4 ms·

Having used the app, if this is the future of discussion then I don't want it. There are innumerable ways to trip concealed content filters that then deceptivel

Having used the app, if this is the future of discussion then I don't want it. There are innumerable ways to trip concealed content filters that then deceptively remove posts in a way that makes it look like community action rather than top-down content moderation. Some of these filters are great for preventing liability for their app, but make the app worthless for useful discussion. I can't wait until they implement their natural language filters so that mentioning suicide or a gun generates an automated community check or police visit, or it's just impossible to violate opaque ideological norms set by app creators in a high-tech version of manufactured consent. Using modern social apps for me is starting to feel like navigating a Kafka-esque bureaucracy. I don't know the right thing to do to be heard, I don't know the wrong thing to do to not bring down sanctions on me, and I often struggle to navigate the app because I don't know the invisible use patterns behind unlabeled buttons and hidden swipe menus, dot-replies, etc. Am I shadow-banned? Does the app hide my replies until I pass some invisible commenting threshold? How do I page through replies? Are my posts being seen by a few people or everyone, and by what criteria? All invisible.

danso 11y ago
Yeah, it's the classic tension between anonymity (also, security) and utility...this morning, a prominent part of Stanford's campus was closed off due to a memorial service for Dave Goldberg. Someone asked what was going on and someone attempted to reply...I don't know what the replier posted, only that it apparently got deleted, so the replier just settled for posting "whatever, just google surveymonkey" I figure there's some kind of NLP entity recognition for names/proper nouns...and perhaps the mechanism/policy is set to be more safe-than-sorry (in favor of deleting content). I just can't see how this will translate into a valuable community, though. I go to YikYak whenever I want to know the kind of things that are going on but aren't covered in a news source, but factual/substantive discussion is effectively discouraged by the content policy. How long until the general userbase just gets bored of seeing ripped-off memes and shower thoughts? (on the other hand, maybe it doesn't need to depend on a userbase that's not in college...every year, YikYak can count on getting a fresh userbase to replace the graduating seniors who end up quitting the app)
- A_COMPUTER 11y ago
  Yes, there is an automated system in YikYak that artificially applies downvotes on a timer if you make a post with an anglophone name in it, faking natural downvotes. I admit there's a tension there, but a large part of my complaint is that it's intentionally opaque to the point of actively trying to hide itself. Ethical, human-friendly moderation systems have at least some degree of transparency.

danso 11y ago

Yeah, it's the classic tension between anonymity (also, security) and utility...this morning, a prominent part of Stanford's campus was closed off due to a memorial service for Dave Goldberg. Someone asked what was going on and someone attempted to reply...I don't know what the replier posted, only that it apparently got deleted, so the replier just settled for posting "whatever, just google surveymonkey" I figure there's some kind of NLP entity recognition for names/proper nouns...and perhaps the mechanism/policy is set to be more safe-than-sorry (in favor of deleting content). I just can't see how this will translate into a valuable community, though. I go to YikYak whenever I want to know the kind of things that are going on but aren't covered in a news source, but factual/substantive discussion is effectively discouraged by the content policy. How long until the general userbase just gets bored of seeing ripped-off memes and shower thoughts? (on the other hand, maybe it doesn't need to depend on a userbase that's not in college...every year, YikYak can count on getting a fresh userbase to replace the graduating seniors who end up quitting the app)

A_COMPUTER 11y ago
Yes, there is an automated system in YikYak that artificially applies downvotes on a timer if you make a post with an anglophone name in it, faking natural downvotes. I admit there's a tension there, but a large part of my complaint is that it's intentionally opaque to the point of actively trying to hide itself. Ethical, human-friendly moderation systems have at least some degree of transparency.