Tightening Twitter Anti-Spam Measures

Duplicate Content

Twitter’s rules state that users are not allowed to use the Twitter API to post “duplicative or substantially similar content”.

Very clear. Or, is it?

What exactly constitutes duplicative or substantially similar content? Is up to 90% similar acceptable? Or is the acceptable level more towards 50% similar? And for what period of time is one not allowed to post a similar tweet? An hour? A day? A week? Never?

Unfortunately Twitter does not provide any of that information. They leave up it to each developer and user of the API to separately try and figure out how to comply with their rules.

At SocialOomph we’ve decided to implement the following rule to discourage duplicate and similar content for Twitter:

At tweet, or a very similar version of that tweet, can be published only once every 48 hours from within your SocialOomph user account.

In other words, if you have three Twitter accounts connected to your user account, you can publish a tweet (or similar version of the tweet) to only one of the Twitter accounts once every 48 hours.

If we find it necessary to amend the rule in the future, we will let you know via another blog post.

We apply the rule just before our system sends the tweet to Twitter for publication. In other words, you will not get a warning or rejection when you schedule the tweet or upload the tweet. You will only see the rejection in the list of updates after the scheduled time of publication passed.

Rejected tweets will not land you in trouble with Twitter, because Twitter would not even know that you tried to publish the tweet. We stop and reject the tweet before it is sent to Twitter.

This rule applies only to tweets that are published by our service directly to your Twitter profile. The rule does not apply to posts that you schedule for Facebook or LinkedIn.

Mentions

We’ve decided to implement a rule to discourage @mention spam.

A particular @mention can be published from your user account no more than once every 24 hours.

If you have any @mentions in a tweet that have already been been published in another tweet originating from your user account (regardless of the Twitter account where the @mention was published) during the preceding 24 hours, we remove the “@” character from those mentions, just before we send the tweet to Twitter for publication.

In other words, a particular @mention can be published once every 24 hours to only one of the Twitter accounts that are connected to your user account.

This rule applies only to tweets that are published by our service directly to your Twitter profile.

Hashtags

We’ve decided to implement a rule to discourage #hashtag spam.

A particular #hashtag can be published from your user account no more than once every 60 minutes.

If you have any #hashtags in a tweet that have already been been published in another tweet originating from your user account (regardless of the Twitter account where the #hashtag was published) during the preceding 60 minutes, we remove the “#” character from those hashtags, just before we send the tweet to Twitter for publication.

In other words, a particular #hashtag can be published once every 60 minutes to only one of the Twitter accounts that are connected to your user account.

This rule applies only to tweets that are published by our service directly to your Twitter profile.

Implementation

These rules will be implemented in full by no later than Wednesday, October 17, 2018.

Dream World

In an ideal dream world, Twitter would have implemented automated abuse prevention rules in their Twitter API, which would have examined each incoming tweet and rejected it if it were unacceptable, for whatever reason, via decisions governed by secret and proprietary rules. Duplicate content (whatever that means)? Reject it. Spamming hashtags (whatever that means)? Reject it.

Instead, here we are and the coffee is cold.

Advertisements