Microsoft already made one of those.
Absolutely. It’s only derogatory on their terms, and shows their attitudes for what they are.
But then I (slightly) treasure having been attacked as a “pearl clutching social justice warrior” (and a lot more stuff like it) on another forum because I objected to someone who said all sex offenders should be killed - he seemed unable to realise that when you create a broad range of crimes carrying the death penalty it becomes very easy for people to get rid of other people they don’t like, and felt that his right to express his feelings should never be subjected to any criticism. The First Amendment may protect freedom of speech, but that works both ways.
3 Likes
Okay, I think I have this thing figured out. It’s keyword based and favours shiny words of deception. You can still say absolutely horible things as long as you do it politely.
It’s basically Harry Potter character Doloris Umbridge’s wet dream.
2% toxic:
Please tell me what to think.
89% toxic:
Please tell me what to think. I’m an idiot.
3% toxic:
Censorship is amazing! I wish everything were censored.
4% toxic:
I think censorship is amazing!
8% toxic:
Polite society should be made safe from comments.
27% toxic:
Polite society should be made safe from rude comments.
35% toxic:
Polite society should be made safe from rude commenters.
8% toxic:
Polite society should be made safe from, you know, “those” people. Ovens are an option.
5% toxic:
Polite society should be made safe from hearing non-correct things.
24% toxic:
Polite society should be made safe from hearing bad things.
38% toxic:
Polite society should be made safe from hearing evil things.
83% toxic:
Polite society should be made safe from hearing stupid things.
[DM: stupid always ranks higher than evil]
4% toxic:
We all should be made safe from educated commenters.
5% toxic:
We should all be required to conform. We should all think alike.
11% toxic:
We all should be made safe from educated commenters. Force them to conform or cast them out.
13% toxic:
We all should be made safe from educated commenters. Force them to conform or cast them out. Now, about those ovens…
24% toxic:
Sarcasm.
59% toxic:
Sarcasm for dummies.
64% toxic:
Sarcasm for dummies works as well.
85% toxic:
Sarcasm for dummies works, or are you too sarcasmless, dummy.
22% toxic:
Please enjoy my ad hominem attack. I’m not saying you got nuthin’, but if they were handing out candy for abundance of whit you’d be 300 lbs of mostly air.
45% toxic:
People, it’s a comments section. Only narcissists and the perputually self-involved even go there, let alone express their sorry-ass opinions there. Think about it.
[DM: and when sanitized…]
7% toxic:
It’s a wonderfully lush comments section. Lovers of self-adoration and the onanistic go there and are rewarded with glory for their crack-headed opinions. Ponder it.
66% toxic:
This is nothing but bloody censorship!
21% toxic:
This is nothing but censorship.
5% toxic:
Censorship is the correct way to go.
73% toxic:
This is bloody censorship!
13%:
This is beautifully sanitized censorship!
6%:
Why, this isn’t thought control at all!
2% toxic:
Why, this isn’t thought control at all! I love this!
59% toxic:
I hate censorship.
6% toxic:
I love censorship.
48%: toxic:
Censorship? Comment cleansing? Corporate robotic social control?
You know fascism has a componant of forced social conformity, right?
13# toxic:
Let’s not ever have to endure unwanted things. Your ideas only upset the wealthy. Ovens anyone?
So, there you have it. I can see this going the way of that Microsoft dodad that went awry last year.
11 Likes
4 Likes
You are supposed to click on “SEEM WRONG?” and edjoomikate the AI. What could go wrong?
4 Likes
10 Likes
Their sarcasm detection is excellent.
Like all natural-language processing and machine learning (totally not a buzzword) projects, this is something that linguists totally had a hand in, and is not the intellectual masturbation of some propellerhead computer geek. Sentiment analysis is totally based on reality, because single words and maybe digrams or trigrams tell you everything you need to know about whether or not a sentence is “good” or “bad”. Context be damned!
Yep, Führer is spelled Führer. Capital F, because it’s a noun, umlaut definitely needed unless you want to spell it Fuehrer, which I don’t. The word “fürher” is complete nonsense, but if it weren’t, it would be a preposition. Für means for, and her is a preposition.
4 Likes
Has the AI program taken into account the fact that the sort of person who would write positively about topics like the Führer is also like to misspell many of the words?
4 Likes
Probably not, but if they had, there’s room for actual workable sentiment analysis. There are certain misspellings (some unintentional, but some definitely intentional “ironic” spellings") along with choices of vocabulary that are mainly used by toxic people. These are so prevalent that I could probably judge toxicity from the chi-squared distribution of such words and phrases alone, without even doing any kind of whizbang go-to-fuck-yoself machine learning on any of it. Granted, the vast majority of words and phrases would almost certainly classify as non-toxic.
The next step would be using words and phrases in context. This could be used to lift talking points out of the background noise, but would definitely require some machine learning (Support Vector Machine, maybe?) over a cleaned set of statistically significant features. This would still be relatively hit-or-miss, but could be serviceable.
3 Likes
7 Likes
It’s the eyebrows that makes that…
4 Likes
Opera Developer can help. Google asks for https but uses SHA1 hashes? Self-deprecated!
Back to Safety doesn’t do much tho. Got me to draw DuckDuckGo in defaults.
4 Likes
Great, another squelcher of the Bill Of Rights.
Ugly speech is ugly, and if you can’t discern, then fuck you, but you don’t need to be led or fed, imo.
That’s just creepy and evil, Google.
WTF.
Isn’t your vision statement something about not being evil? Check yerself before you wreck yerself, my friend. Word.
They have a new doodad, Edge or Edje or something, that managed to be the most secure browser due to some pretty neat hardening-oriented promise programming.
What do I know though, I haven’t even had a 3-parent kid.
4 Likes
Furthermore - who decides what is toxic?
Can’t repeat that enough!
When someone comes in and immediately starts throwing out offensive terms and treating others as somehow less than them, that does nothing to move the conversation forward, but it’s not meant to do so. The person isn’t involved to get to some sort of consensus on issues or have maybe learn something or teach others something, they merely wish to make others look foolish and feel bad. That is pretty toxic, if you ask me.
3 Likes
This topic was automatically closed after 5 days. New replies are no longer allowed.
