4 ms·

The worst thing is that the workflow to report spam has gotten a lot worse since the last 6 months. Now it takes 9 clicks 1. three dots 2. report spam 3. wel

by bko 4y ago

The worst thing is that the workflow to report spam has gotten a lot worse since the last 6 months. Now it takes 9 clicks 1. three dots 2. report spam 3. welcome message start report 4. who is this report for? myself, some specific group, everyone (??) 5. general info (attacked b/c of identity, harassed, spam, ... ) 6. how is he doing this? (Posting misleading or deceptive links, leading to scams, phishing, or other malicious links, +6 more similarly verbose) 7. yes continue 8. submit 9. done It used to be like 2 or 3 about 6 months ago but I remember after Musk started complaining they responded by somehow making it worse. Also most of the reports are centered around harassment or racist comments when 90% of the bad content on twitter most people deal with is straight up scams. There should just be a big this is spam button.

drtz 4y ago
My pessimistic view: making it too easy to report spam may highlight how much spam there actually is on Twitter. Something the company definitely does not want to do at the moment.
- gcmac 4y ago
  I wonder if their A/B test showed that this was somehow "solving" the "spam of spam reports" problem. It would be fairy naive but I could envision a world where they tested this out, saw fewer (but more detailed) spam reports and concluded that the prior distribution of spam reports was itself not reliable (e.g overcounted the problem). This is obviously self-serving but seems like it may at least be a "consistent" world view where they aren't totally cynical.
  - bko 4y ago
    It's just bad UX and confusing. Consider this question "Who is this report for?". Here are the options: 1. Myself 2. Someone else or a specific group of people -- This Tweet is directed at or mentions someone else or a specific group of people — like racial or religious groups. Everyone on Twitter 3. This Tweet isn’t targeting a specific person or group, but it affects everyone on Twitter — like misleading info or sensitive content. For the same Vitalik spam tweet you see everywhere, I'm guessing 3, but its kind of weird question Then "Everyone on Twitter is being ..." 1. Attacked because of their identity 2. Slurs, misgendering, racist or sexist stereotypes, encouraging others to harass, sending hateful imagery Harassed or intimidated with violence 3. Sexual harassment, group harassment, insults or name calling, posting private info, threatening to expose private info, violent event denial, violent threats, celebration of violent acts 4. Spammed 5. Posting malicious links, misusing hashtags, fake engagement, repetitive replies, Retweets, or Direct Messages. Shown content related to or encouraged to self-harm 6. Shown misleading info 7. Offered tips or currency — or encouraged to send them — in a way that’s deceptive or promotes or causes harm I guess spammed, but I'm pretty sure its a malicious link and misleading, so either 4 or 5 or 6 Then, "How is ⁦@... doing this?" 1. Posting misleading or deceptive links, leading to scams, phishing, or other malicious links 2. Misusing hashtags, such as unrelated hashtags and large number of hashtags 3. Sending a lot of aggressive, unwanted, repetitive or unrelated replies, Retweets, or Direct Messages 4. Fake engagement, such as aggressively Retweeting or buying and selling Likes, replies, or other Twitter features 5. Using multiple accounts to interact or coordinate with other people to manipulate accounts, Tweets, or other Twitter features 6. Following and then unfollowing large numbers of accounts so to inflate follower count 7. Something else I guess its a deceptive link, but I'm not sure because I didn't click on it. I also didn't click 5 on the previous step which was "posting malicious links", I clicked spammed. But he's also using repetitive or unrelated replies so maybe 3 as well? And almost certainly the person running the scam is using multiple accounts. That's a lot of words to read every time, and I'm pretty sure it has changed since I started reporting stuff regularly.
    - bombcar 4y ago
      These questions are the kind of questions you put up when you want to funnel some (or most) of the reports into the trash can. One thing Twitter certainly has to deal with is hordes of people reporting tweets, even if they're "ok" - I have to believe every single Trump tweet received hundreds if not thousands of reports immediately upon posting.
anvic 4y ago
How much do you get paid to do content moderation for Google? It may be a good side hustle if the pay is good.
hellomyguys 4y ago
There's a downside to making it easy to report content. You just end up with a bunch of useless reports.
alecco 4y ago
In the past year I reported many of these. The most common pattern was some WhatsApp number with funny characters. In most cases the message stays there. I can't believe it's not easy to filter those. Who uses 10 consecutive weird chars in a message of 15 chars? And most of those are numbers. I stopped reporting because clearly YouTube doesn't care. And there was a more clear case a couple of years ago with inappropriate comments using sexualized emojis. YouTube did nothing until the outrage got to the press. It's like they only focus in finding excuses to demonetize people leaning into wrongthink.
numpad0 4y ago
Because kids, genuinely autistic engineers, politically inclined, and scammers alike were using it as a super-block and were also scripting it. Say something “offensive” to the right person, and your account is frozen by the end of the day.

by bko 4y ago

drtz 4y ago

My pessimistic view: making it too easy to report spam may highlight how much spam there actually is on Twitter. Something the company definitely does not want to do at the moment.

gcmac 4y ago
I wonder if their A/B test showed that this was somehow "solving" the "spam of spam reports" problem. It would be fairy naive but I could envision a world where they tested this out, saw fewer (but more detailed) spam reports and concluded that the prior distribution of spam reports was itself not reliable (e.g overcounted the problem). This is obviously self-serving but seems like it may at least be a "consistent" world view where they aren't totally cynical.
- bko 4y ago
  It's just bad UX and confusing. Consider this question "Who is this report for?". Here are the options: 1. Myself 2. Someone else or a specific group of people -- This Tweet is directed at or mentions someone else or a specific group of people — like racial or religious groups. Everyone on Twitter 3. This Tweet isn’t targeting a specific person or group, but it affects everyone on Twitter — like misleading info or sensitive content. For the same Vitalik spam tweet you see everywhere, I'm guessing 3, but its kind of weird question Then "Everyone on Twitter is being ..." 1. Attacked because of their identity 2. Slurs, misgendering, racist or sexist stereotypes, encouraging others to harass, sending hateful imagery Harassed or intimidated with violence 3. Sexual harassment, group harassment, insults or name calling, posting private info, threatening to expose private info, violent event denial, violent threats, celebration of violent acts 4. Spammed 5. Posting malicious links, misusing hashtags, fake engagement, repetitive replies, Retweets, or Direct Messages. Shown content related to or encouraged to self-harm 6. Shown misleading info 7. Offered tips or currency — or encouraged to send them — in a way that’s deceptive or promotes or causes harm I guess spammed, but I'm pretty sure its a malicious link and misleading, so either 4 or 5 or 6 Then, "How is ⁦@... doing this?" 1. Posting misleading or deceptive links, leading to scams, phishing, or other malicious links 2. Misusing hashtags, such as unrelated hashtags and large number of hashtags 3. Sending a lot of aggressive, unwanted, repetitive or unrelated replies, Retweets, or Direct Messages 4. Fake engagement, such as aggressively Retweeting or buying and selling Likes, replies, or other Twitter features 5. Using multiple accounts to interact or coordinate with other people to manipulate accounts, Tweets, or other Twitter features 6. Following and then unfollowing large numbers of accounts so to inflate follower count 7. Something else I guess its a deceptive link, but I'm not sure because I didn't click on it. I also didn't click 5 on the previous step which was "posting malicious links", I clicked spammed. But he's also using repetitive or unrelated replies so maybe 3 as well? And almost certainly the person running the scam is using multiple accounts. That's a lot of words to read every time, and I'm pretty sure it has changed since I started reporting stuff regularly.
  - bombcar 4y ago
    These questions are the kind of questions you put up when you want to funnel some (or most) of the reports into the trash can. One thing Twitter certainly has to deal with is hordes of people reporting tweets, even if they're "ok" - I have to believe every single Trump tweet received hundreds if not thousands of reports immediately upon posting.

anvic 4y ago

How much do you get paid to do content moderation for Google? It may be a good side hustle if the pay is good.

hellomyguys 4y ago

There's a downside to making it easy to report content. You just end up with a bunch of useless reports.

alecco 4y ago

In the past year I reported many of these. The most common pattern was some WhatsApp number with funny characters. In most cases the message stays there. I can't believe it's not easy to filter those. Who uses 10 consecutive weird chars in a message of 15 chars? And most of those are numbers. I stopped reporting because clearly YouTube doesn't care. And there was a more clear case a couple of years ago with inappropriate comments using sexualized emojis. YouTube did nothing until the outrage got to the press. It's like they only focus in finding excuses to demonetize people leaning into wrongthink.

numpad0 4y ago

Because kids, genuinely autistic engineers, politically inclined, and scammers alike were using it as a super-block and were also scripting it. Say something “offensive” to the right person, and your account is frozen by the end of the day.