Facebook Labelled my Medium Blog as Spam and Then Pulled Their Decision
Trying to understand Facebook’s methods for categorising spam after being categorised as spam and then exonerated
In one way I really respect Facebook’s attempt at controlling what is supposed to be posted and what should not be posted on their platform.
However, at the same time I want to tell you about my experience being categorised as spam.
This article has a long title, but I could not think of a short one immediately.
It was kind of hard to grasp.
Why, unknown moderator?
Why, unknown algorithms?
I want to consider why it happened, although I do not know the answer (yet).
This article will of course be a series of guesses.
According to Facebook, this is spam.
In fact, my entire Medium account had been labeled as spam.
I wanted to tell my MSc class in Social Data Science as well as my friends about starting a master’s programme in the middle of a pandemic.
Apparently, this was not acceptable for Facebook.
I was not given any explanation as to why.
It simply was ‘against our Community Standards’ on spam.
That is a pretty broad reason.
Let me try to investigate why.
Was it the mask?
I have a theory (supported by some articles online) that Facebook has been trying to limit the sale of non-medical masks. This was likely well-intended to avoid fake masks being sold.
In March there were huge issues with people selling ‘non-medical’ face masks:
Facebook banned medical face mask listings, and they're still littering the platform
Last week, Facebook announced that it would begin "temporarily" banning the sale of medical face masks on its online…
Late in June they have an article that says they allow it (with conditions).
Allowing the Promotion of Non-Medical Masks on Facebook
UPDATE on August 19, 2020 at 10AM PT: Today we further scaled back our temporary ban to also allow hand sanitizer and…
“Advertisers must currently be in good standing to advertise non-medical masks, with a minimum advertising history of four months.”
Since I do not know, this is only guesswork — but did Facebook factor in my use of a ‘non-medical’ face mask when moderating my post?
As you can see my mask has typography on the front. It says “SODAS — Centre for Social Data Science”.
Far-out thoughts went through my mind too.
- I have written about Facebook and how they govern moderation previously.
- I have written critical posts about Facebook and its governance.
In my opinion I wrote down these thoughts as constructive criticism and to explore how Facebook works, mainly to consider different types of expression.
How does Facebook define Terrorism in Relation to Artificial Intelligence?
AI and Definitions of Terrorism In Social Media
As well as problems with moderation:
The Failed Facebook Spam Filtering of the Coronavirus
Was it a Fault Within the Field of Artificial Intelligence or Not?
How is relevance ranked?
I would not like to think that this would factor in, and likely it does not, however it makes me pause…
I do not think that would be the reason.
At the same time it is a frightening experience to have an output silenced…
There are some clear guidelines directly from Facebook.
When I send a message to Facebook there is no reply. Except for the ‘closed’ or ‘open’ case status on my complaints.
I sent several messages telling Facebook that something went wrong and contesting the decisions to label my entire blog as spam.
Over the course of the last weeks I have sent messages quite a few times.
Even after my Medium account was cleared I sent the message just wanting to find out how the decision was made:
“I have written to you several times, and it is impossible for me to understand how my Medium page project broke your Community Standards. I have received no feedback and I disagree. I have a project called 500 days of AI on Medium and the account is original. Was this a moderator-decision or was is based on algorithmic decision-making? This is interesting for me to know because I have been doing this project for some time now looking into how decisions are made with algorithms. Right now it feels like I am talking to a void. I have sent you different complaints now with increasing regularity, yet I am not receiving anything back at all. Can you tell me about your governance procedure and where the decision to label my whole personal blog project as spam was made? Some of the posts that I have wanted to share lately have been very personal — I am almost done with a 500 day project on artificial intelligence and then suddenly AI or human decides that my whole page/project is spam. EDIT: I tried to post again and it seems my post was allowed, but I would still want to know what happened if possible.”
I mean, how are decisions made?
I did not even know my account had been cleared.
- Did I post too often?
- Did too many people repost my texts online?
Do they know?
Was it a person or algorithms?
More importantly: should I be able to know?
Would it be too much if a small percentage of users on Facebook started disputing decisions. It is ‘their platform’ after all.
At least that is what it felt like in the given moment.
That being said, this is not much considered how many people have to live with information being managed or their access being denied. As such, my complaint or want for clarification is a minor one.
I think Facebook does a pretty good job, and it seems like they are working hard. Yes, there are issues — and we should explore them, however I am grateful that I (on conditions) can express myself online.
What do you think?
What is your experience?
Please let me know what you think.
This is #500daysofAI and you are reading article 478. I am writing one new article about or related to artificial intelligence every day for 500 days.