Yik Yak Systematically Downvotes Mentions of Competitors | Hacker News Reader

Yik Yak Systematically Downvotes Mentions of Competitors

techcrunch.com

95 pointsburger_moon45 comments

While their response is pretty roundabout, my experience with moderating/owning several community sites is that advertisements for competitors really are one of the largest categories of spam, and I wouldn't default to reading an attempt to mute them as a "shady tactic."

Think about the point of view of a founder-developer. You have a list of bugs to fix, a bunch of people to meet with, etc., all while manning a pool of servers to make sure they stay up. Someone says to you, "hey, we're getting a lot of spam from/about this company called Fade." I'd bet a lot of programmers would just write "if 'fade' in message" before getting back to work on other things.

Fade was viciously spamming yik yak, on the other hand. It's pretty ludicrous of them to pretend like they're being oppressed (and to compare Yik Yak to North Korea!) for retaliation against that.

Sure, but Yik Yak could have had a "report spam" option, or made users type captchas, or reversed down votes if it saw other people gave attention to the post. I can think of a million ways that would be better than what they implemented.

Your comment here on HN would be marked as spam for using the word "Fade". As would mine.

I totally agree that Fade spamming Yik Yak is bad, and pretending that they were wronged by Yik Yak is ridiculous. But that wrong doesn't make this strange censoring right.

It's especially troubling that the posts aren't removed, but rather, get softly automatically downvoted over time, making the process almost invisible.

> It's especially troubling that the posts aren't removed, but rather, get softly automatically downvoted over time, making the process almost invisible.

Seems perfectly reasonably to balance the expression of users with "fading" spam. If a post really has strong merit, it can still get some visibility, whereas a deleted post can not.

How do you viciously spam a localized anonymous message board?

- Your posts a limited geographically - The app knows your device ID - The servers should be able to easily detect location hopping

With all of that in mind, limiting SPAM posts should be simple.

If, instead, your goal is to suppress mentions of your competition, well that's a whole different problem set and you see how the implemented solutions works.

First rule of networked application security is "Don't trust your inputs."

It's relatively easy to deceive an app about it's location; easier if it's running in an emulator environment. And that works just as well for iOS as it does for Android.

Yes, I know that. That's why I specifically mentioned location hopping. You can also fake device IDs. To fight that you'd have to do even more. The point being, a simple word filter is not how you fight SPAM, it's how you silence something you don't want discussed.

You can endlessly create emulators that only spam in a specific area. I only know about Fade because of yik yak spam.

Yik Yak is anonymous and has no concept of users. Fighting spam is going to be hard.

I had never heard of Yik Yak until today (and the mentioned competitors). I tried Yik Yak out, 5 minutes after it was deleted.

I can't express how disgusted I am by the contents/comments within the app. It's full of hate speech, insults, sexism, etc. I find really disturbing that these sort of apps get vc money, really disturbing.

Couldn't you say the same thing for Facebook, Twitter, Tumblr, and Reddit?

On Facebook and Twitter I've never seen a friend or someone I follow make some disgusting comments like I've seen on Yik Yak, specially the top, upvoted stuff. I never used Tumblr (only when visiting a blog) and only visit Reddit sporadically. So no, I can't say the same.

Try running a search on Twitter--you'll see some nasty stuff, depending on the search term.

Yik Yak is basically just location-based search, so there's no social filtering. (As opposed to on your Twitter feed, where you create a social filter by selecting the accounts you follow).

"It's no worse than tumblr" is quite the damning with faint praise.

You should know that Yik Yak is an amalgamation of whatever the users say around you... If it's filled with hate speech, that would be your surrounding environments fault and not Yik Yak.

Oh I know, I also tried the "Peek" feature: used for home, work and my university (it was over a decade ago I graduated). It didn't move the needle. Very disturbing.

Btw, I'm in Europe, so all of the comments I saw were in English, as in American English, so I'm guessing us Europeans aren't getting the best of the US. This is not data, just an observation.

> I saw were in English, as in American English

Entirely off-topic, but what English variant is taught in Europe? I'm going to ask a German colleague to spell the word 'colour' I think.

British english. Colour vs color. Lift vs elevator. Single vs one-way. :)

Are you sure? When I was in Germany kids were learning American English. It might vary country to country.

I'm not saying this is 100% the case in every school. The school I went to was British English, for sure other schools teach in American English, as I'm sure there are plenty of Americans teaching around Europe. I'm just talking from experience/observation.

They pick up American english from tv, British english in school. So it's more of a mix ...

I think it depends on the place. And a lot of it is learned outside of school through TV and the internet, which is primarily American English.

Seems to be British vocabulary and spelling, but spoken with an Americanized accent (due to pop culture).

Depends. My teachers (in Germany) all used British English, but allowed both in tests (as long as you stayed somewhat consistent). For people who don't just learn English in school it obviously varies with the different influences and often is a wild mix.

Although the anonymous nature of Yik Yak encourages people to share things they wouldn't normally share in the analog world.

True, but not every area has a lot of users that feel compelled to post things like that, so seeing a lot of hateful content is still a reflection of the people around you.

Those are the long tail of comments. If VCs judged all user-content services by the long tail of submissions, most wouldn't raise any money.

(I realize that might be a good thing, but I presume then that you have the same complaint against Facebook, Twitter, etc.)

https://news.ycombinator.com/item?id=9070088

Still, that just means the long tail is worse, which necessarily happens when anonymity is granted. The best comments-- i.e. the ones that make using the app fun for users-- are also uniquely possible on Yik Yak (or services like it).

It's got to be your area. I'm in NYC and have never seen any hate speech in the few times that I've used it.

If you're near a college area the posts get ridiculous. I deleted it pretty quickly.

I haven't tried Yik Yak as it seems like a black hole of drama and ignorance. However, I think it is being used in interesting ways. Students were able to use it to severely disrupt a frivolous and disruptive class requirement. It's potentially very democratizing, though perhaps it's only bringing out the worst aspects of democracy.

http://emutalk.org/2015/01/a-new-faculty-challenge-fending-o...

In my neighborhood, Yik Yak is nothing but a bunch of proclamations about social justice related topics, and people humble-bragging about their sex lives. Yik Yak is only as good as your neighborhood.

I wonder, do you oppose the funding of hosting companies that provide email services, or allow people to run IRC servers? Do you have a blanket opposition against the funding of services which are used to disseminate user-generated content?

I tried it in SF (fidi and soma). I didn't find offensive stuff, but I'd actually have been happier if it was /b/ level. Instead it was just... lame? Self references to YikYak seemed to get votes. The top ones were old jokes.

I'm 33 so perhaps I just don't get it anymore.

I've literally never had that experience with Yik Yak.

There's a report button if anyone makes an explicit threat; other than that, when you hold a mirror up to society, it's not the mirror's fault what's reflected.

The podcast "Reply All" had an interesting episode about dealing with hate speech in Yik Yak:

http://gimletmedia.com/episode/9-yik-yak/

I like how Colgate University dealt with the problem, but it's unfortunate that it's still there. On the upside, it allowed the conversations about the hate speech to happen in the first place.

Really the thing that is/was cool about Yik Yak is how unfiltered it was. Facebook has all these algorithms to see who sees what and when, but Yik Yak just worked like a localized 4chan with simple voting. This doesn't really help that image (but also come on what an awful way to "prevent spam")

If this gets enough coverage, we'll see a Streisand Effect for the competitors

True - I had heard of Yik Yak, but not Fade.

I went to the yikyak website and viewed their blog and this was the first paragraph from an article written last month.

"When Yik Yak was created it was intended to give everyone an equal voice. No one user would have an advantage over another based on followers or popularity and posts would be judged exclusively by their content."

http://blog.yikyakapp.com/bullying-isnt-cool/

Hi, CEO of Fade here. Just wanted to jump in and add a bit more color. I'm not trying to pretend that my app wouldn't benefit from not having its mentions blocked on a popular platform used by a similar demographic. But I do want to make the case that there are serious ethical questions about a platform removing its users' speech in a way that misleads them into thinking the content was removed by their peers rather than a bot. And Yik Yak's claim that this is an anti-spam tool is simply disingenuous.

Every account on Yik Yak is associated with a unique device ID. And each post can is visible to only people immediately around you. So spamming any meaningful percent of users on Yik Yak would be very difficult. You'd probably need thousands of accounts. If it ever happened, it would to easy for Yik Yak to detect and stop.

Under my direction no one on my team has posted anything at all to Yik Yak in many months. So afaik, 100% of the posts being auto-downvoted with the word "fade" are from ordinary Yik Yak users with no connection to the company. This is exactly the sort of organic "buzz" that the Yik Yak CEO Tyler Droll claims not to want to suppress in the statement he made to TechCrunch.

Yik Yak must know this because of its unique device IDs. Yik Yak can easily see that the posts that mention Fade are coming from regular users rather than spambots. That's why they are downvoting these posts rather than deleting accounts.

Yik Yak's downvote bot is new. Previously they just deleted posts with words like "fade." Why would they do this? In the GigaOM article just published, Yik Yak CEO Tyler Droll claims that the one-downvote-per-minute bot was implemented to "give people a chance to upvote [posts], which would keep them from disappearing." This is flatly contrary with the way their platform works. As the GigaOM article goes on to point out, even popular posts on Yik Yak average less than one upvote per minute, so everything getting a system downvote every minute is quickly deleted; nothing is "given a chance" to stay on the platform. Yik Yak must understand its own platform well enough to know this.

The only plausible reason for Yik Yak to switch from a simple block to a minute-by-minute downvote bot is to obfuscate to its users why their content is disappearing, to make it seem that their peers don't like it rather than a bot removing it.

The 'tool' seems stupid (every minute if post contains 'fade' downvote), but im genuinely curious - are there any rules against this kind of censoring? The article made it sound like this was a bit of scandal, but is there anything more to it than bad form?

Also Fade's notYikYak thing seems pretty juvenile.

There are no "rules" against it. YikYak isn't the federal government, and it has no such obligation to protect your free speech on its platform. If YikYak wanted to soft-censor -- or even hard-censor -- any mentions of the words "potato," or "HBO," or "Facebook," or "America," or what have you, it could do so.

That said, it's a shady thing to do, and while not illegal, it's certainly generating controversy and bad publicity. Time will tell if users vote with their phones against it. That's the recourse here: not the law per se, but the market.

Now, this would be a different story if YikYak had an overwhelming share of the market, and the DoJ were to view something like this as anticompetitive. But YikYak isn't a putative monopolist in the chat space. At least not at the moment.

I posted "Time to fade away unseen" got -2 downvotes exactly at 60 seconds.

The right way to test this is easy: just park your car between major cities and make a similar post. If it is in a place where there have only been a few yaks for the last month, then a robot vote would be the only explanation.

Reply on news.ycombinator.com