It's also the main way I use HN - I probably click into more HN threads than news articles these days.
I also self-host Nitter (Twitter front-end), Bibliogram (Instagram front-end), and Teddit (Reddit front-end). Each of these server applications allow me to subscribe to social media feeds without having accounts.
Additionally, I self-host RSSHub for a ton of other integrations, but mostly Youtube subscriptions. Using RSSHub for Youtube subscriptions is nice because I can watch ad-free Youtube videos in Reeder when content creators drop new videos.
EDIT: on desktop (Linux, MacOS) I use Newsboat with FreshRSS.
Did you know, Invidious has an RSS feed for your subscriptions? So not only do you not need a YouTube account but you also get private, ad-free viewing alongside that feed!
My biggest complaint about teddit and libreddit is lack of an account. All subscriptions are stored in cookies, so it’s a pain if I want to browse the same content on my phone and desktop. I am actually considering a bit of a stupid simple/quick hack where I could visit a site on my network which would redirect to the URL that restores your subscriptions/settings cookie.
I did not know that! For a Youtube front-end I use Cloudtube[1] and NewLeaf[2] (NewLeaf is a drop-in replacement for Invidious). I'll have to see if NewLeaf supports RSS feeds like Invidious because then I could remove RSSHub from my cluster. Thanks for the tip!
> All subscriptions are stored in cookies, so it’s a pain if I want to browse the same content on my phone and desktop.
Wouldn't using an RSS feed server like FreshRSS mitigate this cross-device problem? I.E. you mark something as read on one device it will sync between devices. I ask because initially I was confused about your description because I don't have the same issue but then after I used my Big Brain® I realized FreshRSS solves this problem for me.
EDIT: Silly me, I guess I requested RSS feeds feature in NewLeaf already[3] last year
[1] https://git.sr.ht/~cadence/cloudtube
Seconded.
My personal setup is a tt-rss instance set up on my personal server, with a Fever plugin, which allows me to use it with other readers—currently ReadKit on my Mac, and Reeder on iOS. I very rarely access my tt-rss webpage directly; it's just there to provide centralized-under-my-control syncing of the feeds and their read status.
I use NewsBlur. It has a free setting, but I like it so much I pay for the subscription.
(available on chrome too)
Interacting with it, sure--at a paid service level it's been easy to monitor service updates this way, but not so much at a personal reading level currently. Though you got me thinking...
Last time I was using a reader, I remember struggling to find a Linux desktop RSS reader with discovery features. I have my own OPML that I like & cart around, but it requires a lot of maintenance over time, and using feeds for their current value seems to subvert the fact that a lot of really great blogs are past their heyday (the good stuff is found by going back in time, not so much now & later).
I also thought it'd be cool if there could be a P2P effect with sharing OPML or RSS feeds, like a general catalog. Or let's say a Gutenberg.org mixed with Github for RSS. Find, explore, & remix others' shared feeds. Not sure if such a thing exists.
Almost some random thoughts here, not sure if any of this helps...
I really hope RSS will be kept alive in the future. Feeds are my daily curated source of news, updates to youtube channels etc.
RSS is the only way I subscribe to any news source and I don't think I'm missing on anything.
Some are news and blogs but also sub-Reddits, Google News searches, and Kijiji searches. Until a couple of months ago I had some eBay searches but eBay killed their RSS feeds. I don't use it for YouTube but they do support RSS.
The bulk of them are feeds with infrequent posts - software release updates and that sort of thing. I try and be pretty ruthless with culling anything with a high noise to signal ratio of posts that interest me.
As a developer, I mostly use Django, so I use the included syndication framework. The major pain point in that library is that adding additional tags that it doesn’t already support, especially if they are complex, is a pain in the ass. of course, you only have to do it once and then you can re-use your subclass in other projects. It’s still not great. Ideally it would support a very wide range of popular tags out of the box, i.e.: the podcast ones.
If a news source does not support RSS itself and there is no workaround to get it otherwise as RSS, I will not subscribe to it.
For reading, I use Feedbro in Firefox and Flym on Android.
I use it to read HN and various blogs & Ars Technica, and to sync my read/unread status between my phone, ipad, and computer (I use NetNewsWire and Feedbin).
I publish a blog that has an RSS feed, although very few RSS subscribers.
I use a service to convert my blog's RSS feed to a daily newsletter for subscribers (who have no idea there is RSS involved). I have around 100x as many newsletter subscribers as RSS subscribers.
I use a different service to automatically link new blog posts on Twitter & LinkedIn whenever they appear on my blog's RSS feed.
I use my RSS/feed reader pretty extensively as a central hub to collect "news" from a variety of sources, including from closed platforms such as YouTube via their semi-hidden RSS feed support. I like having a central place where things that may be of interest to me show up without me having to discover them actively or having to rely e.g. on the YouTube subscription interface.
I personally really like the concept of Microsub, because it allows you to have a single designated server takes care of fetching the feeds, normalizing the content and synchronizing state (e.g. read status, subscriptions, channel/folder organization) and you can then access that data from any client that supports the protocol. I specifically like Microsub, as opposed to some of the other "synchronization protocols" that are more or less supported by some services because it is an open protocol independent from any specific platform/implementation.
[0]: https://indieweb.org/Microsub-spec [1]: https://github.com/aaronpk/aperture
But as it turns out, Thunderbird actually has a built in feed reader, which now nicely aggregates both my e-mails, spam e-mails and both news that I might care about as well as the ones I don't. So the experience is very similar to dealing with e-mails - occasional items that I care about amongst others that I don't, even if some of the others are useful as well, and having all of that in the same application makes a lot of sense!
Here's an example: https://blog.thunderbird.net/2022/05/thunderbird-rss-feeds-g...
I even added an RSS feed to my own blog which can now tell anyone who cares about that stuff about new articles.
In addition, I'm subscribed to some feeds that pull data from HN so I don't miss like 90% of my comment responses, as well as a "Jobs" feed for the "Who's Hiring?" posts.
I currently use this for HN: https://hnrss.github.io/
It feels like RSS is a nice solution for releases of articles and such, I wonder why we don't see too much of it anymore nowadays, though someone would probably claim that it has something to do with ads.
I really like the Ars Technica model, where they have short snippets available in the public RSS feeds, but also provide feeds with full articles for paying subscribers. I don't like to leave the comfort of my reader for the discomfort of a web browser (they're all terrible).
Oh, wait; doh! I forgot the twits got rid of it some years ago, impairing the hell out of the usability of that site.
RSS is useful in the following way: the reader is vaguely structured like an e-mail box, where you have folders representing various feeds. You can dismiss items you've already seen, just like marking e-mail read.
New items are similar to new e-mails. You get some status visuals on the folders like new items unread with a count.
Almost a decade ago, also, I used a certain RSS feed program on an android phone to automatically download news podcasts, and listen to them while cycling. How that works is that the RSS items contain URLs of media files like mp3. Those get downloaded and played. Instead of going into some streaming site or app to get the latest news, the thing is fetching them for you, making them available locally in its UI. The items expire automatically. It made sense with poor mobile networks and data plans; it would sync while you're on Wi-Fi and then you have the programs there when you're on the road without using mobile data.
I was a 100%-RSS-news guy a few years ago, but found myself more and more looking at twitter for a certain kind of news content (mostly rants about how bad agile has become). But I kind of hated needing to use two apps to get my daily content fix. Strangely Twitter didnt provide an RSS feed for the raw timeline feed. So I build myself an Twitter2RSS thingie.
Go / Postgress / Heroku For RSS generation: https://github.com/gorilla/feeds (looks stale now)
Was a great learning experience as I was teaching myself Go back then. Dont use it much these days anymore, but its still working, which I find incredible, since I basically havent touched the thing for over 6 years!
* I near-exclusively access my text-based news via RSS, mostly via thunderbird, using many many RSS feeds gathered over years. I got here from an RSS feed.
* I run scripts which monitor a few web pages and give me RSS output, so that I can see new posts on these pages mixed in with my news.
* I use an RSS reader on my phone to notify me of certain events, and I have an RSS monitor widget for those events in my task bar on my desktop.
* For work, I scrape something on the order of ~200K RSS feeds and ~20mil feed items every ~couple of days.
I've tried elfeed but it's a bit slow and sometimes it's DB go TFU so I end up with Miniflux, just because it's simple and quick enough to be setup and on my homeserver I can see the same feeds on desktop, laptop etc.
I do not use them much more than news, for software I only use them for security advisors and new releases of some projects I track, I've planned an RSS-centric personal website but it's a skeleton essentially empty so far and it's veeeery low priority so...
My most important pain point is publishers who not put full posts inside RSS.
https://discovery.thirdplace.no/?q=https://www.youtube.com/w...
It was a personal pain point for me.
I really like the https://feeder.co/ extension, and even pay for the Pro license. There are a few minor bugs, but overall it works extremely well for me. I don't often use the account Reader feature, but the extension has a nice way to organize various feeds into categories/folders. You can stay on the free tier forever if you'd like, but to sync across multiple machines you'll need to pay.
FWIW, the Chrome extension is named 'RSS Feed Reader' if you want to search for it. For Firefox, Edge, and iOS, it's just named 'Feeder'; Android - "Feeder.co"
The only pain point I can think of has more to do with feeds getting completely re-built from scratch. Haven't quite figured out a way to de-duplicate repeat feeds when that happens.
For the client, I've landed on FreshRSS for the web/desktop experience as it allows me sufficient control to group and tag articles for my own needs. Deployment is simple: docker-compose on a small Droplet that hosts a few other apps. It's helpful that there are a few RSS reader apps on iOS that support syncing with self-hosted FreshRSS instances, as well.
I’ve used a number of blog engines over the years and don’t recall ever having a problem with RSS. It seems like a solved problem on the publishing side.
I do find it annoying when the RSS feed publishes only partial content without making it clear that it’s doing so.
As an example one group I am involved in tried sharing a calendar via RSS but every time they created a new appointment they would copy an old one, update the description, update the headline, then change the invite list and meeting time. Each action resulted in a new record being added to the RSS feed.
Cache invalidation is hard, apparently.
I like it because it gives a very clean, simple interface to multiple sources and keeps track of what i've seen.
I can't think of any burning pain points, off hand.
Then there's a button on my dashboard that opens all of those items in new tabs at once (sometimes a dozen or more!), and also moves them to "seen.txt", which is ignored when fetching new items.
It's as simple as it sounds. I created it when Google Reader died, and has served my needs perfectly so far.
I have https://inoreader.com bookmarked on all my devices allowing me to keep abreast of several hundred feeds throughout the day. It's great that it supports the same keyboard shortcuts as Google Reader (RIP).
I always add RSS feeds to websites and most web applications (for non-proprietary and public info only).
I use Miniflutt on Android instead of the PWA mainly for the mark-as-read on scroll functionality (which I contributed because it's a must-have feature for me). About the only feature I find missing is a global regex blocklist which I may put in a PR for as well.
I use RSS (well, Atom) in my blogs.
I use Liferea on non mobile.
1. I get to choose what I see. No one can hide content from me.
2. It doesn't encourage Platforms. I can just as easily watch content from PeerTube or Odysee as YouTube. I can subscribe to any random blog just as easy as Tumblr or Medium. No need to for every author to be on every site that someone might look at.
Other than for news and stuff i have a google apps script that uses my RSS feeds to add new youtube videos to specific playlists I have.
Since youtube's "subscriptions" page became so bad in the past years that script is a must
I use netvibes for reading.