Filter out low-quality messages

Your patronizing tone leads nowhere either. You just sound rude, and you cannot expect people to chill when you are being patronizing. Chill? #nonoscalmamosnada


Anyone else has a profile on Trustroots?
Whenever I receive a message or request I also get to see this:
Screen Shot 2021-07-03 at 4.59.03 PM

(this was either a copy and paste request or “hi” so that’s why I chose NO).

I have no idea what Trustroots do with those feedbacks later. Maybe when the user gets too many “NO” those users might eventually get a message or email with tips on how to approach other users (read profiles, write personalized messages/requests etc.) or maybe they get blocked? Once again- I have no idea how this works on Trustroots just sharing what I saw :smiley:

Would this YES/NO feedback help educate people on how to write quality messages?

Also, the minimum word count/spam folder was discussed a year ago in July 2020, have you decided whether you like this idea? I think it’s not bad at all. If I was a spammer and knew that my “hi” is too short and goes directly into a spam folder it would motivate me to write a longer message and express myself instead of “testing the waters” or whatever these people are doing.


I have a lot to say about this debate too, and I would love to talk about this topic in the one or both of the online event this week. I think we can probably understand all the angles a bit better if we discuss it live, too!

Thank you for staying civil even though the topic can be really sensitive and triggering :purple_heart:


Let’s stay practical, what automated education is possible on a platform like this one?


@anon14140932, you write really condescending replies and then act like a victim when people defend themselves. If you cannot participate in a forum without escalating it with passive aggression then you are creating a really toxic environment.


I was thinking about this, on one hand it’s exactly what we want to avoid, on the other hand we want to enable people to connect in real life when they are travelling and it’s not easy to distinguish.

A good start, at least, would be having friendly, concise text explaining what messages are appropriate on the screen when you message someone for the first time.

We could also go down the route of confirmation dialogues etc. but there should be a balance, we don’t want to hassle all the time to remind people to be good. (Maybe the first time you send a message? Once a month?)


Seems that easiest way to tackle this problem fast is simply to introduce a 50 character first message minimum.

“I’m in town, let’s meet up for a drink. Somewhere in centre, tomorrow 5-7 pm” is about a shortest reasonable message one may write to a stranger, and that still leaves a lo to desire, like why, who and so on, and it has (if I calculated correctly) 74 characters when spaces included.

In the long run I bet the developers may figure out other ways to sort our useless spam, which is not ill-intentioned, but tiresome and annoying for the recipient. As a male I am lucky I have not received much of such messages, but in CS over the years I have had heaps of pretty useless couch requests, I mostly have declined for the sake of not filled profile and not enough info about the person and why I should host him/her. Also non-solicited friend requests are somewhat annoying. I don’t want to be friends with someone I don’t know in real life and with whom I have not even exchanged first message to find out if there is anything in common.

The gender omit/manipulation I see as not useful, as it has its perks, though in general I think gender in hosting is secondary as long as people do not abuse the system for wrong reasons.

Another probably easy way to tackle this problem could be to limit less than 100 character first messages to one per month. That’s how these people could never bother too many people at once.


I wouldn’t like the person who sends “hi” to random people to elaborate on what he wants from me :joy: :joy: Totally agree with Emily:

I like the idea of just filtering those out as spam - without notifying the sender this is a bad message. I’d suggest one general filter by default that sends “hi”, “hy”, “hello how are you” and all this stuff to spam folder.

Also, what about going about the first message like Instagram did? If you receive a message from a stranger, you can choose between “allow”, “delete” or “block”.

And I don’t like the idea of constant reminders like “hey, it’s not a dating site” or “hey, was this message polite and respectful” at all. I feel they can be a huge turn off for nice people by themselves.


Filtering out etc those “spam” messages would also require they do not count in personal statistics “not responded” to diminish one’s response quota.


I like all the ideas presented here: a short educational message now and then, a minimum of words for fist messages, and a folder for first messages like the IG model.

I also like the feedback idea, wheter it is similar to the one trustroots have, or just an allow/delete/block option. I think if you have too many first messsages being deleted/blocked maybe you could recive a polite informative message about the way to approach people on the community, and that this will be reflected on your community standing. Its educational and simple. And makes clear for the rest of the community this person is not behaving according to the correct spirit.


My last request on Trustroots was literally “Hello!!” and it was from a girl. I don’t think her intentions were bad. Most likely she didn’t know this is not a real life where you don’t need to exchange polite greetings and questions before you write what you want.

I think system messages or emails with tips would be helpful for people like that.

Would it be possible to send these system message only to people who barely get any replies back so we don’t have to bother “experienced” users who don’t need reminders of this kind?

Can we do anything else except 1. ignoring low-quality messages 2. having the possibility to flag these messages as spam/having a spam folder 3. having the possibility to block/report others users or 4. having system messages/emails to help educate users?


Oh, I like this!

Could be different kinds of them also, smth like “we see you don’t get many replies, one of the reasons could be that you don’t tell much about yourself or the purpose of writing, find more in our FAQ here” for those who just don’t get answered, and “we see your messages often get blocked by users. we want to remind you this is not a website for dating or flirting and your account can be banned for such behaviour” for those being blocked.

  1. when counting characters you should also consider the languages. If you use a language with non-phonetic alphabet it is possible that a full messge could be much shorter, in term of characters, than it european counterpart.
    2, just an Hi could be a messge to ask if the person is interested to start a communication.

We should educate people not to do that. A proper message if you want to start communication would be " Hi, I´m this person, I´d like to start communication because of this reason". That´s why I said earlier that calling it “chat” does not help: in a chat system you just say hi and the convo is back and forth, and this may give users the wrong impression about how to use the platform. Calling them messages or something that gives the feeling of the need of a more elaborated content would assist on educate people about the good practices. Is just a silly detail, but seeing the word chat does influence the way we use the tool.

Also, a info page with info about the proper way to use the platform and the message system is needed. I know it´s small now, but technically it doesn´t say anywhere in the platform that it is not a dating platform at this point. If someone gets the link to the app, this is what they see

Yeah, I know there´s an “about” button on the app that takes you to the main main page with all the info about the project, but again: it doesnt´say anywhere plain and simple: “dating is miss use of this platform” or what it is expected as proper behavior from the users. =/


I like the creative idea but as my main role has been a hoster, any sort of message that short would totally prompt a decline. Even the social messages I have sent in the past wouldn’t pass that test. I also realise that the messages I sent about events and such also had prefaces explaining my intentions or what information I was seeking.

I had a look back at the last 5 years of messages, and none of them would have been able to meet a 50 character, or even 100 character criteria. This may be the result of my expectations being crystal clear in my profile. Back in the day this was a defining feature of the platform and the culture was focused on having full and creative profiles, and meaningful requests.

I do realise though that there many that don’t have this preference, but from my own experience I find the accept/reject thought process super easy and straight forward. As a result I’ve not had any sort of bad experiences.

1 Like

50 characters is not a lot. “Hi! My name is Aleja. Hope you are doing well today” That’s 51 characters. It’s totally doable, even for spammers XD

I agree with you that a proper first message should be longer, I never accepted a request that short, but that’s more a personal choice.
(edited because I pressed send too soon)


You’re absolutely right. There are several big updates coming and one of them is explicitly mentioning during signup that is not to be used for dating or hookups, and people will have to confirm that they agree not to use the platform in this manner.

I’m the same way! My first message to a person tends to be a bit longer and I always show the person in my message that I have read their profile (and of course explain why I am contacting them!)

I agree that there should be some kind of character threshold on the first message to someone and on requests (surfing/hosting/meet-up). But what do you think about disabling the character requirement if the person is already your friend? If you already know the person in real life, I think it’s normal to add them as a friend and then just send them a quick “Hey! Great to see you here!” or something similar.

I also wanted to take a moment to share a few upcoming changes that should help improve the whole messaging system overall:

  1. Improving the onboarding process to better educate people of our values and the rules when they sign up for their account. People who signed up before this change will receive this information a different way (either via e-mail or perhaps on their next login).
  2. Prominently linking the community guidelines and help/support links on all pages of the platform so people can easily inform themselves or reach out for help.
  3. Fixing messaging so anyone can message anyone (hopefully reducing the amount of friend requests from strangers).
  4. Enabling a “block” feature so you can quickly and easily block someone if they message you or send a friend request when you don’t know them.
  5. Add more robust reporting features. Currently you can only report someone from their profile page or by clicking the red “bug” button in the corner and sending an e-mail to the address listed there. In the new system you will be able to report profiles, messages, posts on the community pages, events and more. In the meantime, you can e-mail or with a detailed description of the problem and the username or link to the person/content in question.

Please keep the ideas and feedback coming! Our goal is to improve upon the original CS platform and make it 10x better!


Yeah, I think if the character threshold is there, it should only apply to first messages to strangers.


Didn’t know about this Google API until now:

There’s a plugin for Discourse, the Forum software, that makes use of it to queue posts for moderator review and prompt users if they are sure about submitting a potentially toxic post, before submit.

Just sharing in case it’s something interesting for the app.

1 Like

I have seen this before and I have mixed feelings about their API and methodology.

I personally worry about the use of metrics to decide the “extent” to which something is “good” or “bad”. It’s fine if we just use a system to flag things into “good or okay” or “probably bad” buckets, but when one starts scoring every post on a scale, it can have a chilling effect. People are discouraged from voicing their less-conformant opinions, and it can hurt discourse.