The end of tweeted articles | Hacker News Reader

The end of tweeted articles

houseofstrauss.com

162 pointsjger1595 comments

The way I see it, Twitter was one of the last sites standing in the content curation game. Reddit’s great, but balkanised. HN itself is pretty good and broader but Twitter was the best product I used for kicking up stuff from around the web I might actually find worthwhile reading. Now it’s lost that function.

Was it a good business decision? No idea. But it’s made the product worse from this consumer’s perspective.

We're all on goodreads

There's been some recent fuzz on HN about Goodreads too. Is it September over there too?

Has been for what seems like forever.

It's like the fall of Rome. Historians will debate, when did it really start? When did it finally fall? Did it ever fall at all?

It always has been September, always will be, and never was at all; everywhere and nowhere.

Walking is just controlled falling.

Depends on the consumer I suppose. It seems more vibrant, useful, and interesting than ever before at the moment.

Can you give some examples of before and after? Don’t see it myself. I confess I gave up on it once Musk started pushing racists like Tucker Carlson so this may colour my view.

I've been reverting back to RSS and using Pocket as an interim solution (pocket-reader in emacs is really nice and converts well-formed articles into markdown, with the help of pandoc).

I want to host my own RSS server though and then maybe use a native reader to view it, like an RSS of RSS feeds. I don't want to pay an RSS feed company to host that for me and generate ad targeting from it.

The reason is that I want to be in control of the content I consume, and I'm not in control if an engagement algorithm is emotionally manipulating me.

I've been using ReadKit for Mac for several years. It's just an old-school native feed reader app, no services required!

I'm using News Explorer for macOS and iOS. It syncs feeds, articles and read/unread status via iCloud. And for other URLs I use AnyBox which can download them for offline reading (or archiving).

I used to use Pocket, then migrated to Pinboard, and now use Raindrop.

I like Raindrop because it has both golder and tags, plus the folders can be shared.

For example here is a list of tools I have curated over the years, https://raindrop.io/dentropy/tools-31378381

> I want to host my own RSS server though and then maybe use a native reader to view it, like an RSS of RSS feeds.

I've been using Tiny Tiny RSS to do this for years. It works very well. https://tt-rss.org/

I use stringer https://github.com/stringer-rss/stringer

I self host on docker been running reliably for a few years without issue.

I recently abandoned my self-hosted RSS server because I was having a lot of trouble getting past Cloudflare bot checks. I recommend Inoreader as a relatively inexpensive paid solution.

The interesting thing about Xitter was the massive reach it offered over other social networks. So many people on there have Xeeted, "I can't believe this site is free!". I think this lead to other effects like news networks spending half their time covering what people are Xeeting instead of actual news. This no doubt made Xitter as possible as it is today.

I have two big questions about the future of Xitter.

1. Corporate media seems to want to destroy Musk, especially since he bought the social network they invested and depended on so heavily. Does this effectively remove Xitter from the center of public discourse? How will this impact Xitter usage?

2. Everything seems to be trending towards, "pay for reach", which makes it look more like all other social networks. While its certainly possible to build a successful business on this model as Facebook, LinkedIn, and other massive social networks have been done, will Xitter deliver other benefits that are more compelling than competing social networks beyond "lots of reach"?

From my point of view Xitter seems to be going in the direction of being another ad company. Reducing reach, hiring a CEO from the ad industry, and adding paywalls for subscriptions, etc. are all signals that tell me to start looking for a different watering hole.

What will be that next watering hole that gives away an absurd amount of reach in exchange for a relatively modest amount of revenue and corporate media coverage?

> Xitter...Xeeted...Xitter

Could you please please please stop with that it just sounds/reads horrible (and it is also wrong if you talk about the past where people still at least tweeted, and it will also be wrong for the future where people will be Xing on X, if anything, yeah still sounds horrible).

Without pretending the previous Twitter owners were flawless, I do think they had a respect for the "magic" of the site and took care to not damage it too badly. One thing that's been striking about watching Musk talk about Twitter is how...few of his views about what makes the website enjoyable or special are shared by other Twitter veterans. For better or worse Musk is going in a different direction and I think this blog captures one way that's happening.

> Does this effectively remove Xitter from the center of public discourse?

I honestly don't think that Twitter was ever the "center of public discourse". At its peak, it was used by a bit more than 20% of people in the US, far below the likes of Facebook.

There was a streak where it seemed like every other story corporate media (CNN, Foxnews, etc) covered was a Tweet that some prominent person published. That’s what I mean by being at the center of public discourse.

Compare that to Facebook, LinkedIn, etc., which never really achieved that level of press coverage.

LUELinks/ ETI is kind of like Reddit without the balkanization, right?

IIUC, instead of subs you have tags, but tags have moderators and communities around them. Kind of like crossposting, but with a common comment section and no duplication.

I (used to) use Twitter to find interesting things to read and save to Instapaper, but, because of the recent turn down in link virality, Twitter is considerably worse for that. Substack Notes is better for it, but I can't tell if Substack applies a hit to links from other sources (Wordpress, whatever).

X is trying to move content from outside to inside, hence deprioritized link posts, increasing the word limit 100x and sharing revenue. The sharing revenue is 2 fold, one is to give X content creators higher priority and 2 to incentivize content to be created on X.

The rev share has been inconsistent and unreliable though.

Twitter addicts are strange. The attention economy is important, but it is also broadly fractured. Getting loads of clicks all of a sudden never really meant that much.

Yeah the article has the line:

> The attention economy is the real economy

Which is exactly what I would expect a journalist to say... someone addicted to the loads of clicks. Maybe it matters more if getting loads of clicks is your job? Maybe it really is real power? I feel both skeptical and fomo.

I have been a twitter addict since months before i finally subscribed, back in 2013. From time to time I would try to stop but never longer than a few days. Now I went cold turkey with Musk takeover and I don't really miss it at all.

Humans are strange.

Twitter is at best the 15th largest social network, smaller than Quora or Pinterest. Yet because somehow it appeals to the narcissists in the media and celebrities, it gets treated as way more important than it actually deserves to be.

I can't wait for Elon Musk to finish running it fully into the ground.

Same story as with Hollywood, which as an industry used to generate in an entire year only as much money as 2 weeks of the telco's SMS revenues alone. But Hollywood executives are sharks highly adept at getting much more significant tech executives, with stardust envy, to overpay for their marginal businesses.

MySpace and Facebook used to make users enter CAPTCHAs in order to allow them to follow a link and leaves the site with warning banners anyway, so it is part of a social network’s DNA to keep users onsite

Don’t forget about integrated browsers in apps. Gmail, Facebook, and I’m sure a million other apps do this so even when you leave, you’re still in their app.

Twitter will not even allow you to unsubscribe from their emails anymore unless you log into their platform. I'm not sure of the details, but isn't that a violation of the CAN-SPAM act?

You have to wonder what comes next.

The act of killing virality for long form content leads to a reduction in its production.

Then what? Do the deep and important, but not world shattering, stories stop being told?

There was plenty of "long form content" being produced for hundreds of years before the invention of social media. We called them books.

I think long form content will be fine.

Books are one example on the spectrum of types of content here. They're not enough on their own. A book is a poor source for an in-depth look at pending legislation, for example.

Essays, short stories, deep investigative news stories, etc. also all fall into the "long form" bucket. A lot of those had their natural homes in magazines and newspapers, and those have taken huge hits as people's attention has turned to short, viral, online things instead.

Every website looks like TikTok now.

Its hardly surprising that an image would get more engagement than would an article. While an image has less meat to it, by the same token it requires less effort to engage with it. You look at the image, think for a second, then possibly retweet or like. With an article, you can hardly engage with it without following a link, reading some number of paragraphs, reflecting on them, etc. I would consider it quite natural that images and articles get roughly corresponding amounts of engagement. That this was the absolute level of analysis was quite disapointing. Besides trying to evaluate in more depth the viral capability of links today, I would have expected a comparison with days past. I can hardly, having read this article, confidently say that elon is trying to kill external links.

The article seems to miss (or perhaps ignore?) the fact that there has been an incredibly massive exodus from formerly-twitter. There just literally aren’t as many people around to see/share things, which is also a contributing factor to not going viral.

Citation needed

You only need to look up.

https://i.imgur.com/gqmwQmP.jpg

That’s pretty indicative. The blue bird app was always in the top 3.

To me that is only indicative of growth, not decline. People are downloading the app in high numbers. Comparing the downloads of the app across time relative to other apps is unscientific.

Sure, but even if users dropped significantly, they're reporting a drop of multiple magnitudes for the most successful articles. An algorithmic change seems more likely than user activity dropping.

> If your goal is visibility, you are well advised to tailor your messaging to the medium. That means skewing to the visual and away from the linked article.

Isn't that obvious?

Yea I like Strauss but this was a bit strange. He doesn’t seem to understand the profile and iq of the average Twitter user and the type that drives “virality.” Additionally one can immediately engage with an image (which is why every stupid engagement optimizing account would tweet with images), reading an external piece takes more work

One of those filters I want to apply to Mastodon is “images with text”, that is, it is frequently a screenshot of an X (yuck!) or some inflammatory message that I won’t repeat. People do click on them though which is what makes them so annoying.

Haha, there’s that meme about five websites or so that r just screenshots of the other websites.

And yea it enables those stupid posts so I try to refrain from engaging

> He doesn’t seem to understand the profile and iq of the average Twitter user and the type that drives “virality.”

There was still an enormous (and arguably self-sustaining) population of "readers".

No, because it wasn't always true for Twitter. Twitter made a choice to change how it rewarded pictures vs. links, and it happened surprisingly late.

When was this? Sometime in the last couple of years?

I can't confidently say, but the article suggests it's post-Elon.

Ok was just wondering because anecdotally images have been a staple of engagement optimizing posts/accounts. E.g. something like ESPN won’t post a minute clip of an interview or simply a transcript excerpt, it’s always a quote graphic with an animated expression of the athlete. Some fairly mundane quote from Tom Brady will instead include a picture of him yelling

Images/videos were definitely boosted/prioritized, but external links also used to go viral pre-Elon, even a bit after he took over. Remember how mad he got about Substack?

Now all external links are effectively blackholed from the algo feed. It's a massive change.

No, because as recently as just last year, Twitter didn't downrank external links so heavily. Viral articles were possible last year.

Yea I’d imagine the feed change and Twitter fully embracing low effort memes and jokes for its “virality” effect were a big factor

Isn't Mysterious Twitter X curbing "virality" of mainstream media propaganda a good thing?

It’s simple, links are deprioritized in the main home feed, which suggests the home feed is where things go viral now. It’s where Elon gets control, but you should be able to know from the open source algorithm how it is deprioritized

was the algorithm ever released or just the scale analytics code?

TLDR: social media shops are increasingly following the walled garden strategy. From an information searching viewpoint, this is fairly negative as search engines may eventually lose the ability to index the content in these walled gardens - and for some reason(control of what users get to see?), the in-house search capabilities of social media sites tend to be really terrible, see the whole Reddit API fiasco for example that eliminated all the good independent search tools for that site.

As far as X / Substack:

> "Substack this year and last directly targeted Twitter with launches of both a short-form Notes that looked a lot like Twitter as well as a chat feature that moved conversations off social media to its own platform. In response, Twitter stopped allowing users to retweet, like, or reply to tweets with Substack links. It makes sense that Twitter would now take on Substack with a long-form content distribution feature of its own..."

https://techcrunch.com/2023/07/18/twitter-to-support-long-fo...

What an excellent, well thought out piece of writing!! Wonderful. I’m glad I found it on hacker news, and will now repost it onto mastodon. The rest of the social world be damned.

Hmm... virality isn't dying, the idea of measuring it from a single, centralized authority like Twitter is.

The game of any platform is to attract and keep more attention on the platform.

More advanced platforms will be better at this than less developed platforms.

Content that causes users to leave your platform is, from the perspective of the platform, worse content than content that keeps users engaged in your platform.

Articles on X don't go viral as much as they used to on Twitter? That certainly makes conceptual sense because consuming an article involves leaving the platform. If you'd like that same content to perform better on X then "tailor your messaging to the medium" by breaking up the ideas from the article into a series of tweets.

That's the game. If you don't like it then play a different game.

It used to be that the selling prop for a lot of sites was "we'll send you away, but you'll come back every day because we always have a variety of interesting places to send you." In the early days of blogs, The Atlantic would link to other sites, for example, and the Verge is doing it now with their new HP. As social media took over, homepages died, so it didn't make sense to link to a post at blog B from your blog A because no one reads your blog A anymore, they just see some of your Tweets. At this point, the social media giants have changed their algorithms to try to retain more engagement and compete with Tik Tok (which never sends users off platform), but I think in the long run they will fail because there will always be more interesting stuff somewhere else than on your site.

Getting people to play the platforms game is why they end up paying content creators. Twitter recently started paying people per message specifically because they don’t want all monetization to happen off platform.

Pre-Elon Twitter was arguably the last major social media site that actually supported some subset of the "open web" people like to reminisce about here. Articles had the possibility of going viral because the old Twitter didn't downrank external links so heavily, and even encouraged it with the ability to view all quote tweets (another feature X just disabled).

And now it's gone. I think it's fine to mourn that.

> Pre-Elon Twitter was arguably the last major social media site that actually supported some subset of the "open web" people like to reminisce about here.

Unless you disagreed with its politics. In which case you were suspended, or shadowbanned. This has been well covered by Michael Shellenberger and others.

You'd think that piles of concrete evidence would end the denialism, but it seems that people aligned with those politics don't consider stories like the Twitter Files to be real oddly enough. I'm not sure if it's because mainstream media outlets didn't cover it, but there's a correlation.

> Unless you disagreed with its politics. In which case you were suspended, or shadowbanned.

I would appreciate an example about what kinds of political posts Twitter banned. Pre-Elon Twitter did suspend Republicans more than Democrats, but that was because Republicans tended to post more misinformation [1].

Tangentially speaking, Twitter also gave rule-breaking posts regarding the January 6th, 2021 insurrection a bit of leeway [2].

[1] https://www.techdirt.com/2022/04/18/fascinating-new-study-su...

[2] https://www.techdirt.com/2023/06/29/internal-twitter-video-r...

Somewhat ironically for a post talking about misinformation, the first link that you use to support your claim links to a study that says "this study has not yet been published in a peer reviewed journal".

I feel you're looking at twitter from 5-10 years ago and pretend it was exactly like that pre-Elon.

Twitter months pre-Elon was pretty hard on steering the public discussion away from things Twitter didn't want

Ignoring that the major media orgs were still able to post links without being completely blackholed, Substacks and other blogs were still prominent and viral at the time as well. "Steering public discussion" is only partly related, which is why I said "some subset".

Now there is practically nothing on the algo feed that links out of X.

It is in tension with the revenue model that is explicitly about driving people offsite.

Counter point: reddit

They literally send you away to read the article, but with the hope you come back to add comments.

In the case of the new twitter, it makes sense to prevent people from leaving the platform, since Musk wants his platform to replace everything, blogs (hence new unlimited size tweets), YouTube (unlimited length video) and he even talked about a currency.

https://www.theverge.com/2023/7/26/23808796/elon-musks-x-eve...

Do people really read the article? It's a meme that nobody reads the article. Did you read this article? I didn't.

I don’t read most articles. And most articles posted on HN have famously low read rates of probably sub 1%.

I read comments. Comments are the last bastion of pure ideas exchange, articles are just pretty conversation starters.

That was my internet philosophy for a very long time, but for a multitude of reasons I see predominantly noise instead of signal in comment areas I used to enjoy and few better alternatives. I’m now trying to counteract my habits and focus more on the information.

This is a very dated view of Reddit.

I just visited Reddit in incognito mode.

Of the top 20 posts, 3 were links to articles. Most were images, videos, or text posts -- very similar to Twitter.

The way to use reddit is to use https://old.reddit.com AND unsubscribe from everything + subscribe to the specific subreddits you are interested in.

That's the power user route but we're address the business model of reddit which is slowly inhousing all content.

Counter-counter point: for years Reddit has been pushing for fewer article links and more locally hosted content.

They started hosting their own images, then their own videos. This is expensive to do, but perhaps they felt that it was worth it to keep users on their site instead of risking them wandering off to Imgur.

And the new design discourages actually reading the article. When you click on a "link", it sends you to the comment page which has a small embedded thumbnail and blurb. (The old design took you to the article directly).

Counter-counter-counter point: Reddit reached its heyday under the leadership that allowed people to leave and come back to discuss. Its current leadership has intentionally driven away a decent percentage of users within the last year.

I do not think their decisions come from sound thinking but greed. They don't truly understand why they got big and now they're trying to squeeze all the money out of it they can before their luck runs out.

I think they did much better trying to give their users a good experience and they are in the process of killing the goose that laid the golden eggs. It just takes a long time for a zombie social network to die.

> Counter-counter-counter point

(tangent) Does this remind anyone else of Melissa Peterson from Seattle, WA?

[0] https://www.wnyc.org/story/1421-my-dog-hates-you-too/

Yeah, there are a larger number of communities that outright ban link posts now, and others which have a pretty strict guideline against any articles (labeling them as 'blogspam' or 'self-promotion' even when something is not... it's a tough job though because spam is incredibly frequent now).

Note that most popular content on reddit is images, short video clips or plaintext; and the majority of that is hosted on their servers.

I've been enjoying watching the app Artifact progress - https://artifact.news

It's been a news aggregation app so far but with some nice features (article summary, read aloud, reader mode).

They just launched the ability to submit your own links. I have not tried it but it sounds like it may bring back things people liked about Delicious, Google Reader, old Reddit.

I mention this as I personally enjoy this platform as a way to consume content quickly. The article summary is very similar to Hacker News comments, I view that first & see if I want to read/listen the actual article.

This article is a lot of words about the fact that twitter, or should I say X, is promoting tweets, er posts, that keep one on X (posts without a link in them, the example given in the article is posts with pictures) instead posts that send one away from X (posts that consist of a link promoting something on an external website, which used to have the ability to go viral greatly helping the owner/author of what's on the external website).

The article is roughly 1,500 words.. If that is a lot to you, I guess it just means you probably like twitter these days!

I dunno, I also found it sort of rambling and not getting to the point of its thesis.

The article also agrees that 1,500 words is a lot, seeing as they are the leeches who say things like optimize for the picture rather than the content and the attention economy.

Really, this is a bunch of whinging from people who used to get free promotion from others work, now they can't abuse that trust anymore.

As a user, I'd much prefer a concise tweet instead of a huge article with the same exact point. So, I guess Musk's Twitter actually improves my experience.

What if it’s impossible to present that point in a post on X?

E.g. because it’s too complex, or because the author doesn’t post on X, or for some other reason.

Some people just want to live in a world of sound-bites and hot-takes.

They're also the sorts of people who believe most books should be replaced by a paragraph-long summary.

Unfortunately now you can pay to have your tweets be huge-article sized.

Outside of the fact that a platform will always selfishly want to prioritize its own content and keep users on its platform, I feel there is a growing sentiment from web users as well that most content-based websites are just _bad_, and platforms are adapting to this behavior.

Over the past few years, how many times have you:

- Searched Google for something like "Best Coffee Maker" / "STORE coupon codes" / "Best cocktail recipie" and the article was just an SEO'd affiliate link blogspam?

- Searched Twitter for something innocuous only for the top links to be some sort of crypto or OnlyFans garbage?

- Clicked into a website just for 1-3 visual modals to popup to "Disable your Adblock" / "Sign up for our mailing list" / "Login to view the rest of the post" (And we thought popups were dead once popup blockers became a thing)

This article is even an example -- I scroll down just to be interrupted by a call to action modal requesting my email and making me find the "continue reading" button just to read the content.

Platforms provide content consistency -- once users understand the platform, they can know how to consume content on it.

It's perhaps the death of the open web, but I kind of get it.

> This article is even an example -- I scroll down just to be interrupted by a call to action modal requesting my email and making me find the "continue reading" button just to read the content.

That substack popup drove me insane until I had enough and just turned js off for them. Very pleasant now

Reply on news.ycombinator.com