Hacker News Reader

Comment by abeppu | Hacker News Reader

I love this as an idea, but I suspect as a user, I would use either zero or one posts.

I do like the idea that the platform can actively disrupt the "addictive" patterns that develop elsewhere. Other things I've wanted:

- Instagram with an ML layer that auto-rejects pics with faces or text. Landscapes, vistas, animals, architecture etc all would be welcome.

- high latency Twitter, where no post is viewable until at least 2 weeks after it's published. Bickering threads become impractical. People learn to post stuff that will be worth caring about later.

- Clearly just for entertainment and not information Facebook alternative, in which GPT bots produce a significant fraction of posts, impersonating users and making stuff up. Everyone quickly learns you can't trust it, but it can still be cute/fun/humorous.

View on HN

I really like like the idea of a "high latency" social network, effectively bringing back the feeling of old snail mail (but more public).

2 weeks seems unnecessarily extreme to me though, even 24h would probably be enough to severely limit the flame potential. It would also potentially mean that one could have a daily routine of checking for the new content and replies and writing your own stuff and you're good to go for the day.

Although to really be effective I think it'd have to work on a global tick (i.e., all new content is published once a day at a certain time). Otherwise the new content would still slowly drip continuously and you'd still have the addictive nature of social networks.

At this point I'm sort of reinventing a collaborative version of a newspaper.

I like the "collaborative-newspaper" idea a lot.

Though I haven't used it much, this sounds like substack + RSS, that only refreshes once a day.

It's missing some interactivity I suppose

Thus usenet was reinvented

I bought newsto.me like a decade ago with a plan to build a collaborative news site complete with gamification to really help the collaborative bit (eg one user writes a bit about a house fire happening, a photog in the area could earn points for snapping a photo and adding it to the story)

Never found sufficient time/motivation to execute on it. Still think it would be kinda cool though.

I’ve considered something similar, call it The Morning Post or some such. Choose your delivery/send time, and all your content gets posted (and all previously posted content gets revealed to you). Put some sane limit on the number of times you can change your post/delivery time (once a day?).

Is this not email?

I really hope to be able to report back in about a year or so on ways to better communicate. At first I thought, maybe the best online dialectics is just walls of careful text rather than tweets. Now I try doing that, and I make careless mistakes and I can barely parse a paragraph without losing my train of thought.

The solution I think is explicit structure. Arguments usually form a (for now let's say) DAG, let's present them that way explicitly. Metadata, i.e. human emotion can be annotated onto that - emoji, good or bad vibes? If you can diff that graph even better.

Argument Maps are kind of a non-technological approach to this: https://fbistudies.com/wp-content/uploads/2015/04/20150417_R... really impressed me.

In fact, working in a fully-remote, fairly high information density (finance) company, I think the work of the pioneer cyberneticists (and potentially Soviet ones, not sure on that as I can't speak Russian) could be immensely helpful for modelling and manipulating the spread of information and humanity within a company.

No, it's not email.

Email has a latency measured in seconds. Also, email is not usually public.

Explicit structure could be interesting, but structure from whose perspective? Arguments can be modeled many ways, such that there is no one correct answer. Also, that the argument could be modeled as a DAG might be very optimistic in many circumstances.

I picked 2 weeks because I wanted to discourage talking about the news. Or responding/reacting to something dumb a public figure did or said.

Even just 24 hours would do a lot to disrupt the echo chamber.

2 weeks would match up with certain snail mail geography especially around the horse/carriage era.

It might be fun/interesting to try to base the latency upon simulated geography: whether the entirely fictional geometry of something like classic GeoCities where you get assigned a random address in a digital "neighborhood" somewhere when you sign up, or use real world geography and filter it through ancient routing restrictions like "how long would it take in the post by horse/carriage".

A benefit to a dynamic latency such as simulated geography would be that it would make it even tougher to sync together how posts would be arriving from multiple users (unless you were targeting a specific user).

I'd love a high latency app as well, but there is undeniable magic in instant communication. My best writing is often when I'm "hot" on a topic and can quickly go back and forth with others on the idea. Steam could quickly run out for that type of thing without instant communication.

Continuing to brainstorm--

Say, the way some messaging platforms have threads which can branch off of the main channel. Maybe this platform has something like that, too; instantaneous and temporary, where messages disappear N minutes after they're sent. So you can post publicly on the time-delayed cadence, or chat semi-privately in realtime.

I'm not sure if that undermines the desirable properties of the time delay.

I wonder how much sense it would make for there to be some (simple) function you can tweak a number on to set a balance between the latency and lifetime of your message?

For example the lifetime of your message (counting from when it shows up) could be 2 times the latency of it, e.g. a message with a latency of 1 day would exist for 2 days.

It's not necessary to inject intentional latency in the manner the parent suggests via a rather extreme delay.

You can accomplish a lot of the same outcome by merely eradicating following feeds (that is, the combined stream of posts from users that you follow).

In the early days of Facebook when you had to manually visit user pages to see what was new with someone or what they were posting, it was dramatically lower velocity in terms of how fast bullshit would spread and how much content you could reasonably consume. It would reduce page views / content views spectacularly, by including just that little bit of time and labor cost to the consumption. People would focus that effort primarily on the users that really matter to them.

The social networks of course know this. They built the streams / feeds / walls for that reason, to turbocharge the addiction. They'll never give it up.

I currently use Twitter this way. I stay logged out (deleted the password from my password manager) and visit Twitter.com/username directly for a few folks I want to read. It’s indeed way less addicting.

(Twitter really wants you to log in and view your feed but they’ll still let you read the site this way)

Yet, the possible resistance is also on the consuming side. It requires a lot of willpower not to flow with the river, but tools like Fraidycat (https://fraidyc.at/) help a bit in following people not trends.

Regarding high latency twitter: maybe someone should make FidoNet.social. For those who don't remember, back in the early 90's your local BBS's might offer message boards connected to the rest of the US/world. Not a continuous connection. Every day at 2 AM (or whenever), your favorite BBS would dial into some remote node and sync messages.

Practical effect was a roughly 24-48 (or more) hour wait for responses. Didn't stop bickering though. I was quite young at the time. Asked some question about a game. The first response was fairly hostile. Which started an argument. My first time flamed online, and my first online argument. Which went very slowly.

Still, slow social media would be interesting. I kinda like the FidoNet model. Where syncing only happens once a day. Maybe only at a set time overnight. Faster than snailmail but you have the full day to type out your response about why Ultima V did not suck as your opponent claimed. With the ability to submit a post any time, but also with the ability to edit it until sync. I think it would encourage long posting more than the current systems. Which may or may not be a good thing.

I've always wondered about the practicality of doing something like fidonet over AX25/some other packet radio system. It'd be fabulous to be able to ditch the internet and participate in something slower and more humane. Your post has reminded me of that ambition. These days, I'd love an e-ink display to accompany it. Slow computing!

With AREDN, you get a whole TCP/IP ecosystem to play with. This could then be used to run protocols like UUCP or NNTP, that are tailored to disconnected scenarios.

https://arednmesh.readthedocs.io/

FidoNet didn't really mandate syncing once per day. It required the nodes to sync at least that often, and established a common hour for each zone to allow direct node-to-node connections for that purpose; but even in late 90s, larger nodes would already sync more often in practice.

FWIW FidoNet is still around, although most connections seem to be over IP these days.

Let me take this chance to mention the "Unhook" Youtube browser extension, which I recently added and which I think has been having a big positive effect for me. The extension allows you to remove (among other things, it's very customizable) the sidebar of suggested videos from Youtube.

This extension has been more effective for me than any other at cutting down how much youtube I watch. Maybe the biggest factor in this is that it doesn't ban me from youtube entirely - When I've tried extensions like that in the past, I've always ended up uninstalling the extension when I needed to watch a youtube video for work. This way, I can watch a little, but the lack of constant new recommendations keeps me from spending hours and hours on the site.

Seconded, but just a note that if you're using ublock origin you can just block the sidebar with a content filter.

I've blocked the sidebar and the comments section. The former has dramatically reduced the amount of time I spend on YouTube. The latter has dramatically reduced my exposure to the inanity of the YouTube comments section. Both are huge quality of life improvements.

For the last one: https://www.reddit.com/r/SubSimulatorGPT2/

Usenet used to be like this for me. With high cost dialup, I'd download batches of posts, read them and respond offline, uploading my responses at the same time I download the following day's batch.

What about a low-bandwidth Twitter, where posts load immediately, but at 2e-3 bits per second, so that it takes two weeks to load the whole thing.

Call it "Loris"

The combination of your ideas would be Mars rover Twitter: 5-20 minutes propagation delay, 160-800 bps bandwidth using the X-Band High-Gain Antenna - https://mars.nasa.gov/mars2020/spacecraft/rover/communicatio...

For the twitter latency

Have you tried slowly.app? It's an app that simulates snail mail Pairs you with someone based on your preferences and you can write them a letter / email that will be delivered in a few hours / days, based on your distance with said person

Was really into slowly.app for a while, and pleasantly surprised to see it suggested here.

I would recommend it to anyone looking to experiment with a "high latency social network".

The latency idea actually makes a lot of sense

I emulate it by deactivating as many notifications as possible, and manually checking replies after a few days or weeks, when I remember. I have noticed that what was a "hot" discussion often becomes pointless and laughable after enough time has elapsed. I usually don't feel the need to reply to these old bickerings anymore.

>- Clearly just for entertainment and not information Facebook alternative, in which GPT bots produce a significant fraction of posts, impersonating users and making stuff up. Everyone quickly learns you can't trust it, but it can still be cute/fun/humorous.

Aren't we pretty much there now with a so many bots posting?

I think a bunch of people still credulously believe most of what they read. But if some days the bot actively impersonated _you_, and people reacted to stuff it said on your behalf, I think you'd have to grok pretty quickly that a lot of content is fake. So let's do that for every user, unpredictably, a reasonable percentage of the time.

scorched earth, why not? I think this is just about right though. Humans were not ready for social media, so burn it to the ground. however, if what the bot says on my behalf brings disrupte to my name, no repurcursions should be bestowed on me.

> Clearly just for entertainment and not information Facebook alternative, in which GPT bots produce a significant fraction of posts, impersonating users and making stuff up. Everyone quickly learns you can't trust it, but it can still be cute/fun/humorous.

Billy Chasen developed faux-social network called Botnet. Unfortunately, it looks like it's gone and I doubt that name would have lasted long in the App store let alone how he managed to get it through.

https://www.wired.com/story/botnet-social-network-where-ever...

> high latency Twitter

This seems easy enough for current twitter-fediverse-clones to implement.

+1 for the high latency idea. I recently witnessed a really bad screaming match on an NGO mailing list between a bunch of anti-vaxxers and, for lack of a better term, the reasonable people. Emails were flying at a rate of 1 every few minutes for long into the nights. What ended up solving it was appointing a moderator that has to approve all the posts beffore they go out. They don't actually block any, they just have to press the approve button. This meant that it usually takes anywhere from half a day to 2 days for a post to go through, so people started thinking a lot more about what they posted and what they read. It became not only slower, but a more civilised and though-out discussion.

So, Facebook? :-)

The trouble with making a worse option is that people still have choice and will simply use the more entertaining version instead. You can't just make a new social network that's not as engaging as the existing ones because nobody would use it. The moment you're really curious about the thing being hidden from you in the degraded service, you'll sneak a look at the real one to find out. If you had so much self control that you don't, then why do you even need a service for it? Just limit the number of hours you spend on social media yourself.

In this case the community would be a self-selecting group of people with self control and a value for more thoughtful communication. The resulting content would presumably be different from what you'd get if you just viewed mainstream social media on a restricted schedule.

Reply on news.ycombinator.com