Hacker News new | past | comments | ask | show | jobs | submit login
Nextdoor's Heisensubscribe (and Other Dark Patterns) (taylor.town)
186 points by surprisetalk 9 months ago | hide | past | favorite | 108 comments



That's not even the darkest dark pattern on NextDoor

When you try and switch your newsfeed to "recent posts only" they show a toast notification "Your selection expires in 60 days" https://imgur.com/1fZpleg

Presumably because they've found they can drive more engagement by recycling old controversial posts.

They are true innovators in dark patterns.


Twitter started doing that at some point 5-6 years ago, after they introduced the algorithmic alternative to the chronological feed. Super annoying. https://www.reddit.com/r/assholedesign/comments/cttl73/twitt...


Instagram does something similar with suggested posts. Every time I see suggested content in my feed, I click the extra settings and tell it to “snooze suggested content for 30 days”. There’s no option to turn it off entirely, but there’s an option to disable it for 30 days. they know some people don’t want it but they really really want to shove it down your throat.


I activated that 30-day snooze for months in a row. I no longer even have that option.

Others I've showed it to still have it, so either I'm part of an A/B test, or I exceeded my fiscal year allowance of browsing my actual follows without algorithmic fluff interposing.

Joke's on them, I set a 15-min daily limit on the app and spend my time elsewhere.


They probably did some user studies and found out that even users who disable it end up scrolling more when it is enabled. The only way to win is to not play.


I’d take that over facebook’s insistence that you can only turn off individual suggestions. There appears to be no way to turn off all of them, even for a little while.


yuckety yuck yuck, what a crappy app :(


They do have an unsubscribe from all button. It's in the delete my account because i get too many emails user flow. Ask me how i know. Nextdoor is a master class in dark patterns.


Instead of trying to unsubscribe, the best way to punish a business is to use the “mark as spam” button liberally.

Most email sending services have upper caps on acceptable complaint rates, and if done by a sufficient number of users, would result in a suspension of their sending capabilities, and an IP address block on whatever email service you’re using.


Yeah if you clicked the unsubscribe button and they're still sending you email, then it's spam. It doesn't matter if you think the company deserves to be "punished" or not, this is just literally what the word spam means and what that button is for.


Yep. I'm getting more and more aggressive with spam reports every year. If they don't have a single-click Unsubscribe link in the email, I won't even bother trying. I just mark it as spam. If the filter doesn't learn, I add them to my Fastmail rules which automatically reports it. My inbox surprisingly remains one of the most tolerable facets of my internet experience.


> the best way to punish a business is to use the “mark as spam” button liberally

I like the idea but I doubt it. Who do you think is enforcing this? Their transactional email provider don’t want to lose the business, Gmail/MS/Fastmail aren’t going to spamhole a big legit business…


Gmail and MS are particularly aggressive about blocking abusive senders, and it gets talked about a lot in companies that send high volumes of email and complying with their requirements.

I used to work for a email sending provider a few years ago. For smaller senders close to breaching the limit, warning emails would be sent out reminding them about email sending practices; for larger senders these would take the form of meetings set up between the customer and their account manager about the issue.


I think there is zero chance of either of them blocking Twilio (which is what Nextdoor uses), as their users would riot. Twilio aren't going to drop a customer who they prominently feature[0] for sending emails to customers to whom they have an existing relationship.

The only kind of spam anyone really seems to care about is truly unsolicited email. A publicly listed company with an aggressive stance towards emailing their existing users, nobody is going to get excited about.

0: https://customers.twilio.com/en-us/nextdoor-2


All I can say is IPs can and do get blocked by email receiving services, and especially Gmail and MS with their automated feedback loops, with all the implications of collateral damage that it entails. I remember helping provide data points in a similar conversation with another large customer (and a brand name that most would recognize here) who was concerned about increasing complaint rates.

The problem, if anything is that people kinda don’t know about what to do with these emails, and if more people marked these emails as spam, senders such as Nextdoor would sit up and take notice.



After I changed my address to comiskey park I stopped getting anything from nextdoor as, I suppose, the account is suspended until I verify the address, which is fortunately not possible.


"How are you gonna get the band back together, Mr. Hot Rodder, Mister Motorhead? Those cops have your name, your address..."

"They don't have my address. I falsified my renewal. I put down 1060 West Addison."


The worst thing for me about nextdoor emails is I get a snippet of some really controversial thread, and when I click on it, it was already removed by moderators.

I wanted to read what dumb things people in my neighborhood were saying, but instead I only get teased.


The worst thing for me about nextdoor is nextdoor. Everything about the app brings out the worst in people. Nextdoor made me realize that I'd really rather not know that much about my neighbors.


Nextdoor is particularly extraordinary in how people will sign their real name to utterly vile comments in front of the people they live near. Even Facebook was never that bad.


>> Nextdoor made me realize that I'd really rather not know that much about my neighbors who post on Nextdoor.

FTFY

I really enjoy meeting and casually getting to know my actual neighbors in person to person interactions. The local Nextdoor is a cesspit. Its just not the right medium for actual human connection.


This. Nextdoor recruits your local homebody to LARP as stasi for big tech, censoring their friends and neighbors because of minor/petty disagreements.


This is where you just hit report spam and block all their emails. Maybe if enough people do that they'll get blacklisted.


I do this when there is no obvious unsubscribe link in the email. I'll also hard block if the unsubscribe link requires me to log in first.

I haven't needed to hard block in a while, maybe the message is starting to get through!


I mark all marketing emails as spam because I didn't ask to be sent them in the first place. Why place the burden on the user. There should be a hit to your sender rep every time you send out emails the user didn't explicitly and intentionally sign up for.


Agree, however I do understand the desire to keep me aware of their product.

If I ran a B2C sass, I wonder what I would consider appropriate. Maybe a quarterly update on new product features?


... or if the unsubscribe link requires you to type in your email address.


This reminds me of Air Canada’s dark pattern unsubscribe that I just went to.

It takes you to a page of about 5 toggles for what you want to unsubscribe. To unsubscribe you must toggle the untoggled options to on rather than off and then save. Things that you leave untoggled will be what you are now subscribed to.


I found the mention of "Heisensubscribe" three times in the article but it assumed that the reader understood the meaning. I really want to learn this term which is new to me. I have a vague idea from reading the article and I might even be able to form a sentence and use it but I want to know how/where/why this comes from. I did searched but no satisfactory answer.


It's a pun on Heisenberg's uncertainty principle.

https://en.m.wikipedia.org/wiki/Uncertainty_principle


Normally HN isn't the place for jokes, but this seems like the right time for this (very old) one:

Heisenberg, Schrödinger and Ohm are in a car.

They get pulled over. Heisenberg is driving and the cop asks him “Do you know how fast you were going?”

“No, but I know exactly where I am” Heisenberg replies.

The cop says “You were doing 55 in a 35 miles per hour zone.” Heisenberg throws up his hands and shouts “Great! Now I’m lost!”

The cop thinks this is suspicious and orders him to pop open the trunk. He checks it out and says “Do you know you have a dead cat back here?”

“We do now, thanks to you!” shouts Schrödinger.

The cop moves to arrest them. Ohm resists.


Is resistance the law, or currently against the law?


It's potentially against the law.


I assume it's a play on `heisenbug` which is a bug that is not easily reproducible, or more strictly a bug that is unlikely to happen when you are specifically trying to observe it (by enabling debug logging for example).

I assume the use here extends the uncertainty part of the problem to the unsubscribe workflow.


I've never heard of a Heisenbug, but that is clearly a play on the Heisenberg Uncertainty Principle [0]. It's unlikely that the author was referring directly to a Heisenbug, instead I'd guess both are different plays on the same name.

[0] https://en.m.wikipedia.org/wiki/Uncertainty_principle


It is a play on the Heisenberg Uncertainty Principle as is a Heisenbug, which are bugs that happen differently or don't happen when you are observing the code execute. The typical case is you have a crash you can reproduce, but when you attach the debugger to the process the crash no longer happens due to subtle differences in code execution timing.


It's something like settings will change while you look for them and also a play on the ambiguseless text descriptions for their UI toggle switches (reobserved 3 days ago in https://news.ycombinator.com/item?id=39342649).


Edit: My guess is Heisenberg and the Uncertainty Principle.

I can’t find the answer online either. Sounds like a half German half English word that means something like “my name is subscribe” or maybe it’s referring to Breaking Bad. I dunno either.


A few years ago when I just moved to my neighborhood, I got a paper mail from nextdoor. It's a welcome mail and has real people name on it. The mail says something like "John Smith, your neighbor, would like to invite you to join our community". I had never heard of nextdoor before, so I thought it's from my own HOA because otherwise how would they know I'm new here. So I signed up and quickly realized I just fell for a "growth hack".


I think we should solve this problem with some sort of social seal of approval. I call it E-Corp (like B Corp certification)

https://twitter.com/justinprojects/status/151393246732075418...

Companies and projects can get E-Corp certified if they:

- publish their email templates publicly

- are regularly audited by anon user accounts (created by the e-corp project)

- honor unsubscribes per type

- have a unsubscribe from all in emails link

- have a unsubscribe from this type in emails link

- link to their published notification surface area doc

- clicking any "unsubscribe from this type" or "unsubscribe from all" link is honored immediately (meaning "are you sure?" or "enter your email" follow ups after clicking/tapping)

Can work for push notifications too


There used to be an excellent copy-paste “Why your solution to spam won’t work”, that I tried to find for this response, but obviously it’s no longer simple to Google because the relevant search terms have all been “optimised”.

A massive QoL improvement that is implementable — Apple actually enforcing their guidelines on notification-type spam — seems to continue to be a dream only.


Yes, there are 16 pages of notification settings.

Certainly not helped by the ridiculous amount of whitespace that seems to be the awful trend these days, and the insanity of repeating "Allow notifications on Nextdoor" on a separate, double-spaced line for each item. That UI is unfortunately typical "modern" UX with all the elements floating vaguely in borderless abandon, indecipherable hieroglyphics instead of text, ambiguous sliders instead of checkboxes, and cutting off text when there would've been otherwise plenty of space to show it all ("connecti.."?)

There are no “disable all” buttons. In order to unsubscribe from everything, nextdoor demands hundreds of clicks.

Or perhaps a few minutes with a JS loop in the console...


The only way I was able to get Nextdoor out of my inbox was by blocking their domain in gmail - it's absurd how much spam they're capable of sending without getting blacklisted.


"Blacklists" are only permanent for small-time companies. Large companies can always use their networks and back channels to the major email providers to reverse blacklisting or sender score declines.


I’m baffled by how Nextdoor keeps making it through Gmail. I feel like I’ve marked innumerable emails from them as spam over the years.

Maybe they are a major revenue source for Google?


Same for Salesforce.com emails.

I get so many spam emails with bounce addresses like jhvivisvhesnbg@aa85oc16gsbsm2ug.g8k1xdhfdya6f42b.jnwwym.8d-5df9reaw.um8.bnc.salesforce.com, and mark each one as spam. The filter never learns.


Google's inaction here seems to map directly to Cory Doctorow's definition of Enshittification [0].

I just last week re-listened to him describe the process again. It's pretty remarkable how well he nailed it (and how many companies it applies to).

[0]https://en.m.wikipedia.org/wiki/Enshittification


I always thought that gmail ignored you when you marked emails as spam... lately I have gotten loads more spam too


A couple of days ago, I received an email from Nextdoor inviting me to my neighborhood group. It had my real name and was sent to the email address that I only use for business.. I never use it to sign up for services or anything with a privacy agreement.

I sent an email to privacy@nextdoor.com that simply read "Please tell me how and when you obtained my personal information, and then promptly delete it from your systems."

Surprisingly, I received a response within 24 hours:

"Hi neighbor,

Thanks for your email.

From time to time, Nextdoor receives information from third parties about non-users. In your case, we received your information from a third party partner and used this information to invite you to join your Nextdoor Neighborhood. We’re sorry to hear that this email was unwelcome.

I can confirm that we have deleted your personal information from our systems.

Let me know if you have any questions.

Best,

Maureen Nextdoor"

I responded: "I'd like to know the name of the partner and the date of the transaction, please."

Their response 4 hours later:

"Hi XXXXX,

Thanks for your reply.

In keeping with our legal obligations, including our obligations under US privacy laws, we can only inform you that we received your information from a third-party partner.

Unfortunately, we are unable to provide you with additional details at this time."

It's incredibly frustrating. I just want to know who sold me out so I can avoid doing business with them in the future.


Sometimes I'll use a service-specific email address (usually with the name of the service, like XXXNetflix) and it's fun to see just who gets ahold of it.


My normal strategy is to use the name of the service as my last name. But that's with the email that I use to sign up for things. This email address I've managed to keep "sterile" for close to 10 years, until Nextdoor got their hands on it somehow.


They cannot tell you, for reasons of violating privacy. Yours, mostly.


Obligations under US privacy laws that somehow protect them but not you. Typical.


I feel like we need an email client that is basically default drop with conntrack. If users do not explicitly allow inbound emails, they are not delivered (users could get a list of blocked addresses). If users receive a reply to an outbound email, it is allowed.


Weekly emails are too frequent?

Tell that to Gap, who sometimes sends up to five spams per day.


Nextdoor is utter garbage. I had to change my email in the settings to a fake email to get it to stop sending me stuff. A trash company.


Nextdoor sent me a PHYSICAL LETTER that was even shadier

It tried to look like a letter from a neighbor, but it was postmarked from SF.

The letter read like it was someone from my neighborhood, with a real address and name, inviting me to join their community page, but the tone was off. Very weird and corporate feeling.

It was wild to get literal spam mail delivered via USPS


When I created a neighborhood that was an option initiated by the creator.


Here is what you can do (from someone who tried many times to block their emails), write to privacy@nextdoor.com and request that all your data be deleted.

As someone who recently had to implement delete requests at a startup, I can tell you that this process usually exists and is handle by a different team. It's much quicker.


16 pages of notification switches to disable? This is wickedness on par with Meta's current way to navigate your personal/account settings, which is most likely engineered to look descriptive and helpful, but is actually impossible to find what you want to change or disable.


I wonder what kind of HTML elements are those switches on the unsubscribe page.

I have found the snap links extension [0] quite useful in the past for "clicking" on many checkboxes at once by drawing a box around them. (mostly those surveys fast food places want you to fill out to get a coupon)

0: https://addons.mozilla.org/en-US/firefox/addon/snaplinksplus...


Reminds me of Tumblr's early third party cookie list. No reject all, but a long list of third parties which all had to be turned off separately.


I made the mistake of setting up a Nextdoor account once. It was a cesspit of conspiracy theories and blatant racism with little or nothing of real value to be found, so I deleted my account within a week.


As much as I agree that 98% of the conversation on Nextdoor is worthless, it was invaluable to me one day when a stray dog showed up on my doorstep and I had to track down their owner.


I also find Nextdoor to be the most reliable place to find good recommendations for contractors nearby. I no longer trust Yelp and Google reviews.


Are you concerned that the racism, etc. apparent in other Nextdoor discussions (if that happens in your neighborhood board) will play a role in the recommendations?

I find local recommendations often look similar to the recommender. Sometimes it's hard to break out of that bubble.

Edit: I ask because that is the mechanic of systemic racism: You're not doing anything, you're probably not racist and likely not even aware of the 'system'. Yet the system still churns out racist outcomes: You know (e.g.) white people, they know white people, and that's who they recommend to you, and who you then end up recommending to others.


And despite that, I still find more interesting and useful information than I can on X.


I mean, that's a low threshold these days.


Anything "local" that's also unmoderated quickly turns into weirdos fearmongering at all hours of the day. City subreddits are often like this too.


Same


Any company doing it enough times just goes straight into my spam filter. Don’t even bother entertaining their UI maze.


Oh there is a 'unsubscribe'.

It's "delete *@nextdoor.com on arrival". Do this for all domains they use.

Also, report each and every one as spam. Use search to find all, and report all.

And also, delete account. Send shitty email to legal threatening unsubscribe and removal of all data. Come from EU VPN and demand GDPR followed.


If Nextdoor was extra hostile they could swap the “Heisensubscribe” toggle each time you visit the page. Hmmm…that didn’t work. I’ll go back in and toggle it the other way. Hmm…that didn’t work either?


There's a straight line that can be drawn from "dark UX patterns" to "Venture capital funded company who don't care about users outside of how they affect the DAU KPI".


I've never even visited the Nextdoor website and yet they somehow managed to spam me via a neighbor who taped a handwritten letter to my front door inviting me to join Nextdoor.


At this point in the game it’s safe to assume any “social” platform is engaging in dark patterns to drive engagement. It’s the attention economy at work.


Nextdoor banned me because i called out my neighbors for spreading disinformation about made up neighborhood crime. Nextdoor is a disinformation platform.


What do you think is going on there? While I see mis/disinformation about crime in many places, and hear about it on Nextdoor, I'm not sure what the motive is?

The Citizen app works (afaik) by drumming up fear of crime, and therefore emotionally addicting people. According to a Fast Company article (that everyone should read), Citizen has staff adding crime reports to feeds that are peaceful.

But they specialize in crime and have a profit motive. Why would neighbors be drumming up fear of crime?


>What do you think is going on there?

It all goes back to who the mods are, in this case "nextdoor" did not ban him it was local mods.

They get a little fiefdom and then ban dissent in their area, OP just pissed off one of the local "big fish".

No moderation is highly preferable to me if the alternative is orwellian censorship


> No moderation is highly preferable to me if the alternative is orwellian censorship

No moderation, outside specific situations, is unusable.


What do you mean exactly?

Moderation in the classical sense, where spam/off topic content is removed?

or

"Moderation": Consensus manipulation, speech/thought policing, censorship of opposing/controversial ideas, etc?

Because in practice "Moderation" is what you get (and what they spend most of their time doing), but they sell you on the above by fear mongering about spam.


It's a balance, right?

I want to hear from all sorts of people, and I want free expression. But I don't want to spend time with people who behave badly - people shouting, people saying offensive things, people being rude, etc. Humanity universally has manners and social norms for a reason, to enable us to interact.

Public forums are vulnerable to people coming in and behaving badly; we need moderators to keep the place civil without suppressing free expression.


Because they're racist and "crime" is a dog whistle for "existing while not white".

It's incredibly tiresome.


I believe that happens, but have you looked at it and seen the mechanism in action? Is there something else going on too? I'm really intersested in what people's direct evidence is.


I grew up in it. I participated in it, until I got out into the world enough that I realized how wrong I was.

For this specific case, no of course I can't be sure, I don't know the specific neighbors.


For some, I suspect it's just the excitement of it. Watch the likes and the comments stream in. Same reason kids in middle school spread rumors.


I don't find that unusual at all. Doesn't that explain most social media - it's engineered for that excitement. Doesn't that explain HN?

It's not just middle schoolers. Our primate ancestors love to socialize; we've been doing it for seven million years. That's what social media sells. It's certainly not utilitarianism!


Why do people use this service? An honest question.


I check once or twice a week to BOLO any lost pets and get updates on ones that have been found. Of course given how many posts like that crop up it just makes me annoyed that people who "love" their pets so much can't be bothered to do simple things to keep them from getting lost, like keeping dogs leashed when walking them through the neighborhood.


Wondering the same. I bailed on the idea as soon as I saw it has the same problem as Reddit - the people who jump for control are never the people you want in control.


Is it the power what does filtering for a specific people to match their profile? Or does the power actually change those who acquire a control over others?


I don't use nextdoor (and won't)

Just the same when I recently moved, I received mail from them asking me to sign up. paper mail.


Nextdoor emails and notifications suck. I’ll get “safety” notifications that have nothing to do with my neighborhood.


I found the unsubscribe process tedious but not write-a-whole-blog-about-it-tedious.


These tedious practices need to be called out, and blogging about it is one way. If everyone thought like this comment, then we won't complain about anything to others. That's not a method that makes the world a better place.


Nextdoor is a nightmare.


Wait until you reach a live person in Support there (via email).

this bozo told me the reason I wasn't getting the daily email digest must be on my mail client's end. Back and forth, rinse and repeat.

then I went to the Twitter account and asked the same question, and got a completely different answer, which happened to be correct. (Wish I could remember what it was.)


Nextdoor is such a disappointment for me. I've never been that invested in the giant social networks but Nextdoor often has legitimately useful posts from people in my area. But my God, the company behind it is so toxic they ruined what could have been a great product. It's like they have tried to turn it into a low rent Facebook and largely succeeded. I used it regularly for a time but have pretty much given up at this point and can't remember the last time I logged in.


The biggest problem IMO is the local moderators. They were too cheap to hire their own and instead recruited locals who were somehow unfit to run an HOA, then gave them censorship power.


Wanted to unsubscribe but they require you to login to do that. Can't recover account due to some recovery information supposedly being wrong? So I just mark them as spam. A bit insufferable but nothing else to be done about it.


They need to make dark patterns illegal, somehow, maybe.


Just stop doing business with companies which use them. I do not use Facebook, Instagram, Twitter/X, Amazon (I buy electronic books but avoid physical goods because they try to trick you into buying Prime) and next door. All use dark patterns and the solution is to stop using them.


The solution to poor regulation is almost never a boycott.

We didn’t get people to stop buying lead paint, leaded gasoline, or asbestos insulation via boycotts.


I've wondered what that structure might look like. CFPB? CAN-SPAM?

I miss muscular regulatory bodies.


I agree with your last statement, but for the structure, I think it needs to be something else... maybe something like CAN-SPAM, but adapted for a wider range of conditions.


Step 1: Sign up for a hot-garbage website.

Step 2: Get hot-garbage emails from hot-garbage website.

Step 3: Complain?


A single signup shouldn't require sixteen pages of toggle switches to undo.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: