Demetrification: Removing the numbers of likes and retweets from public view | Hacker News Reader

Demetrification: Removing the numbers of likes and retweets from public view

onezero.medium.com

55 pointspaps39 comments

Nice idea, but I think this misses the reality of what is happening. Instagram is removing likes because it incentivises an economic system that they are not only left out of, but is a net cost to them.

Top influencers make a lot of money based on those metrics, Instagram doesn't see a cent of that yet they support these "businesses". Does anyone consider what it costs to distribute a post to 10s of millions of people? As an analogy, consider a system where advertisers paid the actor and the film crew to produce an ad, which was then distributed for free by the television studio. Not going to happen as it's completely unsustainable. Instagram is increasing the monetisation of the platform, and allowing advertisers free direct access to the metrics means you can't sell it.

Once it's gone, expect Instagram to roll out a new system where influencers will need to pay for potential reach and Instagram will start making a lot of money.

It's a cost to them that more people are coming to the platform and viewing ads? I don't get it.

> Instagram is removing likes because it incentivises an economic system

Crackpot theories aside, likes are just a stupid way to look at the world, the bigger scandal is how long it took all these supposedly smart people to figure that out.

It's a loss of potential revenue that advertisers and content creators can make advertising deals directly, without depending on instagram as a middleman.

I guess Instagram follows the opposite of Bill Gates advice on platforms -- strive to capture 100% of all the value created on the "platform".

They will surely offer "promoted content" and claim it to be popular. Honestly, while I don't like point/karma systems very much, I think the development to even more artificially placed content will not be in favor of platforms in the long term.

Advertisers encroach on every platform and it really sucks. Youtube for example didn't really develop anything worthwhile for users. 5 years ago the platform was so much more interesting. It still is, but navigating content was made really difficult with "adjusted" search results.

I’m glad YouTube didn’t develop a lot. Right now it does the job, supports everything I want as a user and mostly stays out of my way.

Sites that constantly evolve and try to reinvent themselves annoy me way more.

They still adjust the algorithm all the time and that tends to piss off a lot of people, too.

That alone would probably be okay if they weren't so quick to bend to pressure from a handful of squeaky wheels who happen to think of themselves as competitors to YouTube. ;- )

> expect Instagram to roll out a new system where influencers will need to pay for potential reach and Instagram will start making a lot of money.

This makes sense from the perspective of the relationship between, say, Kim Kardashian and Instagram. More reach is directly valuable to some people.

But it kind of falls down when you think about the ordinary customer base. A lot of people will want to follow Kim Kardashian. They may not have any other reason to use Instagram at all. But when they try to follow her account, it doesn't look good to show the message "you can't see Kim Kardashian's pictures, because 200,000 other people have already signed up to look at them. Try asking Kim to upgrade her account! (But no, you can't message her here.)"

Summary -- charging popular accounts for potential reach, as opposed to charging advertisers for pushed advertisements, can have strong negative effects on your user base, despite the fact that those users never wanted the reach. They did want their favorite popular accounts to be able to reach them.

>you can't see Kim Kardashian's pictures, because 200,000 other people have already signed up to look at them. Try asking Kim to upgrade her account! (But no, you can't message her here.)

That is now how it works. People subscribe to more than one content producer which allows the aggregator the power to merge and reorder the individual streams. You do not pay for being displayed, you pay for being prioritized in the stream.

Related to this, I remember when HN was showing the amount of upvotes on comments (like it does on posts).

I still miss it, because quickly scanning a very thick amount of content, I'd like to know where I should spend time. I do it with posts (and the list /best is great if you're short on time).

The ordering still supposedly shows us the very recent followed by the 'solid comments', but there is 0 information about how 'solid' that comment is by consensus.

That being said, my favourite part of HN is the [-], so when I'm done with someone's opinion and their 300 responses, I can just skip it.

Funny thing: I still don't know what [-] means, there are no tooltips or any other hints so I prefer not to click it.

It's just a simple collapse thread feature.

Not so simple; it persists the collapses somehow (I think server side attached to your account rather than your session.) That’s a lot more complexity than the equivalent Reddit feature, for a change.

Hm... I always thought it has to do with the comments that get slowly grayed out? E.g., for me that was basically a "downvote".

There is a downvote button, but you need some minimum karma for it to be available to you.

I can understand why it might be confusing if there's no downvote arrow visible!

Yeah, that some kind of UX pardigm - don't hide the control. Disable it, and explain the user why he cannot use it at the moment.

On the other hand, though, I can now simply collapse huge branches without downvoting anyone, so thanks for the clarification! :)

[-] just hides that branch of the comment tree so that you can ignore it if you want to focus on reading other branches.

You've been here since 2012! :)

That's really interesting.

This is so stupid and pointless, all because a few people that over-obsess about essentially inconsequential numbers don't understand how rash generalization is a logical fallacy, so they assume their experiences must be everyone else's experiences.

> 1.5K claps

I think part of the incentive of using a platform is building up a following to leverage ad revenue yourself. It's currently a win-win. Popular people on social media can get sponsorships, the platform gets to show people ads from their network. Platforms are getting the content for free. If they remove the ability of users to negotiate sponsorships, they'll need to start paying for the content.

Title got the key name wrong, it's demetrication.

Thanks. I was wondering what Demetri had to do with it...

Thats my first name which was what drew me to this thread!

This is just throwing the baby out with the bathwater - I'll have even less idea why this or that comment is at the top of the feed, it will be just as gameable for the committed bad actors, and I won't be able to get a quick estimate of whether a social media post is worth engaging with or a crowd has already gathered and established some rough consensus about it.

It's good to keep experimenting with the development of better UIs though - seems to be plenty of demand for that on Twitter today as people react coolly to their new UI.

This seems like a nice improvement.

I don't think the user's experience as a 'nobody' is quite comparable to as a celebrity. I'd waste much more of my time worrying about social media metrics if I got more attention.

Though I also hope it doesn't exasperate competition anxiety, if people just presume those they're competing with are more popular than they are. (My understanding is Instagram tends to be very unhealthy for adolescents).

Even though I left social networks long time ago I welcome this change. Even if I tried really hard not to get influenced I still ended up reading posts with a high number of [interaction] somehow differently.

IMHO HN comments or some subreddits are good examples following this principle (not exactly the same environment though) while maintaining some sort of ranking.

I worry about this as a part of a larger trend of more proprietary and enforced ignorance in service of vested interests instead of transparency in service of all who can look and think.

The article touches on this at the end, which I think is the most important part:

"For social media titans, full demetrication would require a more radical abandonment of faith in data, and a disavowal of the numbers-driven “growth mindset” that has powered Silicon Valley for so long."

I'm conflicted on this change. On the one hand, we can see how visible like and follower counts incentivises wrong behaviour on the user side (let's leave aside ethical concerns about "nudging" users to the "correct" usage).

But on the other, these data points are still there for the company. They still know how many likes, and followers you have. They randomize your timeline to increase engagement. So this is not a radical change for them, they are still going to use all the data points they can to push you in the direction they want. The main thing this changes is how transparent they are about this. For Instagram, this means pushing you to use stories more, because it's what's driving engagement now.

Also, it's worth noting that for these companies, metrics were a vital part to convince investors to buy into the companies: they could point to them and say: "see? we're growing, people are engaging!". Now that they've made it, they can afford to tune it down.

The New York Times ran an article about this back in May[0]:

"[T]oday, what you see on Twitter and Instagram already depends on a mixture of signals — things you’ve liked in the past, how much time you’ve spent looking at a particular user’s content, whether you communicate privately with a given user and whether you have an affinity for some topic or another — not just chronology, likes or retweets. Those signals are all metrics too, of a sort, invisible to us but very much legible to the platforms themselves. Imagine a ticker in your Instagram app counting up the number of times you’ve scrolled, or tallying the number of times you’ve tapped, or counting up the seconds you’ve spent looking at an image. These already exist, somewhere, and may inform what you see every day. They’re just not for you to know.

Understood this way, the idea that metrics are the problem sounds an awful lot like these companies saying their users can no longer be trusted, not even with the scraps of actionable data they’ve been allowed to see for years."

[0]: https://www.nytimes.com/2019/05/31/style/are-likes-and-follo...

Metrics with hard numbers are always too precise to be useful. Even the well-meaning will micromanage based off the difference of a 5 or 6 on a scale of 50.

I find the key is metric obfuscation. Instead of "1023 likes", color code the post. A deep crimson for the most downvoted and a sky blue for the most upvoted on a site. The specific value is so useless anyway.

Which significant sites use that method?

How do you handle the high range of values? In one part of the site 1000 likes may mean a very high number while in other, a high profile could regularly see millions.

Good point, but the overall system could be adjusted any number of ways. The point is to communicate to the user, not throw numbers at them and let them decide what it means.

I actually particularly dislike this, transparent reporting of metrics gives users the ability to have their own interpretation. Instead of guessing why platform Y gave post Z a deep crimson color, which you'd probably not be able to do because of 'algorithms'.

Most people using metrics aren't great at interpreting them on their own in the first place. Look at Chile's notorious misuse of waiter ranking (you fill out a survey after your meal). If a server gets less than 5 stars, their job is on the line. Nevermind the fact that each customer will use the 5 star system differently to communicate different feelings. Some people will never give a 5 star rating.

If it involves people other than you, your metrics likely don't mean what you think they mean.

Mastodon does this. It’s really nice.

So’s the non-engagement-optimized chronological timeline, of course.

In a decentralized network, you don't really have a choice. Such metrics can only be imposed with central login and "unique identity" policies.

Facebook, Twitter and Instagram all have shadowy businesses to buy likes, followers and whatnot. Now just imagine how that would turn out in a decentralized setting where any script can create many accounts on several different servers of the fediverse.

Big corps have to show the metrics because their "engagement" model relies on those number to keep people addicted. On the fediverse, there is no incentive to make people addicts to your service, and showing the numbers would just encourage the same kind of fake-fame industry as on those big networks.

Reply on news.ycombinator.com