Hey queer twitter: we don't actually need to extrapolate from a single data point.
-
-
I'm pretty sure the system that's at work here is: [Term] is listed in a [MAY BE SLUR] chart, rather than a [LIKELY SLUR] or [MUST BE SLUR] chart; Tweet containing term gets mass-reported; Account gets throttled for [Time*anti-whuffie]
-
If I were writing an automated system to review text, I would train only the [MUST BE SLUR] fallthru on umlauted/diacriticed variants, on the justification that atypical quantities of character set use, can and is already part of the metadata flag process, & demonstrates "guilt"
-
I would also have the system learn to recognise [LIKELY SLUR] from ongoing evidence of attempts to avoid a presumed-naive static realtime slur-filtering/flagging system
End of conversation
New conversation -
-
-
"programmers have in fact heard of diacritics" That doesn't mean they *care* about them :) (Personally, I recommend U+211A for aesthetic reasons ;)
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
i like quëër tho, more metal that way
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Tøtesdisguised
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
What about “kweer”?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.