HNHacker News
TopNewBestAskShowJobs

lonk11

395 karma · joined January 9, 2021

My hobby project that this account is for: https://linklonk.com

Email is "lonk" at the above domain.

submissionscomments
lonk11··on Show HN: RSS Brain
A simple algorithm is to sort content by the signal-to-noise ratio (SNR) of the source feed.

Basically, the more you "liked" past content from a given feed - the more priority is given to new content from that feed.

A more detailed description of the algorithm I use: https://linklonk.com/item/3292763817660940288

As I understand the OP system does not collect/use your content ratings, only external upvote counts. But this is something I think they should try.

lonk11··on It's time for an RSS revival (2018)
You can get an RSS feed of posts of any Mastodon user. Just add ".rss" to the profile url. For example, mstdn.social/@feditips -> mstdn.social/@feditips.rss

Details: https://mstdn.social/@feditips/108357998963885456

The conceptual difference between RSS and Fediverse is that RSS is read-only while federated protocols are read-write. For example, you can add comments through ActivityPub, but not through RSS.

lonk11··on This is the year of the RSS reader?
The process you are describing can be automated a bit.

Imagine your feed reader automatically subscribes you to feeds of articles you liked (star/thumbs up/favorite). Let's say you start subscribed to an HN feed of items that got more than 100 upvotes: https://hnrss.org/newest?points=100&count=100 Then you find the OP link and like it in your feed reader: https://www.niemanlab.org/2022/12/this-is-the-year-of-the-rs... Your reader automatically subscribes you to the source feed of this article: https://feeds.feedburner.com/NiemanJournalismLab

You like more content => you get subscribed to more feeds automatically.

Your feed reader can also take care of unsubscribing you from feeds that are posting too often or are not posting interesting to you content. If you don't like the content of a feed for awhile then your connection to that feed becomes weaker and the content from that feed is ranked less prominently. The content from feeds with a high signal-to-noise ratio (ie, how often you like content from that feed) is ranked higher for you.

This is how my little project https://linklonk.com works. If you'd like to give it a try: "thumb up" a few items from the above HN feed: https://linklonk.com/feed?url=https:%2F%2Fhnrss.org%2Fnewest... and then check your recommendations on the home page.

Through this mechanism I am currently subscribed to 201 RSS feeds and I know that I won't miss content from high "signal-to-noise ratio" feeds. Here is a sample of my top feeds:

https://blog.piekniewski.info/feed/ https://bam.kalzumeus.com/archive/rss/ https://karpathy.github.io/feed.xml https://blog.cryptographyengineering.com/feed/ https://www.inference.vc/rss/ https://openai.com/blog/rss/ https://stratechery.com/feed/ https://medium.com/feed/@yonatanzunger https://rachelbythebay.com/w/atom.xml https://openai.com/blog/rss/ https://randomascii.wordpress.com/feed/ https://computer.rip/rss.xml https://timharford.com/feed/ https://idlewords.com/index.xml https://www.youtube.com/feeds/videos.xml?channel_id=UC9-y-6c... - Computerphile videos https://www.granolashotgun.com/granolashotguncom?format=rss https://daringfireball.net/feeds/main

lonk11··on Protect me from what I want
What you are describing is similar to how my hobby project works: https://linklonk.com

When you upvote content (+1) you make your connection stronger to those who posted that content and also those who also upvoted it. When you downvote (-1), your connection to the poster and other upvoters becomes weaker. The content in your feed is ranked base on how strongly you are connected to those who upvoted the content.

For example, I upvoted the OP post on LinkLonk: https://linklonk.com/item/3406294698351951872. If you upvote that item then you will get connected to me and to two RSS feeds that posted that content: the blog's feed (https://www.tbray.org/ongoing/ongoing.atom) and the the HN's feed of newest items (https://news.ycombinator.com/rss). As a result you will start to see other content from those RSS feeds and from my "main" channel more prominently in your recommendations.

Users can post/submit links under different user channels. So if your hypothetical E2E person were to post E2E stuff under their "E2E" channel while keeping the furry stuff under their "furry" channel then you would only get connected to their E2E channels if you only +1'd content from that channel.

lonk11··on Ooh.directory
LinkLonk supports comment notifications for any item that you have subscribed to. You get notified through browser notifications and email. And you can mute/follow individual branches of the comment tree. Details: https://linklonk.com/item/396040801387544576

There is definitely more to improve, but I'd like to see more use of the comments section first to know what needs to be improved next.

As for ActivityPub, I just saw https://go-fed.org/ on HN and it looks like a good fit to add ActivityPub support to LinkLonk (since it is written in Go). There are multiple ways to do it, so let me know how you would like to see it working. Do you want to follow posts of Mastodon users on LinkLonk (similar to how it shows you RSS content)? Do you want Mastodon users to be able to follow your channels on LinkLonk (e.g. @s3000@linklonk.com)?

lonk11··on The New Smoking (2021)
The way I am trying to solve this is by building a different kind of content feed. One that is not optimized for "time-spent-on-site" but for "will you be happy to have spent time on it".

This is how https://linklonk.com works: you upvote content that you found useful and the system gives more priority to users and RSS feeds that posted this content in your future content recommendations. As a result you see content from those who have brought to you useful information in the past - a feedback loop that is better aligned with your needs.

lonk11··on Ask HN: What are your “scratch own itch” projects?
I'm building https://linklonk.com - a news aggregator that prioritizes content from sources (RSS and other user channels) that you upvoted before.

The server part is written in Golang (sqlc, pgx, gorilla/mux) and I think server development is a great fit to learn Golang.

I want to expand the idea of personalized ranking ("others also liked") from content feed to group decision making. Basically, let a group member delegate their vote to other members that voted like them in the past.

lonk11··on Ex-Reddit CEO on Twitter moderation
This also seems to be gameable. Suppose you have blue and green camps as described in the linked paper. And if content gets ranked high when it gets approval from both blue and green users then one of the camps may decide to promote their opinion by purposefully negatively engaging with the opposite content in order to bury it.

This seems no different from "popularity based" ranking mechanisms (e.g. Reddit) where the downvote functionality can be used to suppress other content.

Maybe the assumption is that both camps will be abusing the negative interactions? But you can always abuse more.

How is such system protected from someone manipulating the consensus by employing a troll farm (https://en.wikipedia.org/wiki/Troll_farm)?

lonk11··on Ask HN: What is something you built but never marketed?
I built a social rss reader that ranks content based on how much you liked other content posted by the same feed or user: https://linklonk.com

I haven't marketed it much except for a Show HN. I spent $20 on Google Adwords to see how effective it is. The result was a few clicks but no new signups. I'm curious how other folks in a similar situation have marketed their side project? What channels have you found useful?

lonk11··on Things I don’t want to do to grow my side project
I used Material for my hobby project [1] and got this feedback:

"Maybe I'm the only one, but I don't like the UI. It seems like a demo for a Material design app. I think the site has lots of potential, but without an original, personal style it might put off users. It doesn't have to be complicated, just something recognizable."

So a standard component library can get you started quickly, but you might want to tweak it. A few things I changed - made icon buttons smaller (the default sizes are way too big), changed the pills to be white with black border instead of just grey background.

[1] - https://linklonk.com

lonk11··on Stabilization of gamma sulfur enables 4000 cycle Li-S batteries [pdf]
"TLDR is that with just a new anode made from carbon nanofibers and sulfur"

The new material they discuss is for the cathode side (not anode). That's side where the lithium ions go when the battery is charged. The cathode of traditional Li-ion batteries is made of metal oxides: Cobalt, Nickel, Manganese (lately trying to get rid of the expensive Cobalt).

The sulfur cathode promises to remove all of the expensive metals in the cathode and make the cathode much more energy dense.

The anode is where the lithium ions go when the battery is being discharged. In traditional batteries the anode is made of carbon fibers (burnt coconut shells). The holy grail of anode is pure lithium anode - there is nothing more energy dense. But the problem with pure lithium is that it tends to get deposited not in an even layer, but in the form of dendrites that eventually poke through the dielectric separator creating a short. In this work they explored lithium anode, but as I understand that was not the main focus. I don't think they solved the dendrites formation problem.

lonk11··on China has pioneered a law to empower people over algorithms?
"Art 6: “… actively disseminate positive energy, and promote the application of algorithms to be good.”"

It depends on who it is "positive" and "good" for. For the society, the government, or the user?

I would like to see algorithms that focus on the user. For that you want to have algorithms that are understandable and controllable by the user.

Unfortunately, this is not compatible with the algorithms employed for recommendations today. They use neural networks that are hard to understand not only to the users but to the developers of these systems.

Look at the abuse prevention technique used by DALL-E 2 [1]: "Our content policy does not allow users to generate violent, adult, or political content, among other categories. We won’t generate images if our filters identify text prompts and image uploads that may violate our policies."

I worry that the algorithm regulation will focus on forcing recommendation systems to use similar controls and then those who control the filters of what is "not good" will effectively control what is acceptable to learn and talk about.

I'd like to propose an alternative - design the recommendation algorithm that gives control to the user to decide what is good for them. I'm building https://linklonk.com as an experiment in this area and the algorithm is detailed here: https://linklonk.com/item/3292763817660940288

[1] - https://openai.com/dall-e-2/

lonk11··on Ask HN: What is your “I don't care if this succeeds” project?
Indeed, RSS is a great minimal API and we need more tools that can create and consume feeds. For example, you can create a feed in siftrss.com and submit it to LinkLonk so it could rank content from feeds you follow based on their signal-to-noise ratio.

Regarding the OP, I'm building LinkLonk to test out a different mechanism for cultivating trust among strangers and for discovering interesting stuff along the way. I wouldn't say I don't care whether it gets attention. That's the whole point of the experiment - to find out if it is a viable alternative to the AI driven feed. I am prepared though to maintain it for years and not looking to "fail fast".

lonk11··on The web starts on page four
Yes, that's the focus right now. Make it a good tool and then make it a good network. The functionality that supports single-player mode is:

- bookmarking - you can search your past liked items

- subscribing to rss feeds

Please let me know if you have any other suggestions along those lines.

lonk11··on The web starts on page four
Your assumption is correct, user retention is difficult. I've been running LinkLonk for a year now and out of >1000 users that created accounts we have ~9 daily active users on the site and 43 users are getting recommendations by email. But I think this is in part the chicken and egg problem. The more users join the more useful the system should theoretically become.

If you don't want to create an account, consider creating a guest account - it does not require any personal information from you. The account is just a number that is issued to you and is stored in your browser cookies. If you don't end up using it - all information associated with your guest account will be deleted after 30 days of inactivity.

If you don't want to create a guest account then the best I can do is to show generally popular content - which is what the home page is filled with for brand new users.

lonk11··on The web starts on page four
LinkLonk can also send new link recommendations by email.

There is also https://refind.com

lonk11··on The web starts on page four
At the end the author talks about curated link collections as a solution - create them using Pinboard.in and consume them via RSS.

I think the part that is missing is how to discover the link collections to subscribe to.

I'm building https://linklonk.com that merges the collection, the consumption and the discovery:

- You add links to content you liked to your "channels".

- LinkLonk connects you to RSS feeds and other user channels that posted the content you liked and shows you new content from them. The more content you have in common with a feed/channel - the more weight their new content gets in the "For you" list.

You can also share your channel via a public link (example: https://linklonk.com/channels/programming-yUlhkwDXFb) or an RSS feed (example: https://linklonk.com/api/publicChannels/programming-yUlhkwDX...) .

lonk11··on Ask HN: Who wants to help promote RSS?
Hello future! Glad you like it. Do you have any suggestions on how it could work better for you?
lonk11··on Twitter Home made me miserable
Yes, this is a very natural conclusion to make. We all have definitely seen people upvoting content as a way to promote their opinion, even if the content is not accurate or an outright lie. This happens on Facebook, Twitter, Reddit. If people behave this way on LinkLonk then we can imagine how it would connect people with the same opinions into echo chambers.

But I would argue that this behaviour is the result of the incentive systems. Our likes/retweets/upvotes don't have much effect on us. The recipients of these actions are other people - they affect what other people see. And when we are given tools that can influence other people it should no surprise when we use them to do just that. The content that is good at influencing (we are clearly right and they are clearly wrong) is often sensationalized and misleading. The means justify the ends. And the other side lies too.

On LinkLonk the incentives are different and therefore the behaviour would be different. Your upvotes have a limited power to affect what other people see. Instead, your upvotes primarily determine the type of content your future self will see. I believe that you would be more likely to upvote content that informed you, than content that simply says how right your side is (which usually has little new information to you).

I commented some more on this topic here: https://news.ycombinator.com/item?id=28834278

lonk11··on Twitter Home made me miserable
How about a recommendation algorithm that works like this:

- When you upvote an item, the algorithm connects you to other people who upvoted it and shows you what else they upvoted ("more like this")

- When you downvote - the algorithm disconnects you from those who upvoted it and you see less content from them ("less like this")

Some properties:

- You have an easy way to expand your connections - just upvote/submit content you liked.

- Because it is so simple, the algorithm can explain to you exactly why you are seeing a given recommendation - because it was liked by people who liked X, Y and Z (where X, Y and Z are things that you liked before).

- Since your connections are a result of your explicit content ratings - you don't have to worry about poisoning your recommendations by content you merely viewed. You have active control over your recommendations.

If that sounds interesting, check out my hobby project that works exactly this way: https://linklonk.com - it's free, no ads, no tracking.

lonk11··on Ask HN: Who wants to help promote RSS?
You are describing the project I'm building: https://linklonk.com

You connect to RSS feeds when you "thumb up" or directly submit a link that this feed posted. The more of the content you like from the same feed - the more other content from that feed will be prioritized. So feeds that post too frequently naturally go down in the ranking. And feeds that have a high signal-to-noise ratio are at the top.

On LinkLonk I am subscribed to 202 RSS feeds (you can see your subscriptions here https://linklonk.com/feeds), and I don't get overwhelmed by too much content.

You also connect to people who like the same content as you did. In that case each user is similar to an RSS feed (a user has "channels"). The more in common you have with a user - the more content you see from them.

No obnoxious adverts, no trackers and it's free. Would love to get any feedback (email in profile).

lonk11··on Ask HN: How do you get articles/papers/talks recommendations?
Let me plug my hobby project: https://linklonk.com - it is a collaborative RSS reader and I did a Show HN recently for it: https://news.ycombinator.com/item?id=29445764

I think it may strike the balance you are looking for between: 1. following popular content, 2. subscribing to individual sources and 3. getting personalized algorithmic recommendations.

From the Show HN post:

"LinkLonk is a novel mechanism to subscribe to RSS feeds and discover content - upvote or submit a link to anything you liked and you will get connected to RSS feeds that posted this content. The more content you upvote from the same feed - the higher other content from that feed will show up in the For You page. This helps you see content from feeds with the highest signal-to-noise ratio first.

In addition to RSS feeds, you connect to other users who upvoted the same content as you. This way other users help highlight great content from feeds you are already connected to and discover other feeds that they are connected to.

To sum up: upvote content => connect to RSS feeds and users => discover great new content => repeat."

lonk11··on Ask HN: What do you read for general knowledge?
Good to see that the newsletter works! Let me know if you have any improvement suggestions for the newsletter.
lonk11··on Show HN: Collaborative RSS Reader
When I find an article that is particularly insightful or informative I often wonder who are the other readers of this article, how did find it and what else did they find interesting? LinkLonk is a way to connect me to the other readers.

I guess it could be trivialized to the good old "Other people who bought this also bought that" but for articles. But that was not the source of inspiration.

The algorithm itself is inspired by Personalized PageRank (which Pinterest used to use 3 years ago: https://blog.acolyer.org/2018/05/23/pixie-a-system-for-recom...). Essentially, you do a walk on the graph that has sources of information (RSS feeds, user channels) as nodes and upvotes as directed edges. The walk starts at you => into one of the items you upvoted => into one of the sources that upvoted it before you => into one of the other items that source upvoted. You sum up all the probabilities to end up in each item after these 3 steps - and that gives you the ranking score of the item.

lonk11··on Show HN: Collaborative RSS Reader
From the FAQs on https://linklonk.com/about: "No, at the moment I don't have plans to open-source it."
lonk11··on Show HN: Collaborative RSS Reader
Thank you! I'm focusing to make it a better tool first to make it useful in "single player" mode before we have enough users for network effects to kick in. Please let me know if you have any suggestions on how to improve the single player experience.

By the way, you can group your upvoted content into channels. For example, if you are interested in tech news and food, you can create a channel for each interest. And then people who like you "food" upvotes will see your other "food" upvotes but not your "tech news" upvotes and vice versa. This would create more coherent groupings.

More on channels in this FAQ post: https://linklonk.com/item/2369676162410512384

lonk11··on Show HN: Collaborative RSS Reader
The popularity based recommendations are only shown to brand new users - so they have some content to rate.

A new user starts off weakly connected to all other users and to feeds the existing users are connected to.

But as you submit or upvote links to content you liked - your recommendations will be based on your explicit upvotes, and will be less popularity based.

> aren't you recreating some of the same issues with social media?

1. The algorithm.

Social media algorithms are black box ML models that optimize for time-spent on the platform (ie, ads watched). LinkLonk also uses an algorithm, but there is 0% of machine learning. All it does is keeps track of how useful the past recommendations of every RSS feed and every user have been to you. You define what is useful to you based on what you upvote. This makes it so that each user can make decisions of what they want to optimize for. Social media algorithms make these decisions for you.

2. The types of actions

In social media we have two types of actions:

- Inbound: follow/subscribe - they affect what I see as a user. The target of these actions is myself in the future.

- Outbound: comments/likes/dislikes/retweets/etc - they affect what other people see. The target - are other people.

The problem is largely with the outbound actions - social media give us tools to influence other people and we use them for that. In other words, they give us megaphones and we each crank up the volume of our own megaphone to the max. There is little incentive to not set the volume any lower than the max.

I understand your perspective - traditional RSS readers don't have any outbound actions. You only decide for yourself who you trust to get information from.

And your concern is that LinkLonk provides tools to influence other people and that these tools will be inevitably abused.

What makes me hopeful about LinkLonk is that it does not have separate outbound and inbound actions. There is only an upvote which is both inbound (determines what your future self sees) and outbound (it affects what other people see). So you may want to be more careful about what you upvote - because you don't want your future self to receive useless stuff. For example, maybe you would be less likely to upvote an article without reading it purely based on the title that says "Look how terrible the other side is".

3. Proof of work

The amount of attention someone gets from you is directly proportional to how useful their past recommendations have been to you. In other words, to get your attention they need to prove to be good curators of content for you. They cannot fake it by upvoting the most popular content. That’s because you connect only to those users who have upvoted the content you like before you did it - ie, before the content became popular.

lonk11··on Show HN: Collaborative RSS Reader
The page with your personalized recommendations does not have an RSS feed itself. The recommendations are prioritized based on how strongly you are connected to sources that posted it - RSS feeds and users. This prioritized list does not have an obvious way to be translated to a chronological RSS feed.

If such RSS feed includes content from all sources that you are connected to - then your RSS feed would have too much noise. The only way I can think of converting it to an RSS feed is if you choose how many top recommendations you would like to see in that feed per day.

This is how the email newsletter works today (you can subscribe here: https://linklonk.com/profile) - you choose when (daily/weekly and at what time) and how many top recommendation you want to delivered to your email inbox. And you can rate items from the email.

> Can it be seeded with known liked feeds or content?

Yes, you can seed your recommendations - simply submit (https://linklonk.com/submit) links to content that those feeds posted or directly submit links to those feeds. If you use RSS readers then you can the list of feeds you are subscribed to as an OPML file. I haven't built an automatic way to import OPML files, but if you send your OPML file to me by email, I would be happy to import it for you.

> And can you 'pin' liked feeds in some way, to say 'I always want to see content from here'?

To soft-pin a feed that you liked you need to upvote content from that feed. That will cause your connection to the feed to be stronger than to others and items from that feed will show up at the top for you.

There is no explicit pinning functionality - something that would effectively set the connection weight to infinity. The value proposition of LinkLonk is that it keeps track of how useful the content from each feed and user has been for you and prioritizes new content accordingly. I know this is very different from a traditional way of subscribing to RSS feeds, where you are either subscribed or not (boolean). On LinkLonk your subscription has a weight (float). In the days of Google Reader, there was something similar I think if you sorted your content "by magic". Only on LinkLonk, there is no magic - the mechanism that connects you to feeds and users is transparent and controlled by your upvotes.

> Is this a reader, or another source? It is a different kind of reader - one that automatically manages your subscriptions.

lonk11··on Show HN: Collaborative RSS Reader
I appreciate this and I understand your concern that a project that does not have a sustainable income is likely to be stop working one day.

At this scale the project doesn't cost much to run. The best motivator for me to develop it further would be your feedback and any help spreading the word to bring in more users. After all, in order for the "collaborative" effects to kick in we need more users.

lonk11··on Show HN: Collaborative RSS Reader
LinkLonk is a novel mechanism to subscribe to RSS feeds and discover content - upvote or submit a link to anything you liked and you will get connected to RSS feeds that posted this content. The more content you upvote from the same feed - the higher other content from that feed will show up in the For You page. This helps you see content from feeds with the highest signal-to-noise ratio first.

In addition to RSS feeds, you connect to other users who upvoted the same content as you. This way other users help highlight great content from feeds you are already connected to and discover other feeds that they are connected to.

To sum up: upvote content => connect to RSS feeds and users => discover great new content => repeat.

This is my hobby project that I am building in my spare time. My stack is: PostgreSQL + Golang server (sqlc, pgx, gorilla/mux) + Angular client + Firebase for auth. It runs on a small VPS instance on OVH that costs ~$12/month. I am planning to run it for years to come and I will be happy to have you as a user.

I did a Show HN for linklonk.com 3 months ago (https://news.ycombinator.com/item?id=28405643), but I think I over complicated the title and the description, as exemplified by this comment I received: https://linklonk.com/item/8281929870071398400?comment=173480.... That’s why I’m reposting it.

← PreviousPage 2 of 4Next →