If people are using 3rd party apps because the 1st party apps suck, why don't Reddit just acquire the 3rd party apps. The 3rd party developers get an exit, and the users get to keep the experiences they like.
It's not accidental at all that the first party apps suck. Any banner that says "it is better with our mobile app" should have a steaming pile of poo emoji next to it.
The whole point of the new reddit (and many other sites like Anandtech and Tony Ortega's blog) is to make you click on ads accidentally. Someday the world will wake up and realize this is not just annoying but it is good old fashioned fraud... but it will still take a while.
One advantage of a mobile app over a mobile web site is that you can use the accelerometer and other sensors to predict when somebody is going to click on something and you can trigger a layout shift and... oops!
> One advantage of a mobile app over a mobile web site is that you can use the accelerometer
Accelerometer API access has been available on mobile web for ~ a decade. I think it'd actually be more trouble than it's worth to listen for a "tilt" to shift the view and misdirect the user's tap. Most UI's just put the ad dangerously close to an element that's also interactive, or style the ad to look nearly identical to the other content in the feed, like reddit. It's to make you encounter ads and maybe tap on them purposefully or accidentally. They will happily take either.
In a mobile browser, this would also be not as easy as in an app because the permission would have to be requested on-demand (or on page load), if I am not mistaken.
Me neither, it just worked in chrome on my pixel 6. And in Brave it doesn't work and does not ask for permission. Both still ask for permission for using geolocation API.
Disagreed on your last paragraph. The average website is constantly rescaffolding. Because of ads, lazy loading, shitty frameworks, whatever. I'm constantly clicking on ads and unintended buttons/elements because of it. It's literally why mobile newspaper websites are unusable.
So if the goal was to get clicks on ads, they'd be pushing you hard to mobile web.
The real reason is the app can get much more interesting data from you.
I think both can be correct, rescaffolding probably was the original reason for stray clicks but once identified, I'm sure profitable issues may be fixed while plausible deniability remains.
Not disagreeing with you, I can smell the incentives for this sort of behaviour, but:
> you can use the accelerometer and other sensors to predict when somebody is going to click on something and you can trigger a layout shift and... oops!
Is there anything out in production that’s provably doing this? That’s pretty aggravating if so!
It’s happened to me a few times… I think? But hard to know if it’s just “ads load slowly and I decided what to tap on equally slowly”.
This has been a very frustrating experience with new YouTube which now hides comments in the vertical layout. I used to pay for Premium, so it’s some random recommended video that I constantly accidentally click, it’s very frustrating, esp when it doesn’t save my location in an hours long video. It only clicked for me that it’s actually for adverts when I saw someone else’s app.
Now, instead of paying Google $10 a month for a better experience, as I was happy to do, I’ve just spent a few bucks on Vinegar and donate the $10/mo to a couple of creators I like on Patreon. I’ve been successfully converted from an extremely happy paying customer to an annoyed leech.
Great point. I downloaded the official Reddit app just to see if the negativity surrounding it was warranted. As soon as I tried to scroll past my first ad, a tap was registered and the ad website loaded. It was suspiciously erroneous in its handling of my scrolling up.
The problem is that BigCos destroy the 3rd party apps that they acquire. The same thing happened after Twitter acquired Tweetie.
Indie devs and BigCos have vastly different incentives. Simply acquiring an app won't help, because the app will no longer be developed and managed by an indie dev. Wreckage is inevitable.
not that I particularly disagree with your premise, but alien blue had been somewhat abandonware for quite a while before reddit bought it. It had trouble with basic markdown like tables, let alone new features that made it increasingly less great before reddit bought them out.
In all this news, I completely forgot that I used to use Alien Blue before Reddit acquired it. That was actually what made me try out the official reddit app for a while, before switching to Apollo.
Because reddit is running out of money, and the investors have stopped giving any more. At least, that is what I infer from their direct statement of "Reddit must be self-sustainable". I have no inside info to know for sure.
So buying apps would just shorten their runway even further. And even if they did, they would just be bringing in more users of the same type they have today - users who are unwilling to pay for reddit accounts.
They need to figure out what their business model is, not just buy apps that keep them running down the same path they have been on for years without profitability.
The third parties are incentivized to deliver a good user experience. Reddit has incentives that inevitably lead to the opposite. Any good app under Reddit's control would cease to be a good app.
It's not an accident that Twitter just went through this too. We're in a new era of too-big-to-be-disrupted enshittening.
That would solve user issues, but nothing for Reddit the company.
If have seen versions of this serveral times, and it assumes that what is good for users is good for Reddit. It is like asking "if people like free lunch, why dont all restruants serve free lunch. Imagine all the customers they would have."
In reality, there is a trade-off between happy customers and customer revenue.
Many of these 3rd party apps are made by single developers, some are open source. Reddit could clone or recreate them easily if they wanted.
reddit (and big tech in general) doesn't want users to have experiences that users like, they want users to have experiences that advertisers like. Ideally there's some overlap there (otherwise why would you use a service) but usually companies making money from advertising lean towards the ideal experience of the people who give them money, ie, not third party app users
They're not necessarily trying to do the optimal long term thing right now. They're making short term moves to boost their valuation for the IPO. Doing things like showing an increase in ad revenue is a nice boost for that, no matter how it's accomplished.
I believe the API changes are not about third party apps or LLMs (at least directly). I think it's all about bots and spam. They only want to do business with real enterprises through their API. This raises the value of purchasing real ads through their platform vs buying upvotes and other subversive ads on grey markets. Once the API access is restricted I think there will be a huge dropoff in spam and bot accounts. You'll have to scrape endpoints and fight the built in anti spam detection on the app/website in order to game the system versus using the free highly available API that has fewer built in protections.
I think they've done everything to steer the conversation away from the topic of bots, spam, and grey market ads. If reddit is selling ad space but you can get it cheaper via vote bots, why would anyone buy ad space through reddit?
Well the Reddit CEO lies so I wouldn't trust anything he said, but here is what he said about 2 months ago w.r.t. AI and Reddit: https://archive.ph/m3KBt
Being a source for LLM's gives the platform relevance. Reddit are happier for openAi to make trillions off their data than for a 3rd party reader to make 0.99c
The whole point of the new reddit (and many other sites like Anandtech and Tony Ortega's blog) is to make you click on ads accidentally. Someday the world will wake up and realize this is not just annoying but it is good old fashioned fraud... but it will still take a while.
One advantage of a mobile app over a mobile web site is that you can use the accelerometer and other sensors to predict when somebody is going to click on something and you can trigger a layout shift and... oops!