Help with our Twitter account [Update: we're back]

Our Twitter account, @clicky, was suspended for unknown reasons on June 24. Tomorrow will be 1 month since that happened.

UPDATE AUGUST 10 2020: After 5 appeals and 47 days, Twitter finally responded re: our account suspension and informed us that it was in fact to do with our search API usage, as we surmise at the very bottom of this post. So that feature is disabled for now. 😩 /UPDATE

Twitter is our main outbound communication method with our customers. We never send mass emails and only blog about major updates and features, so Twitter fills in the gap between those two perfectly for all minor things.

Before I write a novel, I want to ask you all – if any of you work at Twitter, or know someone who does… if you could get this post to the right person to help get our account reinstated, that would be amazing.

We’ve submitted 4 appeals and heard nothing back. I don’t know what their normal process is but I can’t imagine it’s been impacted too much from COVID since most of their employees can easily work from home, and from what I hear, they are in fact doing that.

It’s beyond frustrating to not know why. On the third appeal I submitted to them, just 7 days ago, I noticed that their “enforcement philosophy” page contained the following section:

When it comes to enforcing these rules, we are committed to being:
- Fair – we will enforce our rules impartially and consistently, considering the context involved.
- Informative – we will inform you about actions taken against your account and why.
- Responsive – you can appeal decisions that have impacted your account.
- Accountable – we will be transparent about actions we take to promote healthy public conversation, including by publicly reporting the metrics we are using to measure health and by publishing a regular transparency report around violations of our rules.

In my appeal, I pointed out to them their claims on this page of “informative” and “responsive” were blatantly false, since we were never informed why our account was suspended and at that point it had been 3 weeks so they were certainly not being “responsive” either. And then today (July 23) when I went to check again in prep for this post… THAT ENTIRE SECTION HAD BEEN DELETED! Did they seriously delete this section because I pointed out the falsehoods in it, BUT STILL HAVEN’T DONE ANYTHING ABOUT OUR ACCOUNT…??!?!?! WHAT!?

You see, the link above is to an archive.org copy of that page, from July 11… 12 days ago. Please compare to the current version. It’s 100% the same, except for the deletion of the section quoted above, and the addition of the word “newsworthy” further down. Those are the only differences!

Anyways… let’s go over ALL the Twitter rules to see what we could possibly have done:

Safety – No threatening violence, targeted harrassment, sharing sensitive media, facilitating transactions in illegal goods or services, etc. Nope, we’re good.
Privacy – You can’t share someone’s private information such as home address or phone number, or non-consensual nudity. Nope.
Authenticity –
- Platform manipulation and spam – Nope. We only post relevant updates on our business and service.
- Civic Integrity: You may not use Twitter’s services for the purpose of manipulating or interfering in elections or other civic processes. – Nope. Why would we talk about anything political on our business account when we make internet software.
- Impersonation: You may not impersonate individuals, groups, or organizations in a manner that is intended to or does mislead, confuse, or deceive others. – Nope. We pretend to be no one but Clicky.
- Synthetic and manipulated media: You may not deceptively share synthetic or manipulated media that are likely to cause harm. Nope. The only media we share is screenshots of Clicky, and the occasional animated GIF in response to customer tweets.
- Copyright and trademark: You may not violate others’ intellectual property rights, including copyright and trademark. Learn more about our trademark policy and copyright policy. Nope. We’ve owned the US trademark on “Clicky” as it relates to software for almost 10 years. Although our account was not “verified”, so it’s possible that if someone submitted some kind of trademark or copyright claim, Twitter may have been quick on the ban hammer if they didn’t spend time investigating.
- Third-party advertising in video content: You may not submit, post, or display any video content on or through our services that includes third-party advertising, such as pre-roll video ads or sponsorship graphics, without our prior consent. Nope. We’ve never posted a video on Twitter that I know of, but if we did, there certainly weren’t any ads in it.

So those are literally all the rules. The only one that could possibly apply to us would be the copyright/trademark rule, and that would at worst be a misunderstanding and a failure to investigate on Twitter’s behalf. As mentioned above, we own the US trademark on “Clicky” as it relates to software.

Twitter also has a page called “our range of enforcement options” that talks about what might happen if you break the rules:

Tweet-level enforcement. We take action at the Tweet level to ensure that we are not being overly harsh with an otherwise healthy account that made a mistake and violated our rules: Limiting Tweet visibility … Requiring Tweet removal … Hiding a violating Tweet while awaiting its removal … Nope. Never got a notice about any specific tweet.
Direct Message-level enforcement. Nope. Never used DMs on this account.
Account-level enforcement.
- Requiring media or profile edits: If an account’s profile or media content is not compliant with our policies, we may make it temporarily unavailable and require that the violator edit the media or information in their profile to come into compliance. Nope.
- Placing an account in read-only mode: If it seems like an otherwise healthy account is in the middle of an abusive episode, we might temporarily make their account read-only, limiting their ability to Tweet, Retweet, or Like content until calmer heads prevail. The person can read their timelines and will only be able to send Direct Messages to their followers. Nope.
- Verifying account ownership: To ensure that violators do not abuse the anonymity we offer and harass others on the platform, we may require the account owner to verify ownership with a phone number or email address. This also helps us identify violators who are operating multiple accounts for abusive purposes and take action on such accounts. Nope.
- Permanent suspension: This is our most severe enforcement action. Permanently suspending an account will remove it from global view, and the violator will not be allowed to create new accounts. When we permanently suspend an account, we notify people that they have been suspended for abuse violations, and explain which policy or policies they have violated and which content was in violation. Yes, I believe we are permanently suspended since it’s now day 30. However, as I mentioned previously, we were absolutely not notified of the suspension, or were we explained which policies we may have violated. Even if they refuse to reverse their decision, not knowing the “why” is what’s killing me inside.

There’s only one other thing I can think of, and that’s our usage of Twitter’s Search API. The Search API allows read access to public tweets. If we were using the write API to create tweets or DMs that broke the standard Twitter rules… sure, they could suspend your account. But we only use the search API to read public tweets that match keywords as entered by our customers. So the only “abuse” I know of here is too many API calls. But the developer agreement only mentions revoking API access if they don’t like your usage. There’s nothing about account suspensions. But in the interest of full transparency, in case it is relevant to our suspension, here’s how we use the search API:

Clicky has a feature that monitors Twitter for certain keywords you specify (typically related to your web site), and we build reports for you out of that data. We’ve had this feature since 2009. The API is throttled though. When you pass too many queries in too short a time, Twitter starts responding with an error and telling you how long to wait to try again. We’ve always respected these API limits as far as I remember, if not from the start then very shortly thereafter.

We have over a million sites, not all of them using this feature but as we grew, we needed things to run in parallel to keep the Twitter reports as up to date as possible. So from very early on, we’ve run multiple containers just for Twitter API calls, so they can get data faster. Before Twitter required authenticated calls, maybe 5 years ago, I assume they were throttling based on IP address. Our IP address would have been the same for all containers since they were all behind a firewall. So although we were running multiple instances, as a group they have had the same total API limit as a single instance – we were just able to search faster, while still respecting the limits.

When they switched to authenticated API calls only, it was still a pretty open system. E.g. you didn’t have to “apply” for anything like you do now. We had 4 Twitter containers, so we created 4 Twitter “apps” in our dev account, and updated our containers to each use one of the apps’ auth keys. Looking at it now, I believe this let us get around the intended limits since I would assume they changed enforcement to a per-app basis now. I don’t know for sure how they’ve enforced limits now or previously, these are all best guesses, but if I’m right then our ability to skirt the limits was an unintended side effect of the API evolving.

I just realized now that the cron jobs for this were still running, all API calls were being denied though since the account is suspended, but I just turned off these cron jobs so there should be no more hits to their API.

And I believe that’s all of the relevant information on that, to the best of my memory and knowledge.

That’s everything. I can literally think of no other possible reasons why we would have been suspended.

If you’ve read this far, thank you. I’m hoping one of you knows someone who knows someone who can help us out. I’m more than positive this is all a misunderstanding. So please, Twitter, if you’re reading this, tell us why so we can fix it if necessary, and get us back online.