Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Snap’s Drop in Active Users Could Signal a Social Media Peak

nytimes.com

Post reply on HN
Print view View original

141–150 of 300 posts

Re: Snap’s Drop in Active Users Could Signal a Social Media Peak

#141

Post by DJHenk »

quotemstr wrote: post #128
florabuzzword wrote: post #118

Earlier quoted context omitted.

I bailed at the advent of forced non-chronological timelines. I would have been less insulted if Facebook hired someone to spit in my face. Easily the scummiest design decision I have ever whitnessed. Actively manipulating the structural representation of time in a communication platform is begging for chaos. Yes, of course I know the time stamps are still presented. That is not my point. Communication is sensitive;…

Chronological is just one ordering. Why are you so attached to it? Why must deviating from this one specific ranking approach lead to "chaos"? Algorithmic ranking is not evil. A tiny minority of people rail against algorithmic ranking, but the vast majority, as demonstrated by their engagement , prefer the algorithmic model. Should we let this tiny vocal minority dictate policy, data be damned?

Algorithmic ranking might not be evil per se, but it is if the algorithm is hidden and there is no way to adjust it, or to check what it filtered out.

Add to that a perverse incentive (money from ads and 'engagement') for the company designing the algorithm. How can that not lead to evil?

Re: Snap’s Drop in Active Users Could Signal a Social Media Peak

#142

Post by sullyj3 »

mygo wrote: post #113
danepowell wrote: post #54

Earlier quoted context omitted.

Sounds like an example of Goodhart's Law: "When a metric becomes a target, it ceases to be a good metric." Optimizing for something as abstruse as "engagement" seems like a great way to completely ignore everything that actually makes social networks compelling. https://en.wikipedia.org/wiki/Goodhart%27s_law

they’re just optimizing for the wrong metric. If the metric was “user satisfaction” they could target it day and night and nobody would be complaining.

The issue is that something like this difficult to operationalize. User Engagement conflated with user satisfaction, with poor results. Maybe you could force your users to fill out Likert scales all the time, I bet they'd love that.

Re: Snap’s Drop in Active Users Could Signal a Social Media Peak

#143

Post by andyroid »

quotemstr wrote: post #128
florabuzzword wrote: post #118

Earlier quoted context omitted.

I bailed at the advent of forced non-chronological timelines. I would have been less insulted if Facebook hired someone to spit in my face. Easily the scummiest design decision I have ever whitnessed. Actively manipulating the structural representation of time in a communication platform is begging for chaos. Yes, of course I know the time stamps are still presented. That is not my point. Communication is sensitive;…

Chronological is just one ordering. Why are you so attached to it? Why must deviating from this one specific ranking approach lead to "chaos"? Algorithmic ranking is not evil. A tiny minority of people rail against algorithmic ranking, but the vast majority, as demonstrated by their engagement , prefer the algorithmic model. Should we let this tiny vocal minority dictate policy, data be damned?

Yeah, having to refresh my instagram feed 2-3 times just to make sure I didn’t miss anything from my friends certainly has me engaged. But it’s the exact kind of “math-friendly refrigerator magnet junkfood” engagement GP was talking about.

Re: Snap’s Drop in Active Users Could Signal a Social Media Peak

#144

Post by fragmede »

ravenstine wrote: post #82
leon_sbt wrote: post #69

I really wish social media would "peak" to towards meeting people in real life semi-spontaneously. No profiles, no media uploads, no digital showboating. The best relationships I have with other people fall in this category. Real life,common interests,no digital uploads. Just add attributes of your geolocation and hobbies [biking, yoga,tensorflow,etc] or let it crawl your digital profile(s) to autobuild your attribut…

We seem to be on the same wavelength. I might be giving up a viable idea here, but for a while I've been thinking about a sort of hybrid between Meetup and online multiplayer matchmaking where groups of people can meet up and do stuff without the need for an actual organizer. Here's how I envision it: - A user joins and provides a materials used to perform a background check. - People with accounts sign in and are au…

It's a good idea, but with too much emphasis on the technical side, and too little on the how-people-actually-interact side. (think XKCD #538)

No one's going to use QR codes, and people are already shitty and shallow; you're fooling yourself if you think an app's going to stop them.

There were a couple group-dating apps that were tried out for a short bit; Grouper is the one I remember from 2016 but there were others.

Re: Snap’s Drop in Active Users Could Signal a Social Media Peak

#145

Post by Vinnl »

leon_sbt wrote: post #69

I really wish social media would "peak" to towards meeting people in real life semi-spontaneously. No profiles, no media uploads, no digital showboating. The best relationships I have with other people fall in this category. Real life,common interests,no digital uploads. Just add attributes of your geolocation and hobbies [biking, yoga,tensorflow,etc] or let it crawl your digital profile(s) to autobuild your attribut…

Heh, this makes me think of this anti-viral startup idea I had - which I'm not going to make because it's... Anti-viral and thus not something I think I could pull off. Anyway, the idea was pretty funny: a meetup-like website where you'd sign in with e.g. Facebook, and then you could only see and attend meetups that none of your Facebook friends are attending. In other words: everybody goes to these meetups with the expectation of not knowing anybody and getting to know new people.

That said, I think there's only so many people interested in actually meeting new people, at least of a certain age, and even those who are, are only interested in getting to know a small amount of new people.

Still, I quite like the idea. I don't even think you need to group people based on interests (though of course the subject of a group allows people to self-select); just being open to meeting new people can probably go a long way.

Re: Snap’s Drop in Active Users Could Signal a Social Media Peak

#146

Post by florabuzzword »

quotemstr wrote: post #128
florabuzzword wrote: post #118

Earlier quoted context omitted.

I bailed at the advent of forced non-chronological timelines. I would have been less insulted if Facebook hired someone to spit in my face. Easily the scummiest design decision I have ever whitnessed. Actively manipulating the structural representation of time in a communication platform is begging for chaos. Yes, of course I know the time stamps are still presented. That is not my point. Communication is sensitive;…

Chronological is just one ordering. Why are you so attached to it? Why must deviating from this one specific ranking approach lead to "chaos"? Algorithmic ranking is not evil. A tiny minority of people rail against algorithmic ranking, but the vast majority, as demonstrated by their engagement , prefer the algorithmic model. Should we let this tiny vocal minority dictate policy, data be damned?

> Chronological is just one ordering. Why are you so attached to it?

This is a very deep question you’re asking, and I don’t know the answer. But, this boat only goes in one direction so let’s make the best of it.

But seriously, sure other sorting methods are useful. I would like to have many options, but I can’t think of a better default for this use case. Can you?

> ...as demonstrated by their engagement, prefer the algorithmic model.

You just don’t understand my point. Maybe read it again. I dunno. But ...

As demonstrated by their engagement, people prefer doing whatever their boss tells them to do.

As demonstrated by their engagement, people don’t actually like sex very much at all.

Re: Snap’s Drop in Active Users Could Signal a Social Media Peak

#147

Post by Vinnl »

krisroadruck wrote: post #28
askafriend wrote: post #13

Earlier quoted context omitted.

To add to this, there's also Nextdoor, Discord, YouTube, Telegram, Life360 etc depending on how you define a social network. I'm pretty sure all of these are growing significantly both in the US and abroad. Not to mention all the Asian social networks that are growing like crazy. My theory is that Snap just doesn't have enough differentiation either as a product or as an audience to be sustainable as an independent e…

Genuine question... how is whatsapp a social website vs just a chat application? Perhaps I've missed something critical about it but the way I've used it and seen anyone else use it is just as a replacement for SMS

As someone who doesn't use WhatsApp in a country where everybody does: the primary use case appears to be group chat. If it was just a replacement for SMS, not having it would not be a problem - I could still text with people. However, I'm "missing out" on things because I can't be in some WhatsApp chat groups.

Hence the network.

Re: Snap’s Drop in Active Users Could Signal a Social Media Peak

#148

Post by cyphar »

lsalvatore wrote: post #134
cyphar wrote: post #89

Earlier quoted context omitted.

This is a ridiculous idea, that falls apart with even cursory scrutiny. For a trivial example, almost all of Shakespeare's plays were amalgamations of other people's plays or were based on other people's stories (Hamlet, for instance, was a retelling of a 16th century version of the 13th century Scandinavian legend of Amleth). Would you argue that Shakespeare was not creative? Or is it possible that stories are based…

You're not going to convince me that someone narrating over themselves play Half-Life has the same creativity as Shakespeare, because "everything is a copy". To say that no one actually owns the rights to Pokemon, that someone copying Pokemon frame-by-frame is of equal creative value as someone drawing their own monsters, well I just am deeply saddened by that.

(I'll be honest that I didn't originally didn't read that the topic under discussion was modding of games. Passing off a game as your own is obviously not right, but that isn't the only topic under discussion here.)

Of course modding is not the same as Shakespeare. My point was that stories like those of Shakespeare would not have been possible with our modern notions of copyright. I do agree that many (but not all) examples of such work are just simply rip-offs of other people's work. On the other hand, does it make sense for any work to be untouchable for an essentially unlimited period of time?

So while you might argue that mods are all shameless copies of another person's work, I direct you to 'The Stanley Parable'[1] which was a mod of Half-Life 2. I think most people would argue that it is a creative work that is unique despite the fact that it re-uses assets from a copyrighted work. There are many counter-examples to the idea that the only way to have creative value in a game is to "draw your own monsters". Jim Sterling ran a contest where people were to come up with original works based on an overused Unity asset, and the winning entry was a game made entirely out of textures based on that asset[2]. Is that not a creative work?

I think that ultimately it's a disservice to argue that all such works should not be allowed because many examples of people building on previous work is lazy or a transparent rip-off. That the requirement for a creative work be a strict sense of originality (not to mention that you need to decide how do you define sufficient originality -- is "Harry Potter Told Using Pokemon" an original work?).

[1]: https://en.wikipedia.org/wiki/The_Stanley_Parable [2]: https://www.youtube.com/watch?v=jTL6NlPF0Lw

Re: Snap’s Drop in Active Users Could Signal a Social Media Peak

#149

Post by CamperBob2 »

egfx wrote: post #117
CamperBob2 wrote: post #76

Earlier quoted context omitted.

This does seem like a powerful idea. The problem of networking friends, family, and co-workers has more or less been solved, but the problem of putting together complete strangers with shared interests and coincident physical locality really hasn't even been attacked except in very constrained use cases (i.e., dating services). It's possible that the next big wave of social networks will be less "Facebook" and more "…

The problem of putting together complete strangers with shared interests is solved by Reddit.

No more so than it was solved by Usenet, though, or by BBSes before that.

Re: Snap’s Drop in Active Users Could Signal a Social Media Peak

#150

Post by _wmd »

florabuzzword wrote: post #146
quotemstr wrote: post #128

Earlier quoted context omitted.

Chronological is just one ordering. Why are you so attached to it? Why must deviating from this one specific ranking approach lead to "chaos"? Algorithmic ranking is not evil. A tiny minority of people rail against algorithmic ranking, but the vast majority, as demonstrated by their engagement , prefer the algorithmic model. Should we let this tiny vocal minority dictate policy, data be damned?

> Chronological is just one ordering. Why are you so attached to it? This is a very deep question you’re asking, and I don’t know the answer. But, this boat only goes in one direction so let’s make the best of it. But seriously, sure other sorting methods are useful. I would like to have many options, but I can’t think of a better default for this use case. Can you? > ...as demonstrated by their engagement, prefer th…

I'd offer you one blindingly obvious explanation for why it's putrid: because it is a machine deciding which of my human relationships I should pay the most attention to.

Machines can't do that, silly. They can only promote hate speech with 100 comments and lolcat pictures with 100 likes. The only option is for them to take a back seat and present an ordering that is semantically meaningful to the human computer operating the tool -- the only one qualified to perform any kind of filtering and sorting.

"Relevance ranking" in social networks is a result of engineers having zero clue what applications they're working on, but that's just typical IT people divorced from reality for you. As you said, they find it acceptable because it's written in Angular v7 and uses ML and A/B testing revealed a 3.2% increase in clicks. The idea of the software industry helping real relationships is in general hilariously contradictory (but of course there are exceptions!)

Post reply on HN