Currently watching the #COMOatScale stream and thinking it's providing an unusual amount of access into to what's going on in the black box but not much re: policy insight
-
-
Try to form hardline robot-like rules around either spam or nudity and you're going to have just as hard of a time as you would with moderating harassment, it's only "straightforward" when you apply gut instincts based on exposure to many examples over time
Afficher cette discussion -
I'm also beginning to suspect that a lot of the ~ AI ~ stuff around moderation focuses on content instead of metadata??? When the history of anti-spam tech clearly demonstrates how rich metadata can be in screening for unwanted content????????????
Afficher cette discussion
Fin de la conversation
Nouvelle conversation -
-
-
Yes!! I’ve talked about this from my POV of being one the four men who made up the first Twitter anti spam team!
-
We don’t even talk about anti-harassment tooling with the maturity we do for anti-spam. With anti-spam, for instance, we had the common discussion point of “we’re not trying to end all spam; that’s impossible and ends up with bad acts. We’re trying to make it uneconomical”.
-
me: well my book isn't relevant anymore and i should rewrite it also me: *watching a panel* goddammit
-
You can find my tweets hoping for another one, too btw. Or just a chapter on Dogspotting Court
-
well i have the general outline of a new one, it's just a matter of finding the time for it
- Fin de la conversation
Nouvelle conversation -
Le chargement semble prendre du temps.
Twitter est peut-être en surcapacité ou rencontre momentanément un incident. Réessayez ou rendez-vous sur la page Twitter Status pour plus d'informations.