avoid more fraught cases (several mentioned in the screencap and thread you link). Where there is an active cultural war waged on and through beliefs, the picture rapidly gets more complicated.
The Internet erases space. Space provides boundaries, barriers, and regimes with distinct and conflicting practices.
(Again: the who and what of your question are unclear.)
Doc Edward Morbius ⭕
Woozle Hypertwin
Mina