Hacker News new | past | comments | ask | show | jobs | submit login
It’s Time to Get Back to RSS (danielmiessler.com)
798 points by danielrm26 10 months ago | hide | past | favorite | 365 comments



RSS is dying (or dead) because it was incompatible with the dominant business model on the internet -- advertising. This is why Google killed it. This is why lots of professional publishers hated it. With HTTP you'd be able to earn money via embedded ads but you'd earn exactly $0 via RSS since the feed was stripped of ads, just content. This forced publishers to put useless blurbs, redirecting to the HTTP version, which was a bad user experience and just sucked.

I'd like to see new innovation around protocols and client 'browsers' that were made with monetization built-in as a first-tier specification.

1) client sends request for content with some header with payment information attached. 2) server verifies payment transferred. 3) server responds to client with content after payment verification.

If this existed, RSS would be alive and well. Internet publishers would be alive and well. The internet would be a more beautiful place with a viable first-party alternative to ads.

Challenges here would be:

- Sufficiently low transaction costs to make micropayments viable. (Bundle payments?) - Verifying proof of payment extremely fast

Someone(s) should create a new protocol.

FTP was invented in 1971. SMTP was invented in 1982. HTTP was invented in 1989. RSS was invented in 1999. Bitcoin was invented in 2008.

The amount of innovation around protocol has been abysmal relative to the explosion in creativity around applications on top of these protocols. And SMTP/HTTP are the only ones with any real mass adoption today.


Maybe the world needs an unmonetizable space.

I keep hearing how RSS is dead for many years. My favorite blogs seem to be doing fine, its not like my newsblur stream of posts has decreased over the years.

Many people are unhappy about non-monetized readers and authors just being happy off by themselves. The people most able to "fix" that are the happy readers and authors, and the readers are not very motivated to spam and tax themselves, and the authors obviously don't mind not monetizing or they wouldn't be blogging to begin with.

There doesn't seem to be an obvious disruptive force or angle to apply force to "improve" the stable situation of a distributed decentralized happy unmonetized ecosystem.


For the blogs and writers you follow that use RSS, are they really choosing to forgo a meaningful monetization opportunity?

Unless a blog really has ~50,000+ daily uniques, or a lucractive audience niche, I suspect there’s not much opportunity to monetize even if they wanted to.


Their headlines in my aggregator are advertisement.


Well, there are still affiliate links, you can still promote your own products like courses or books and invite readers to subscribe to paid newsletter. Most of my subscribed authors have Patreon now and it generates much more than ads would.


I often ponder about the current push towards federation in tech such as Mastodon, and wonder if the issue isn't more around the issue of finding the sites in the first place, rather than being able to access sites via RSS. RSS is usually a default in many web publishing tools (wordpress, hugo etc).

As someone getting into RSS more recently, I've definitely struggled to find decent sites to follow (though trawling hackernews totally helps:)).


Building a list of RSS feeds takes time. I find that when I stumble across a new site (regardless of how I got there), I poke around to see if the content is interesting. If so, I seek out an RSS feed for it.

Note there are ways to get an RSS feed from some sites without them (RSS-Bridge[0] is what I use), but the experience is much better if the site natively produces one.

In time, you'll find yourself with hundreds of feeds on a variety of topics.

[0] https://github.com/RSS-Bridge/rss-bridge


Last month Inoreader introduced this feature too: https://blog.inoreader.com/2020/04/convert-almost-any-webpag...


You might like Baldur Bjarnason's link blog: https://notes.baldurbjarnason.com/

He always seems to have interesting links to good blog posts.


People say that RSS is dead, but actually I think it's doing pretty much the same, if not better, than before Reader shut down.


Quality content doesn't happen without costs.


A lot of top-quality dev blogs are written by people who have existing well-paying opportunities. A lot of these people aren’t making an income writing blogs, just sharing the lessons they’ve learned from their jobs. Dan Abramov’s blog is a good example.


That is a highly disaggregated long-tail group of content with lots of low-quality content mixed in. Not only that, there's still the opportunity cost for the writer, and secondary non-monetary benefits that "pay" for the effort, be but let's ignore all of that for now.

It isn't practical for consumers of content to try and find the quality content mixed among the mass of dross when content is spread out among individual creators. This gives rise to aggregators. And even aggregators of quality content cost money to run. Maybe you'd use HN as a counter example but HN isn't run out of the benevolence of ycombinator, there are secondary benefits that justify the costs.

So my comment shouldn't be taken as "there's no such thing as quality content without costs" and if I made it too easy to interpret that way instead of more charitably, that's my fault.

So let me amend it: You don't get concentrations of quality content without costs.


idk but that does seem like it comes from costs, just not from viewer funding.


"Costs" does not have to mean "advertising".

Here's some quality content: https://www.raptitude.com/

I've been following that blog for years, via RSS. Every article he writes, I read. He gives it away. You can support him by becoming a patron or signing up to courses, etc. he offers. No advertising, and very high quality.


Right now the top reply's first sentence is this: > Maybe the world needs an unmonetizable space.

I have noticed that people in hackernews generally tend to look things from a, excuse me for saying so, narrow business "make money" perspective. Or maybe it's somewhat US related, I'm not sure.

There are thousands of sites that don't live from advertising. Government, universities, and every public or even private institute has a site to provide information, news, announcements etc. There are also hobbyists' sites who are never going to make money from adverts.

Not everything should be reduced to a Facebook page.

Not everything should be about money.

First and foremost I think there is a need to recalibrate what matters and what actions that requires.


I do a lot of things for free because I have the luxury of a job that pays well. Not everyone has that kind of freedom and if someone wants to put food on the table, then money has to be a factor. Otherwise you'd lock a lot of things to only people who have extra time to take risks on activities that won't get them anything in return. Saying not everything should be about money is a pretty privileged position.


Not every activity needs to be a viable way to make money. If you're struggling to put food on the table, maybe you should explore other avenues for making money besides blogging. If you want to write, there are lots of ways to get paid for writing.

Everyone has activities they enjoy that don't produce any income. Just because people enjoy an activity doesn't mean that there has to be a way to make a living from it. People enjoy playing video games, riding horses, sailing, painting. For most people they will not be able to support themselves with these things and we generally don't think that we have to go out of our way to make it possible for them to, so why is blogging different?

The general order of business goes "Here is something people are willing to pay money for -> You can do this and make money" not "Here is something I want people to pay me money for -> Pay me money for this"


Not only this, but blogging can also be a marketing tool, where you (or a company you're writing for) publish articles not as an end in themselves, but as a way to show you're an expert in a given domain. For SEO reasons, but not only. These blogs don't need advertisement to make money, they are the advertisement. I'm pretty sure nowadays, this is where money is in blogging. If you want to make money as a blogger, I think it's a way safer road than the advertisement-centered approach.


This threshold has been measured. Everything stops being about money when household income hits about $70k-90k. https://www.marketwatch.com/story/this-is-exactly-how-much-m...


I agree, I think good thing to point out is that incentivizing and monetizing is not the same thing. If you would excuse a tangential argument, I notice that in Eastern culture 'social cache' is much stronger contender to monetary benefit than west; and in Western incentivization = monetization is more stronger. (my theory is that with more individualistic society with less strict societal norm money is one thing most can agreed upon)


The blog itself seemed like an example of content marketing, As is evident from "I challenge you to signup for Feedly" at the end of the article.


Yeah, I can't decide if that was a "this is convenient" or "I'm getting paid to hock this service." Either way, there are other ways to get RSS feeds; I run my own FreshRSS server, but there are apps for your phone/tablet that can use instead.


This website was started by a venture capitalist firm and its original name was "Startup News". The user base certainly reflects that.


> This forced publishers to put useless blurbs, redirecting to the HTTP version, which was a bad user experience and just sucked.

Is this that bad? I personally don't mind this at all, I subscribe to RSS feeds so I can easily tell at a glance who has an update in one place. If they then want to redirect me for the full content, then so be it, especially if the alternative is me loading every single one of those sites every day to check anyway.


I do mind such RSS feeds that just try to force people to load the webpage. With the vast majority of my RSS feeds, I never have to leave Emacs’ Elfeed RSS reader to consume daily news. If a feed won't let me see the full content in RSS, then I am more likely to delete the feed than become a pair of eyeballs for the website’s advertisers.


I work on a project to convert partial feeds into full-text versions. It pulls in the article content from each feed item and then creates a new full-text feed. Feel free to try it out here: http://ftr.fivefilters.org/


If somebody chooses to make their (complete) writing available only on their site, I can choose not to read it (and often do!) But I don't think it's right to pull it from their site and repackage it.


Absolutely agree. I just can't get why in every discussion about RSS somebody mentions that y it's dead because publishers had to give away content for free. Feed could contain full article, but don't have to.


Agreed. The RSS can easily be equivalent to a Tweet with a link attached. I like Thunderbird on desktops including RasPi, have yet to explore cross platform + floss mobile or web readers, but have built my own little youtube-like stack with RSS that is similar to YT’s feed format and can aggregate them, in the iSpooge Live project in Clojure + ClojureScript.


Thing is, RSS didn’t “die”. It only ever filled a couple of niches, and arguably it is more popular than it ever was because not only are people still continuing to use RSS readers, podcasts are arguably more popular than they’ve ever been (this period of time with the coronavirus possibly excepted).

I get that when people think of “RSS”, they’re thinking of tech that downloads text and maybe some images into a reader, but it’s just another means of syndicating media. The same places that disseminate the full text continue to do so because they’re not invested in making money directly off the writing, and the same places that have an investment in you going directly to the site continue to operate their feeds in that manner. If you have a browser extension that can discover it, or are willing to do a bit of guesswork, it’s amazing how many new sites launched in the last year that might not expose it on the website (because the webmaster didn’t think it was important to expose even when the CMS has it) actually have a feed you can pass off to a reader.

I don’t think RSS is ever going be competitive with a service like Instagram or Reddit for what they do, but the full value of these services doesn’t lie entirely within a feed and wouldn’t be neatly exposed within the constraints of the RSS or Atom specs to begin with. It’s different, and therefore have other, albeit less popular than social media, uses.

If someone prefers a stream of photographs to a stream of blogs and news sites, and that’s a lot of people, then RSS isn’t going to be competitive for their attention. If somebody prefers the discussions on Reddit and the semi-randomness of sources, RSS isn’t going to be competitive for their attention either. If you don’t care about an image stream (with comments, reactions, stories and whatever else Instagram has), and you can do without the Reddit commentary, well, you may very well have a use for RSS. Or maybe not. If it hasn’t gone away yet (and it really hasn’t), I have a hard time believing it is going anywhere in the near future.


> podcasts are arguably more popular than they’ve ever been (this period of time with the coronavirus possibly excepted).

There's been a worrying trend of some podcasts shifting to a "get our app" mode of delivery. I've been boycotting each one, and been encouraging the podcasts I donate to to explicitly keep the RSS delivery.


This is news to me, and I’ll keep it in mind. Personally I’m not so sure I would boycott for the sake of effecting a policy change as I simply wouldn’t continue listening because the only thing that makes podcasts tolerable is that I’m filtering them through Overcast’s smart speed. Can’t stand talk shows at the pace of a normal conversation.


> This is news to me, and I’ll keep it in mind.

Granted, I only have two immediate examples:

* Season 2 of The Onion's hilarious true crime parody (https://en.wikipedia.org/wiki/A_Very_Fatal_Murder) is exclusively on a platform called Luminary. I can't even figure out exactly what Luminary is, because they serve me with a special can of "Luminary is not yet available in your location" crap. Not that I care anyway. The first season is on regular old RSS and is fantastic (if you like the true crime genre and The Onion-style silliness).

* Wind of Change (https://crooked.com/podcast-series/wind-of-change/) investigates the (serious!) conspiracy theory that the CIA created the hit "Wind of Change" by The Scorpions. Now, this is borderline, but the podcast is available over RSS – however they're clearly trying to push people onto Spotify by releasing all episodes there at once and only dripping out an episode a week over RSS. I'll allow it.


I woke up today, checked Hacker News and saw Joe Rogan setup an exclusive deal with Spotify or more specifically, Spotify setup an exclusive deal with Joe Rogan. I was thinking about this thread all day. I've never listened to any of his podcasts because there's so many podcasts out there that I'm more interested in I just haven't gotten around to trying the Joe Rogan Experience. I like the idea of someone that just gets someone into a room and let's them speak from the heart or shoot from the hip about whatever their pet issue, policy, thing they are known for is, but I also have plenty of other discussion format podcasts I'm subscribed to and working through the backlog of.

Luminary by the way is one of those "we're going to do podcasting but it's not really podcasting because we're going to lock things up behind our paywall." kinds of "podcasting" apps. Think of it as Spotify but without the music or (I believe) without the free tier. Mind I've never signed up, but I remember there were a big story last year, or maybe it was the year before? Either way, they aspire to be the Spotify of Podcasts and Spotify is also trying to be the Spotify of Podcasts. I think Spotify is probably winning that exchange.

The strange thing is, a lot of these podcasts could probably get away with charging more directly if they wanted to. Joe Rogan has literally millions of fans, and charging for episodes isn't exactly incompatible with RSS. It just means you need to authenticate for access to the feed which is fine. I do that with TechMeme Ride Home subscribing to their ad-free feed. I think I'm paying $5 a month for that at the moment. I also pay for podcasts that don't have a premium feed per se and can be found in every podcast directory, but are products of news and political commentary organizations to which I pay a monthly or annual sum to support, primarily because I enjoy their podcasts (and newsletters) so much. I get access to their websites and dead tree print editions, but really it's the podcasts I find I like the most out of the deal and am happy to pay for just those.

And Joe Rogan is bigger than almost all of those, so selling out to Spotify to try to be the next Howard Stern feels a bit like he's selling himself short. Welp, we live in a free society with free markets and he's free to do that if he wants to. I get that he's a controversial figure for some, but so are Howard Stern and PewDiePie and last I checked, they're doing fine.


I used to speed them up as well but I stopped doing it for now as I feel like the time is not only there to hear the words and the surface meaning (I can do that at 2x), but to digest it and take it further inspire thought in myself. I sometimes even pause and think.

Depends on the type of show of course. But if they are talking about some interesting deep topic, I find it useful to use that time to think more. It's kind of the difference between skimming and flipping through a book vs reading a paragraph and looking up, pondering a bit, then reading along, after a chapter standing up and walking back and forth a bit in the room, thinking.


While RSS doesn't replace a site like Reddit, it's great that Reddit has RSS support built-in.

Much of my browsing of the site at least starts off in a feed reader, to make it quick to find what I'm interested in. From there, I'll open links to the site proper, with all the comments and such that make the experience what it is.


Actually forgot about this! That said I do remember trying it some years ago; Reddit RSS just ended up providing too little signal for too much noise. I came to the same conclusion with Twitter RSS and certain types of Tumblogs.


Twitter RSS works fine if you want to follow certain users and/or searches, which is what I use it for.


No, I understand that basic premise, it's what I did too, it's just, well this is an old outstanding complaint I've had about Twitter and not so much Twitter RSS, and really the main reason I don't consistently use it anymore. Every now and then I re-install Tweetbot on my phone, create a new account, carefully curate a list of Accounts to follow based off whatever I'm thinking Twitter might be useful for this time.

I personally think it's really cool that I can follow people I respect and see a little bit of this, and a little bit of that, but inevitably some stupid controversy erupts and all of them have to get their opinion in on the subject.

The other problem I have is that generally, people tweet too much. It is too easy to compose a tweet and fire it off into the void and and while they're at it, about five or six retweets. RSS didn't improve this for me, I look for ways to make Twitter work for me, and it's just unworkable. The one time the list of accounts I came up with was political (that is to say rather than being constantly exposed to the abhorrent politics of people I otherwise respect for other reasons, I picked people who I respected in the field of politics), it was still an irredeemable crapshoot.

If Twitter works for you, I have no doubt Twitter RSS can also work for you, but Twitter doesn't work for me. Too much noise, not enough signal. Same with any kind of firehose feed, I think I even followed an HN firehose feed at one point six or seven years ago which just overwhelmed my reader within a week. I mean if you tried following Tumblogs as I did at one point in my youth, I tried to treat them like blogs, but what I didn't realize was that nobody was really interested in using Tumblr like a blogging platform. Your typical user that posted at all, was generally going through their feed and reblogging whatever they liked, and they'd do this 30 or 40 or 50 reblogs at a time. It was asinine, I gave it the old college try because a few of my friends really seemed to like it and I wanted to see what they saw, and it was just asinine.

Anyway it sounds like you're still using Twitter RSS, and, I mean this sincerely, kudos to Twitter that they're apparently still maintaining RSS feeds. I really thought that would have been killed off by now. More choices are generally better, so I'm glad that's still a thing.


Twitter got rid of RSS years ago, I use a service for that. I did the Tumblr thing for a while as well, and the hole reblogging thing was extremely annoying. It's a blogging platform, which is exactly how I used it.


I use userscripts to try and find those hidden feeds, but sometimes they still can't be found. Sites like that are the bane of my existence!!!!!!


I mean if browsers hadn't gone out of their way to remove RSS discovery, it wouldn't be necessary. I don't remember if I mentioned this in my above post and I don't want to re-read every word I wrote, but I've got the NetNewsWire Safari extension which finds feeds for me. Pretty sure I also have some extension in my current Firefox install, but I haven't opened that in a while. Sometimes the feed really doesn't exist anymore, which is unfortunate.


I use a service to create feeds from pages if they really don't exist. It's a paid service, but worth it!


Micropayments have been tried. They all failed. The fair market price for content is $0. What is happening is that the only publishers that will exist shall be only those who create it for free. There is simply a glut of content out there because the barrier to creating content is completely gone.


>The fair market price for content is $0.

That's a gross oversimplification. Lots of content isn't worth anything but some is. If you're a stock trader, for example, certain timely information is well-worth paying for. And there's a market for that served by Bloomberg, Thomson-Reuters, and so on.

The term "content" obscures the differences by suggesting its all interchangeable. That's certainly in the interest of the Googles and Facebooks of the world: When all content is equal, no content creator has any negotiating power.

But there's a distinction between a multi-month investigative report and hastily paraphrased rewrite at some fly-by-night website intent on capturing algorithmic ad dollars.

Pricing news content is hard because it has a different value to different people and often the value is only apparent after it has been consumed.

It's worth asking how much we will pay for reports on political corruption, civic injustice, and so on. If the price for content is $0, the signal to noise ratio will disappear.


I think journalism is going to have to clean up its act a lot before I'm willing to trust it enough to pay for it.

I'm seeing more credible journalism done by bloggers than the mainstream media at the moment.


I meant "content" as in cultural content--articles, Q&A submissions, blog posts, takes, memes, recipes, how-tos, tutorials. Content that makes up >80% of G search results. At one time, there was a price for such content because you could only get it from a print publication or from broadcast television. Now that there are no barriers to posting, we see how many people actually want to publish content even if they have to pay out of pocket to host it (i.e., making the price of their content negative since the consumer gets to freeload bandwidth without ads and consume content).

> Pricing news content is hard because it has a different value to different people and often the value is only apparent after it has been consumed.

You're suggesting that journalism is a public good because the value is captured by society as a whole as opposed to by the individual newspaper subscriber. I wouldn't argue against investigative journalism being important for democracy etc., but why does it need to cost anything (even by the public)? What about the whole Harvey Weinstein/MeToo saga which started after some individuals made some tweets without being paid by a journalism company to do so. There is no shortage of outraged citizen activists who will do "investigative journalism". I don't know why their free work should automatically be considered inferior just because they don't have a journalism degree and aren't being paid by a journalism company.


How do the patreons of the world fit in there? There is such content creators that I enjoy enough to give them a few bucks every month to make sure they can keep doing it


Stuff like Bloomberg, Thomson-Reuters, etc is valuable to a specific type of professional and will always have a niche audience. That isn't really what the discussion is about.

Reporting on corruption and injustice is critical to a society, but doesn't have any monetary value. The way I see it, we already have a way we choose to fund that kind of thing: non-profits/NGOs. Some media outlets (Vox, The Guardian) are already turning to just asking for money in this way, and over time journalism might become more formalised as something funded by the public for civic ends, not as a commercial product. To that end it should be unbundled from stuff like sports coverage and celebrity gossip - which might really end up disappearing as a viable career and be done exclusively by amateurs.


ProPublica seems to break a couple of huge stories every year operating as a nonprofit. The fact that they aren't ad funded means they don't have to churn out lots of Content to keep eyeballs coming, so I know every story I hear from them is worth my attention.


Yeah, it's a good example.


>The fair market price for content is $0.

I would have agree with this two years ago but things have changed. I'm just worn down with the huge volume of free dross that doesn't tell you anything. This year I've started subscribing to a bunch of resources and stopped with google news/fb noise.


> Micropayments have been tried. They all failed. The fair market price for content is $0.

To be fair, most were aweful and never reached critical mass. And many were to early and badly placed. Today situation is different. Patreon, twitch, youtube, netflix, spotify and all the other paid services proof that people are willing to pay something for content, to their conditions.

I think a well implemented micro-payment could work out today well enough to be viable. Something build into the browsers and aggregator-sites (reddit, hackernews, google news, facebook...) first. Most users don't wanna waste their time with micromanaging their bills, so make it optional, and automate it for the rest of the time.


Free, loss leader, or honeypot.

Most places use content to get you to look at ads, but some places write content as a sort of ad itself - we said something thoughtful to build brand recognition, consumer confidence.


Ja, I've seen enough of these sites, the content is awful but filled the first pages in my Google search.

I used to call them malicious SEO sites, but your point makes better sense than my hatred.


What about games? People pay tons of money for them, and they are content too. When Cyberpunk 2077 comes out later this year, try telling people it's worth $0. I think the difference is social norms. The internet having no real built-in payment mechanism has obliterated the norm of paying for most content. Games have held the line, but things like Google Stadia and Apple Arcade are working to change that.


A game is intellectual property like a movie. If I copy and paste 1/5th of your game, you can sue me and win. If I copy and paste 1/5th of your blog post, as long as I put quote marks around it, I'm safe.

Games won't become free until it's easy for commoners to make them. Even with Unity and all the easy plug and chug game dev resources, it's still hard and requires multiple people with different talents.


It's social norms and perceived value (which are interrelated). From sampling music to quoting literature - there is case precedent and fair use.

Software is no different. When Compaq reverse-engineered the IBM PC BIOS, the standards were incredibly high. If I violate a GPL license or steal some copywritten Unity code for my game - I'm probably pretty safe. Unless, of course, a bunch of people agree that my software is valuable. I could be in big trouble if that's the case, even if I give it away for free (as in beer) and never make a cent off it.


The legions of small publishers making a living on substack beg to differ.


1) Subscription is not the same thing as micropayments (as they are commonly understood)

2) It's pretty unclear to me that Substack actually lets "legions" of writers make a living. https://substack.com/discover lists the top (income making?) writers, and it drops off very quickly. By number 25, it is "hundreds of subscribers" paying $10/month. That's a nice hobby income but isn't really "a living".


> Micropayments have been tried. They all failed

Would be interesting to read about that — you happen to have any links?

I websearched for "failed micropayments" and found: "One type of micropayment that does work — one you might not even think of as a micropayment — is in-app-purchases (IAP). IAP are a huge source of revenue [...]"

And: "One reason users don’t like micropayments for content [to read] is it requires a decision ... waste the users’ mental effort ... costs so little that its implied worth is almost nothing" (here: https://blog.applovin.com/why-micropayments-fail-and-one-not... )


Sure, just look around, do you see any successful micropayments processing companies? Do you use any of them? If any of them were successful, you wouldn't be asking this question.

There was a wave of micropayments startups about 6 years ago. The only one I remember is Flattr. The crypto bubble came and gave new hope because maybe the reason micropayments failed before was the credit card processing fees. Nope, still didn't work. I don't use Brave Rewards and have never heard of anyone making more than pennies with it.

Xanadu was one of the original visions of the internet from the 1960s. It sounded like a lot of the idealistic plans people come up with for micropayments where content creators can collect micropayments from viewers. Of course it didn't work.

Micropayments is an absolute cemetery of a business model.


Thanks for the reply, didn't know about Xanadu. Didn't come to think about Flattr.

> If any of them were successful, you wouldn't be asking this question

If you change from to: "successful" to "hugely successful in all English speaking countries" then I agree ... almost,

But there're also different continents and countries and different languages

... Which prompted me to now search for "Chinese micropayments" — and:

> "[the company's] growing profit margins are riding the wave of China's exploding media micropayment economy"

https://www.musicbusinessworldwide.com/tencent-music-uses-ti...

Reading a bit more on different sites: Apparently this started 5 years ago and has been working well, in China. Micropayments. A little bit I got the impression it's currently being used primarily for tipping.


> This is why Google killed it.

Is this true? It certainly doesn't feel like it. Reader was a first party branded reader. Google could put ads on Reader relevant to your interests (which they'd have been more keenly aware of than if you weren't using Reader), and they didn't have to pay out the publishers.

Google killed Reader, I think, for the second reason you mentioned: publishers were neutering their feeds. Reader had gone from a tool for consuming content to a tool for being notified about content.


Google says they killed Reader because of declining usage[1], and there was little reason to doubt that at the time. It's very true that people were switching to mobile consumption and no longer reading lots of news all at once.

I was a heavy user of Reader (and did huge amounts of work on RSS/Atom/etc), and I agreed with this assessment at the time.

More and more publishers were just putting headlines in RSS, so you had to click through anyway, and commercial blog sites were publishing a much larger volume of content, most of which I ignored. But I didn't want to unsubscribe, because occasionally there was something good.

Given that, I was finding much better content through aggregators like HN or via Twitter/FB.

Maybe they could have iterated and built something different, but what they had really wasn't useful.

[1] https://www.wired.com/2013/06/why-google-reader-got-the-ax/


"In December 2007, Reader linked up with Google Talk (the chat feature in Gmail) to display shared streams from friends. Within the context of feed reading, it fomented something of a Neolithic Revolution. Foragers, hitherto gathering headlines on a crude and solitary basis, became farmers, cultivating streams of information for their neighbors. Sharing increased 25% overnight." [1]

[1]: https://www.buzzfeed.com/robf4/googles-lost-social-network


What they didn't point out there was that feature was a disaster for many users.

There was no separation of the chat friends and the people that you followed.

I've still got random people in my address books from that debacle.


Google would have been rightfully sued to hell and back if they started putting ads in the reader client with no compensation back to the publishers.


Why? Many RSS readers have ads, and the publisher is making their feed available to readers.


Google never made an public attempt to show ads in their Reader, unlike GMail. Publisher feeds were also still prospering and widely used when Google killed the reader. It all started to die fast in years after.

Also, IIRC they specifically killed it because of Google Plus, which according to their plans should replace Google Reader and the demand for RSS.


>1) client sends request for content with some header with payment information attached. 2) server verifies payment transferred

What you describe looks like authentication and authorization rather than payment processing. Which may be a good thing. Adding support for authentication to the RSS protocol (OAuth for RSS anyone?) and RSS clients can potentially make it more interesting for publishers and solve the problem with monetization.


RSS is very much alive and aggregators like Feedly are way more powerful than Google Reader. All media website still use them, and even some major ones post full-text, The Verge does that, for example.

Monetization is rather simple, people visit a website if they want to read more about the news. In exchange you get guaranteed access to their attention not intermediated by Facebook, Google or Twitter.


Someone explain to me why the following combination of steps wouldn't solve all problems - for both content publishers, and RSS app creators.

Content publishers: Update the RSS feed generator to include the ads associated with the post. Write tools for dynamic ad insertion that actually updates the syndicated feed.

RSS App creators: Build ad networks that work with both ad buyers and content creators to match content-relevant ads, and either replace the dynamic in-post ads (and share revenue), or add additional intra-post ads.

Someone explain to me why this couldn't work with the existing technology stack - no new protocol, or client-side changes required.


It wouldn't work because it doesn't support the intrusive tracking (including "telemetry" and "analytics"), giant slide-over windows that obscure the content, repeated pop-ups demanding your email address, announcements that "you have 1 free article remaining unless you pay for a subscription," social media integration buttons, solicitations to "download our mobile app," CAPTCHA loops, "DDoS protection" delays and redirects, ad-blocker-blockers and autoplaying videos that are integral to web sites in 2020.

Also it is too lightweight in terms of cpu, memory, network and power usage, so it wouldn't push users to upgrade to the latest and greatest hardware (and data plans) every year just to maintain the same performance they had last year.


But the thing is, it’s not dead nor dying. I don’t think very many major websites have stopped their rss feeds. Maybe it’s just you that isn’t using it. There is rarely a webpage that I find that can’t be followed via rss, either natively which is most always the case, but there are also services that an rss-ize content.


Agree here. I don't think many sites have either. You have to dig a little, open the code-behind, but they're there.


How is it any harder for a publisher to embed a display ad than do what John Gruber does and have a weekly sponsor in his RSS feed? It's easy to insert ads into any RSS feed. The publisher/advertising argument for the death of RSS doesn't hold water. Folks prefer Twitter and Facebook to get a feed of news. The majority of people don't want to do the work of manually curating their feed. RSS just didn't win on the merits as a feature. I still love RSS and use it every day, but it wasn't "killed by publishers."


The protocols are designs as small layers and on top of one-another. I wouldn't expect a new protocol to be invented every year or even decades.

I think what you mentioned could be accomplished using existing protocols. There isn't a need for a new protocol for this specific purpose.

For this specific purpose, what if we just had personalized/signed RSS feeds per payer?


> 1) client sends request for content with some header with payment information attached. 2) server verifies payment transferred. 3) server responds to client with content after payment verification.

This is exactly the recently-proposed LSAT protocol[0]. It uses the HTTP402 response code to prompt for payment over the Lightning Network in exchange for a cryptographic bearer credential that may be used in future requests to the server.

[0] https://lsat.tech/


I am building my business fully based on RSS. It is a news API. I got all my data solely form RSS. And I can confirm that the vast majority of news outlets still have it.

https://newscatcherapi.com/

I probably should write a post with exact numbers: * how frequently it is updated * how many websites still support it * etc


I think you gotta connect the dots with UBI here. You wouldn't have to monetize if you didn't have to monetize.


While I agree with this take on UBI. There are middle grounds to explore.

The streaming and game services seems to point at monthly subscription fee from an aggregator as a working model. Not sure how they distribute royalties though, I suspect there are many models I won’t agree with based on what they incentivize.

Then there are the Kickstarter/patreon. Where money is given more in support than for a specific product. This is more like the UBI approach.

I think there is room for a hybrid approach here, merging the two.


I'm keeping my eye on a few people, like Abbadon who makes "Kill Six Billion Demons" and Andrew Kelley the creator of Zig.

But really I think we need a major reconfiguration of our global economic structure.


> RSS is dying (or dead) because it was incompatible with the dominant business model on the internet -- advertising. This is why Google killed it. This is why lots of professional publishers hated it.

> I'd like to see new innovation around protocols and client 'browsers' that were made with monetization built-in as a first-tier specification.

I don't think the solution should be to cater to what big platforms want.

If we did that, then the logical conclusion of that is that we would have to download some bloated locked down app for each platform we wanted to visit, and we would have a plethora of walled gardens. Copying and pasting text would be very limited, and we certainly could not "view source". The best we could do to save content was take screen shots, we certainly could not click on an image and save it. Basically what they are doing now.

Platforms would like nothing better than to completely deprecate access by web browsers all together. (Didn't Instagram do that recently?) Perhaps in the near future, websites will require that you use an "Apple approved" web browser, if they let you access them outside of the app at all.

But, they fact that they present any of their content at all is because of the ubiquity of web browsers. They could probably make more money if they had complete control of their platform, and could do things like prevent ad-blocking.

So, we should have pushed more for RSS to be de-facto requirement of serving content. Firefox, and other browsers, should have advertised when RSS was available, and make it highly discoverable for users.

> The amount of innovation around protocol has been abysmal relative to the explosion in creativity around applications on top of these protocols. And SMTP/HTTP are the only ones with any real mass adoption today.

I don't personally care that much about the protocol itself, I care about the content that the protocol makes available. If reading an RSS story required unpacking a bloated js runtime and fetching even more content, then why not just use a browser?

Publishers also hate SMTP and IMAP, and would love to force users to log into their platform and view ads, just to send an email to someone. And Google is certainly doing their part to eventually kill off these protocols.

AOL and many 90s ISPs did not support these protocols either (even though they used them internally) because they wanted to make users log into their platform instead of using their own mail client.

But the reason that SMTP still exists today is because of its ubiquity. The more RSS is adopted, the more popular it becomes, and the more platforms had to support it, even if they did not want to.

Being able to programmatically send emails is incredibly useful and helpful. I'm sure that when the last SMTP server shuts down, they will tell you that it is ok, because you can still use the mutually incompatible GMail or Outlook.com APIs. Pending approval.


> Firefox, and other browsers, should have advertised when RSS was available, and make it highly discoverable for users.

Firefox did, with a subscribe button in the address bar. https://mariolurig.com/wp-content/uploads/2013/01/firefox-rs...


> Firefox did, with a subscribe button in the address bar. https://mariolurig.com/wp-content/uploads/2013/01/firefox-rs...

Is that from Firefox 3.6? I would have to confirm this, but I'm pretty sure they started de-emphasizing RSS around 2011 with Firefox 4.0.

It was many years ago, I was looking at some discussion about it here: https://decafbad.com/blog/2011/01/15/how-to-use-feed-auto-di...

I remember being rustled about it back then, also.


Re: rss being dead, many US govt sites still have RSS feeds for data updates and even your saved searches:

- PubMed

- Dailymed

- clinicaltrials.gov

Slack has a slash command for ‘/feed subscribe https://foo.bar/baz.rss’ so we use that plus the above sites RSS feed for real time notices on when new data is added.


+1. This is one of the areas where the profit motive leaders to a worse user experience in a competitive environment.


So maybe a combination of RSS and Brave?


I've seen a number of news RSS feeds that only give you the first paragraph or a summary in RSS, and to get the rest, you need to log into the site (paywall). I think this is quite an OK RSS business model - it's like a preview, and you pay to get the rest.


The single worst offense against the usability of Atom/RSS right now comes from Apple and iOS.

If you click a link to an Atom feed in Mobile Safari, iOS will launch the Apple News app. Which will then show you an error message saying the content is unavailable.

As far as I can tell there is no way for an installed reader app to take over handling of feed URLs. It just makes the entire feed ecosystem look broken for anyone using an iPhone or iPad.


I wonder if the EU could, in theory, take action about this behavior.

Apple is basically monopolizing the whole market for rss readers.


How many EU countries are Apple News available in?


After UK exits, none.

Apple News is available in a whopping 4 countries, 4 years after its introduction.


Yeah so the EU can't do much about it.


Only the UK.


The UK is not part of the EU anymore.


Thats neither factually nor theoretically correct, as all the EU laws apply until the transition period has expired, which I wouldn't be surprised will be extended given corona has taken so much focus and time away from it.


Officially they are no longer part of the EU. They have agreed to abide by the rules of the EU for now, and the EU has agreed to allow them to trade as though they were part of the EU, but they officially left on Jan 31.


Duck typing says they're currently a member.


It's both factually and theoretically correct - the UK has left the EU, but laws are extended until the transition period is over.

Note the absence of the UK in the below sources:

https://www.gov.uk/eu-eea https://en.wikipedia.org/wiki/Member_state_of_the_European_U...


It is theoretically correct. The factuality is the point under dispute...


What's your definition of 'in the EU' and 'fact'?

If a nation state doesn't recognise itself being in the EU, and the EU doesn't recognise the nation state being in the EU, It's not in the EU.

https://europa.eu/european-union/about-eu/countries_en#28mem...


Well maybe there’s one way to resolve this dispute?

Did the UK MEPs serve their last day already?


All of them, I guess?


Apple News is not available in Germany. The only thing we get is the non-customisable News-Widget.


Well it might not be available as a service but apple might be handling the RSS URLs anyway.


Which is really awful. Always showing tabloid nonsense that you can't get rid of.


>Apple is basically monopolizing the whole market for rss readers.

You can't be a monopoly with such a small userbase.


Of course, you can. The company is the monopoly not the userbase.


No you literally can't..one of the legal tests is market share.


Oh so, you meant Apple's userbase not the entirety of RSS users?


i don't know about you how you use your iOS device if you do but after installing an RSS reader (NetNewsWire is an excellent open source app) you can share the webpage/website/blog using the share menu and it will be added to your RSS reader


I dislike a lot Feedly, almost like Inoreader and kinda like bazqux.

Not sure why though.


I used bazqux for a while and paid for the annual subscription. It just ended and I was looking around for free alternatives, and found CommaFeed, which reminds me a lot of Google Reader.


I really do like CommaFeed, but they seem to limit the amount of feed items.


CommaFeed is great ! thanks for sharing this


Inoreader (at least their paid version) supports private rss feeds--that is feeds with HTTP Basic Auth headers for non-public content. Private blogs and private RSS are a space where monetizing your content just isn't important. This is where people can write personal, vulnerable things to share with friends and family without creating a burden of "now I need to visit 50 different friends' websites"


Stop using Apple products. Stop developing for Apple products. Stop treating Apple like they matter.


Your other choice without lots of friction is Google, an advertising company. I hate that this is the world we live in.


At least with Android (not Google) I have a choice on what software I install.

Besides that you can still use a desktop PC. Most people do anyway.


> At least with Android (not Google) I have a choice on what software I install.

I'm have very negative outlook on Apple with their vendor lock-in and walled garden, but how long do you think Google will let you make this choice if they gonna have no competition at all?

Google was slowly making Android less and less open with every release. Now "security" features like SafetyNet decide whatever you're allowed to use software or not.

Think on it! Just look at what Google doing with web because of their monopoly on search and browser markets.


F-Droid exists as an alternative app store or you can distribute APKs. Also running Android from a vendor is a horrible experience. I recommend installing LineageOS or any other alternative ROM.

Google might have a monopoly on the web, but at least they have given us tools to work around it, unlike Apple.


> I have a choice on what software I install

No you don't. Much of the software preinstalled on an Android device is permanent. And you can try stripping out Google Play Services, but good luck having a "smart" phone.


Those services are not whats making a smartphone "smart".


> Stop using Apple products.

Sure.

> Stop developing for Apple products. Stop treating Apple like they matter.

The market, as it actually exists, disagrees (especially in terms of profit, as opposed to number of devices). I understand hating Apple, but if I write software for profit, I care about where the customers/money are and Apple absolutely qualifies. Should you also support other platforms? Absolutely. But removing support for a platform like you're suggesting is at best prioritizing ideology over profit, and even the ideology argument is... not one-sided.


This is why I was so heartbroken when Mozilla removed first party RSS support from Firefox, for what seemed like an extremely flimsy justification.[1]

RSS should be ubiquitous, and seen as an essential part of any service that serves structured incremental content. People should be emailing webmasters asking why there is no little orange icon.

It also serves as a back door form of accessibility. But I strongly suspect that RSS goes against the interests of big tech who don't like RSS, because companies like Facebook go through so much trouble to make it difficult to scrape or modify their content.

I just wish that Mozilla would stand up more to their corporate underwriters. Now RSS is relegated to add-ons, and is on the same tier gopher (no offense to gopher).

[1]: https://news.ycombinator.com/item?id=17613051


Smart Bookmarks were fantastic. Add your favorite sites' RSS feeds to your bookmark toolbar and you'd have all the recent headlines from all your favorite sites at one click. Fortunately I wasn't the only one that appreciated this long neglected feature so someone created Livemarks (https://github.com/nt1m/livemarks/) that mostly replicated its functionality. I highly recommend it as I've been on the web a long time and have yet to come across a faster way to check all my favorite sites at once.


Thank you so much for that, I am/was a HUGE fan of that feature in Firefox, and it was actually one of the main features that got me to switch to Firefox in the first place. I was gutted when they removed it, and while I still swear by Firefox over Chrome, I'm finding they're making more and more questionable decisions lately when it comes to their supposed love for a free/open Internet.


I once wrote a little python script that did something like this in the systray.


Innoreader can make a virtual feed from changes that appear in any site -- a new feature I haven't tried yet. Used to use Google Reader, but now I pay for Inno, which I'm happy to do.


That is pretty cool.

But unfortunately that is not really a solution. It is like saying that it is ok that a website removed screen-reader support, because you have a screen reader that can still parse the website anyway. The problem is RSS not being made available at all.

RSS being made available less and less, and they have less of an incentive to do so. And I am saddened that a lot of the good work Mozilla did was abandoned by them and that the web is regressing.

Additionally, having to make your own scraper is really not a solution to RSS not being available. Scrapers are very high maintenance, and can easily break with updates.


Sure -- not saying it's as good as presenting make-for-rss posts -- may be helpful in some cases though. At least I'll know that content changes, and if I don't like the virtual feed, I can just link out to the source and view it directly.


I appreciate you sharing. I didn't mean to shoot you down or anything.


No problem -- I didn't take it that way at all!


as far as i can see, RSSHub does exactly that, and supports as many as 536 scraped sources (of varying caliber) at https://github.com/DIYgod/RSSHub/tree/master/lib/routes.

It's not an outlandish amount of work, if lots of people chip in with their favorite source.


I see what you are saying, but I still think it is a far cry from having content providers simply providing the feeds themselves.

In the same way that I don't think that YouTube allowing users to submit closed caption transcripts, or machine generating them, any substitute for the content creator providing them in the first place. I'm sure in the near future, smart TVs will be able to machine generate closed captions from the audio, but I still don't think we should let television producers off the hook for providing captions.

RSS should be the default. And it is not hard to generate RSS.

I happen to think that big platforms only reluctantly adopted RSS over a decade ago because it was a "standard", and because they felt that it was popular enough to justify the traffic from it. But they do not like RSS. It works against their analytics, their ads, and their control of the presentation.

And while it is cool that people are crowd sourcing scrapers, I think the real solution is to promote RSS itself and encourage more platforms to simply provide it. And organizations like Mozilla taking Facebook's position that RSS is obsolete has been profoundly unhelpful to the web.


Maybe what we need is an accessibility equivalent to the SSL Server Test. Input your domain and it gives you a letter grade on how accessible your site is. RSS access should be heavily weighted, of course.


When I look for an RSS feed, it's usually to see if it's updating, not because of any accessibility concerns. A lot of times, I'll find broken feeds. However, if a site is really giving me trouble with navigation, I'll rely on the RSS feed for reading purposes.


I used to be a big fan of RSS feeds but with their demise I started using Feedly.

It let's me curate sources into different customizable feeds like news or science. I pay for the pro happily since they let me add specific twitter accounts too. Really saves me time!


Feedly is an RSS reader. It’s not one or the other.


I should've been clearer above, I meant the google RSS reader [feeds]. Either way, your point stands to anyone wondering about the identity of feedly :)


Mozilla _critically_ lost its way. No RSS, no NNTP, no EMail; but a paid news service, opt-out services and so on.

The data services which provided content-without-markup were abandoned, and the fluff and garbage were embraced.

It was supposed to be that Mozilla would be a beacon of light in a murky web, but it has lost itself in the void.


Mozilla have been very clear that they don’t have the resources to win every battle. They’re slipping in the browser wars. I too would love them to fight every battle, but I understand the importance of them managing their limited resources and fighting those battles that can help them towards ‘winning the bigger war’ and staying alive.


RSS and bookmarks bypass Google, meaning less money for the browser. Yes Firefox is funded by ads (Google ads).

It seems to me that Mozilla executives has no fortitude. They prefer revenue rather then invention and what's best for the user. KaiOS is becoming the third biggest mobile OS, guess what it's FirefoxOS, but Mozilla was too afraid to give it a shot. Then there's the Rust programming language that is taking the world with storm. It seems there are great talent, and if they would be allowed to work in the user's best interest people would switch over from Chrome - and Firefox would become big enough to matter. As for revenue, a lot of purchases are initiated from the web, but they leave the browser for a short while and takes a 3-5% cut. Browsers could work with banks and offer a secure wallet. And micro-transactions could become a thing. Publishers are crying for a solution! The web have been funded by ads for over 20 years now, with diminishing returns. And users hate it! The web is ripe for invention!


> RSS and bookmarks bypass Google

Bollocks!

Google - and any search engine - cannot help you find an exact web-page you found after hours of researching while web-surfing earlier.

And RSS feeds are for when you’re already interested in a content source. Google searches help you find something new: they won’t help you automatically be informed of new posts. They just save you the time of having to manually sort-out new content from the old when you visit an article website.

Google isn’t to blame for the drop in popularity of RSS (Google Reader’s closing was a symptom, not a cause), it’s the content websites’ webmasters who saw that by allowing machine-readable access to their content index means that users wanting to get to their new content can bypass the advertising on their home-page, effectively halving the pageviews and thus halving their revenue - or if they included their whole article content in the RSS feed then they’re missing out on potentially all of the advertising revenue - that’s why some content authors, like Daring Fireball’s John Gruber, as an example, only provide their full RSS feed to paying subscribers.

RSS still works for podcasts though - as podcasts wouldn’t be popular at all if people had to navigate through a webpage to download each audio file each time a new release is made - so the halving of web banner ad revenue is compensated-for by having a much larger audience for the in-audio advertising baked into the podcast content.

Twitter - and centralised content platforms like Facebook also was/is a major part for the reasons I described above: allowing direct access to content means less pageviews. Somewhat concerningly, we’re seeing people use Twitter to do things that RSS was originally designed for: such as posting links to new articles posted to a blog or for things like live service uptime status updates.

Finally, there’s the usability issue: it’s difficult to describe what RSS is or why it’s good to a layperson. Ssure, today we can just say “an RSS feed is just like a podcast, but for normal web content, or anything at all” - but back in the early 2000s when RSS awareness (or hype...) peaked, I had difficulty understanding what a “syndication feed” was - the terminology “feed” implied to me it was a unidirectional continuous push-style connection (like a HTML/HTTP EventSource) - not a pull-style index file. Don’t forget the format-war with Atom too.


> that’s why some content authors, like Daring Fireball’s John Gruber, as an example, only provide their full RSS feed to paying subscribers

This is false. I am not a paying subscriber, but I still get the full content of Daring Fireball articles in my RSS reader. In fact, the RSS feed is one of the links on the site’s sidebar. https://daringfireball.net/feeds/


That's not the "full feed", that's $19/yr and mentioned here: https://daringfireball.net/members/info

> However, paying supporters do get access to a few members-only perquisites, including separate full-content RSS feeds for articles and the Linked List (my daily list of links and blurbs related to Mac, web, and design nerdery).


I don't understand why they removed support for it. Isn't RSS a "solved issue" - what possible updates can be made to it? Why couldn't they just keep it available and forget about it.


The functionality was never polished, and there were some fairly serious technical problems with the implementation due to lack of maintenance, seen often in apparently minor things like https://bugzilla.mozilla.org/show_bug.cgi?id=337897.

I can readily understand why they removed it—it was implemented in what had become the wrong way for such a feature, and fixing it would have taken more effort than they wanted to expend on such a niche feature, and it was starting to hold back other improvements. (Similar deal to why they broke old extensions: they were holding the browser back technically, and a couple of years later I think it was fairly clearly the right decision, painful though it was.)


> The functionality was never polished, and there were some fairly serious technical problems with the implementation due to lack of maintenance, seen often in apparently minor things like https://bugzilla.mozilla.org/show_bug.cgi?id=337897.

Then that is a great justification for improving it then. Or at least bundling an RSS add-on. Mozilla felt that bundling Pocket was ok, but not one of the many great RSS add-ons?

> I can readily understand why they removed it—it was implemented in what had become the wrong way for such a feature, and fixing it would have taken more effort than they wanted to expend on such a niche feature, and it was starting to hold back other improvements.

The truth is, they did not see it as an essential part of the web, worth implementing. If you look at the other stuff that Mozilla is allocating resources for, it becomes clear that maintaining RSS support would be a drop in the bucket.

RSS could have been fixed for a fraction of the cost of one of their many dead-end research projects, or they could have swapped out the canapés for a cheaper finger food at one of their events.

The fact that anyone would call RSS "niche" is part of the problem; something can still be important even if not a lot of people use it at the moment. But that kind of nuance is something lost with this toxic market-driven mentality. Accessibility features are also "niche", should they be removed? Many people consider RSS to be a type accessibility feature. Should all accessibility features be relegated to add-ons that must be manually sought out and installed?

I don't know the details of the high level decisions at Mozilla, but I also can't help but notice that all of the decisions made by Mozilla seem to align perfectly with the interests of companies Facebook, Google, Apple, Netflix, Amazon.

Sending alerts to users and getting them to navigate to your platform is what big content wants. Mozilla agrees that that is the future and that RSS is obsolete and apparently holding them back. Mozilla has no problem implementing whatever Google wants and always being behind Chrome. Mozilla also decided to legitimize web DRM with its embrace of EME.

> (Similar deal to why they broke old extensions: they were holding the browser back technically, and a couple of years later I think it was fairly clearly the right decision, painful though it was.)

Honestly this is a whole other discussion, but I would dispute that they made the right decision. If browser extensions are just toys to you, then I'm sure you appreciate how streamlined and simple they are now.


Mozilla should be focusing on RSS within Thunderbird, because that was the original location for RSS functionality, and is a much better fit anyways (due to the expected presence of an almost always open left hand sidebar.

Unfortunately Mozilla has dropped thunderbird almost completely so that’s not been a real option either.


I completely agree. I never actually used Firefox as an RSS reader, but I did get my mom started on RSS through Firefox, and later she moved on to a dedicated RSS reader. I don't remember, but I think that Firefox would offer to subscribe to a feed via Thunderbird if you didn't already have one set up.

But I think Mozilla's abandonment of Thunderbird is very much in line with their abandonment of RSS, and their loss of commitment to an open web.


Slight correction where I was unclear: I don’t call RSS niche, but RSS reading within the browser itself.


Yea it competed with Pocket, that is the actual reason it was removed


This article is completely wrong. Slashdot, Digg and Reddit were already at 100% of full-power way before Google Reader ever shut down. And him citing that the causes are unknown is so completely naive. It's obvious. And it was obvious at the time to many who used RSS in 2012. Google is in the ad business, and RSS doesn't do adds. Not really, and not then. They realized they were competing with themselves and closed it; it's that simple. God, I feel like I got baited to reading that post because there was no new or insightful information whatsoever other than the title line. We should bring back RSS, and make the web more about conversations and communicating, than listicles and click-bait. 5-10MB of Javascript per page load, 1px tracking images, endless stupid ads, and now every single site that I go to has a pop-over that I should sign up for something, which gets in the way of the content that I am only going to spend 30 seconds reading anyway and then never return to that site ever again. The web has very quickly become a cesspool of non-information. It's like a bad shopping mall.


I completely agree except that I think google did it intentionally. They did a fine job shutting down other things in similar ways like newsgroups.


> listicles and click-bait. 5-10MB of Javascript per page load, 1px tracking images, endless stupid ads, and now every single site that I go to has a pop-over that I should sign up for something, which gets in the way of the content that I am only going to spend 30 seconds reading anyway and then never return to that site ever again

This exactly describes my experience on this website.


I would like to take one comment in this post to recognize Aaron Swartz https://en.wikipedia.org/wiki/Aaron_Swartz


What's his connection to RSS?


It's literally in the first paragraph of the Wikipedia link above...


RTFA has never been more appropriate.


From the link: "At age 14, he became a member of the working group that authored the RSS 1.0 web syndication specification."


RSS seems to me to have use cases far beyond website updates, if it was extended a little.

Event syndication. Say that I'd collect the event feeds from a load of cinemas, music venues I like around the world, why not, plus those of musicians. I want my RSS-based events reader to narrow down the date field to be this weekend, location to be my town, and ticket field to be available, and why not price to something I can afford, while I'm at it? Bam, everything I could dream of doing this weekend, no Facebook and using a slightly modified version of a two-decade-old tech.

Similar functionality for shopping.

Why couldn't RSS be extended to something like this?


It's really not that RSS couldn't handle (/be extended to handle) this use case, it's that the parties publishing this information do not want you to have that much control over the feed.

See how much effort Netflix puts into making their catalog hard to browse, to obfuscate the actual size of the catalog and promote specific things they need to show a large return on, or all the sponsorships/ad-deals/promotions that inevitably begin to clutter almost any commercial news feed.

We have the technology, but publishers will fight tooth and nail to keep control over the platform away from the end user.


Remember the heady days when Twitter, MySpace, Facebook, et all had almost open access to their 'social graphs', allowing people to grab content from one place, post to another, build connections using tools like Yahoo Pipes? I sometimes long for simpler days when companies were more open with their data. Didn't Netflix even have a more open api back in the day that you could rank and sort, and see new release dates, etc?


You realise what the most successful thing built on top of that open social graph was, right?

Cambridge Analytica.


The issue there was collection of massive amounts of public data by Facebook. Open data APIs can be used to build nice things, if the data isn't user information.


No, it was Cambridge Analytica using the old FB social graph API ("Open Graph") to build a network of people that they could make predictions on.

The app "This is your digital life" let people do a psychological profile linked to their FB profile. The Open Graph API then gave Cambridge Analytica to (some of) their friend's data as well,

https://www.cnbc.com/2018/04/10/facebook-cambridge-analytica...

https://medium.com/tow-center/the-graph-api-key-points-in-th...

https://about.fb.com/news/2018/04/restricting-data-access/


The problem here is that we don’t really have a consensus on what should happen if people share the information they know about you with others without asking you, not that we should not have API access to our own content on social networks.


No particular argument there - just noting that the original comment was calling for return to "the heady days when Twitter, MySpace, Facebook, et all had almost open access to their 'social graphs'".

Turns out that wasn't a great idea. Disappointing because it's what I wanted too.


But Netflix is different from a gigging band, venue and cinema, whose financial models are based on actual ticket sales not whatever the hell Netflix's is. It takes so much labour and money for them to get eyeballs filtered through social media and paper ads and to attract customers to their sites. So for them this would simply be almost a free extra way to drive traffic.

I think you're making a valid point about Netflix, but not seeing how it applies to more traditional financial models?


The parties in that case would be the venues, cinemas, concert halls, etc. They don't to control any platform, they want that information to be freely available. They want lots of sites to show case them (remember, comoditize your complements).


This is it exactly. Realtors sued Zillow for essentially publishing MLS data. They lost, thank god.


From one direction 99% of the world's VCR clocks blinked "12:00" for the entire lifespan of that media format. The proposal is a lot of interactivity and cognitive load to demand from most of humanity. Only a very small segment of society, mostly engineers, can utilize parametric search. Even engineers are lazy or in a hurry sometimes, in theory investing time in a parametric search would benefit me, in practice I needed a USB cable and going to amazon for a generic search of "amazonbasics usb type-c cable" works well enough.

From the other direction there's not much middle ground between the proposal and a REST API. You're asking for developer.ebay.com, I've fooled around with that a little and its fun. Sometimes I think the business people don't understand how much the devs are exposing in their APIs, which makes me worry about the staying power of public APIs. There are businesses where their business model and front-end UI could all be replaced by a very small shell script and I don't think the business people understand that weakness. Of course an API can be shut off with the flick of a switch once it eats into profits.


Yes, yes, yes! This is such a good idea. Working at a University, I’ve often thought this would be the ideal way of advertising seminars. I suspect the problem is chicken and egg, the technological barrier is just a little too high for this to be implemented when RSS readers are not widely used.


You can publish an RSS feed for shopping search results.

You can put a future date in RSS for events.

To lazy to search for examples sorry.


Customizable HN feeds (inofficial): https://edavis.github.io/hnrss/

It's really easy to get posts on the frontpage only if they have more than x points:

https://hnrss.org/frontpage?points=x

Or contain certain keywords:

https://hnrss.org/newest?q=git+OR+linux


This is awesome, configurable RSS feeds are critical for large-scale publishing platforms, thanks for sharing!


I use this every single day.


As one that loves and RSS and hated that many websites don't offer them anymore, I created a middleware that transforms the static HTML of most websites to an RSS/Atom feed. Its just a proof-of-concept, but maybe you like it :)

https://github.com/damoeb/rss-proxy/


This looks interesting, thanks for sharing the link! I work on a project that's somewhat similar but users have to be explicit (using CSS selectors) about the elements that will be used to create the feed.[1] I like that yours appears to try to pick out the best elements without user input.

[1] http://createfeed.fivefilters.org/


damm that what i was looking for?


The reason RSS failed to reach mainstream adoption by users is because it is not user friendly at all. While I love RSS myself, no amount of tech nerd nostalgia is going to make it popular enough that your mom starts using it.

Most sites still have implemented RSS in a terrible way. For example, many blogs I follow only show excerpts in their feeds. So the feed is worthless to me. Others put every podcast episode they do every day in between their posts. Annoying and worthless.

Then, if you want to follow a site that publishes a lot of content, often you have to subscribe to everything or nothing. Sorry all mainstream tech news sites. I don’t want to read 1,000 low quality articles every day.

Then comes the UX nightmare of actually finding the feed on each website you visit. If the site even has one.


> The reason RSS failed to reach mainstream adoption by users is because it is not user friendly at all. While I love RSS myself, no amount of tech nerd nostalgia is going to make it popular enough that your mom starts using it.

I think it is a false premise that something is only valid if "mom starts using it". That is the profit-driven mentality. eg. "how can we market this? How can we expand RSS market share into valuable demographics?" etc

Also, I actually did teach my Mom to use RSS a decade ago and she still uses it today.

These concepts are really not that hard. I told my mom it was like she was getting a newsletter from her blogs delivered to a special dedicated inbox, but without cluttering up her email. She was delighted.

I think if someone knows how to use email, and knows how to browse the web, and knows how to sign up for email newsletters, they can handle RSS. I would argue that it is in many ways more useful for less computer literate people.

> Then comes the UX nightmare of actually finding the feed on each website you visit. If the site even has one.

This UX nightmare was solved 15 years ago. Browsers displayed a little icon in the corner when RSS is detected.[1], Firefox later displayed an RSS icon prominently in the address bar[2].

The UX nightmare was then re-introduced as the RSS icon was de-emphasized[3][4] and eventually dropped completely, with a dubious justification.[5]

[1]: https://www.hanselman.com/blog/FeedAutoDiscovery.aspx

[2]: http://scripting.com/images/2011/01/15/rssicon.gif

[3]: https://decafbad.com/blog/2011/01/15/how-to-use-feed-auto-di...

[4]: http://scripting.com/stories/2011/01/15/mozillaPleaseKeepThe...

[5]: https://news.ycombinator.com/item?id=17613051


I certainly agree, it's not that hard.

Most things are not that hard. The problem isn't the difficulty level. It's friction.

I mean, cooking healthy food for yourself and vacuuming your floors are both cognitively easy things to do. But most people don't have enough motivation to do these things all the time. It's why meal kits and the roomba exist.

The minute you introduce the slightest bit of friction, you lose people. RSS contains enough friction to remove a 95%+ of potential users before they even get started.

The reason why people prefer social media newsfeeds is because they have zero friction.

I'm not arguing that the problem with RSS is that it should be driven by a profit mentality. I'm arguing the problem is 95%+ of people will never benefit from what it can potentially offer the world: a better way to consume the internet.

I personally love RSS. Alongside email, it's my preferred method for reading the internet. However, I think the world would benefit more if the RSS ecosystem could be made viable for the average person. Whether that's a better protocol or a better client, I don't know.


High-volume sites are not a good use case for RSS. We already have various social ways to filter. Like the site you're on. I think RSS should be for high-quality niche sites where you care about every single post.

I subscribe to 200+ feeds, but only read a dozen or so stories a week on them: http://akkartik.name/feeds.xml


Oh, I have to disagree, I think high volume sites are a great thing for RSS. In fact, I tend to think the more items in a feed, the better. No, I usually don't go through all of them at once, I'll go through several hundred at a time.


What kind of reader can I import this into? I'd like to get back into RSS.


Most of them, I hope! It's a pretty standard OPML file.


> actually finding the feed on each website you visit

The site should just have

  <link rel="alternate" type="application/rss+xml" title="RSS Feed for Bob's Blog" href="/rss/" />
Does that not work anymore? Or sites don't include it? I haven't used RSS in a few years.

Back when the address bar had an RSS icon in it, this is how it got the info.


Not all sites that offer feeds actually put the appropriate link tag (things that come to mind: youtube channels, reddit, a lot of wordpress blogs, ...), which is extra annoying. It actually takes a lot of heuristics to be able to say with some sort of certainty whether a site offers a feed or not.


Other than the issue of posting abridged content, it seem like all the other issues can be handled with a capable client. It's better to post more (inlc. the podcasts, etc.) and then filter as you desire, no?


Whenever you put the burden on the end user to endlessly customize everything you’ve just lost 95% of the mainstream public and [insert thing] remains a niche tool used by people on HackerNews.

...and then mom still ends up getting her news from Facebook.


5% of users is fine -- nobody is saying that RSS should be the only interface to the internet-at-large. We just want to enable the 5% (or less) of users that are invested enough in the process that it's worth it for them to get their specialized content efficiently, but not lose out on valuable content.


It doesn't solve everything, but WordPress will generate an RSS feed for most pages. Just tack /feed/ on the end. You can follow tags and categories.


If anyone wants to help, here's an open source project I worked on quite a bit: https://github.com/GetStream/winds Goal is to build RSS for regular users instead of the power user audience that RSS readers tend to cater to. I think this is part of the problem. The market for RSS shrank. All commercial RSS readers focused on the people who pay (IE the power users). Creating a user experience that is just not viable for most consumers. You end up in this vicious cycle because of that. RSS usage drops, RSS readers become more power user focused, sites drop support, continue the cycle.


I don't mean to sound harsh but if a user is either required to run their own React/NodeJS (which is not going to happen for your stated target audience of "regular users") or to use a centrally managed RSS service that could go away at any moment... doesn't this defeat the purpose of syndication / federation a little bit?

If


Yup, also picking up the tab to host it: https://winds.getstream.io/

Not sure if I'll keep on doing it though, bit expensive for a side project


To reroute much of web-based feeds consumption to RSS - check out https://docs.rsshub.app/en/

It works quite well, has vibrant community, and support for ripping twitter and instagram feeds among dozens of other sites. It also has a browser extension to help discover available feeds on sites that it can digest - https://addons.mozilla.org/en-US/firefox/addon/rsshub-radar-...

There's also a Chrome extension, but that one is not translated from Chinese yet.


I'm with the author in needing better reading tools, but the 'firehose' of RSS isn't by itself the answer

aggregators solve a different variety problem vs RSS -- RSS gives you access to random sources that you curate yourself, whereas HN or link-heavy blogs give you access to a meaningful amount of high-impact articles from high-diversity sources (i.e. more different websites than you subscribe to in feedly) that everyone else is reading

at minimum, I need a tool that lets me tame the RSS firehose with some kind of ranking or priority queue, plus mix in some aggregator reading so I don't miss things


That was why the friend-of-a-friend social features of Google Reader were such a big deal for me — I got curation and aggregation from a person I trusted, and the people they trusted.

I didn't have to subscribe to anything to get content, I just had to follow people I knew, and even if they weren't sharing a lot, their first-level connections collectively shared plenty. The interface let me subscribe to whatever feeds they were sharing from, which is how I discovered a lot of content I never would've on my own, and in a lot of cases likely not through other aggregation methods either.

Add the content they were clipping content with the bookmarklet and even sites that weren't syndicating their content were getting my regular traffic via shares.


+1

all kinds of trust networks are good


Devil's advocate: A public trust network is great for advertisers. They would know exactly who the influencers are and how to cover the population with a minimum of influencers.


If influencers choose to work with advertisers, then jeopardize their circle's trust due to advertiser affiliation, they would lose that trust. It self-corrects; as long as the system doesn't reinforce the network effect.


I half-agree

the accidental trust networks that dwell in existing social networks (fb, twtr, IG) are gameable

a trust network where real-life experts gain and lose reputation based on their behavior should (in theory) punish corruption


RSS works pretty well for me. Every single website I follow supports it, both the mainstream and the niche ones. I use on my desktop and phone NetNewsWire, which is a great, fast and simple piece of software. Overall, I’d say RSS is not the latest and greatest anymore, but it keeps working for me.

In fact, I use RSS for this website as well.


I remember that brief time when free email addresses were being offered up before RSS was created. You could get daily news emailed to your inbox from such as USA Today, Wired, and so forth. An electronic newspaper tossed at your digital front door every morning. I actually looked forward to it.

Then RSS was invented. Now RSS is supposed to be dead, and/or killed, what appears to be several times over. I never did get on the Google Reader bandwagon because I was subscribing to RSS feeds via other means.

Nowadays, I have the app ‘Podcast Addict’ [free with ads, $0.99 once for no ads; I'm just a user here] installed on my tablet through which I subscribe to over 50 RSS feeds – podcasts, webcomics, even a comedian’s “upcoming shows” feed. There are countless other RSS feeds available via in-app search and discovery let alone the “add your own” option.

By missing out on an RSS apocalypse folks seem to love to talk about from time to time, have I accidentally become one of those “welcome back” guys?


I felt very similarly.

If you’re interested in getting “the news” delivered to you daily, instead of having to check a reader app, I’ve built something that aggregates all your feeds (and websites without feeds) into a daily digest via email.

I don’t want to “spam” so I won’t link here, but you can check on my submissions for “News, calm”.

On the “RSS apocalypse” front, I can say that on this product I noticed that most websites “non-tech people” follow don’t offer RSS feeds, apparently over monetization concerns.

I was definitely in a filter bubble as everything I followed had RSS.


We should go more old-school and resurrect NNTP instead: Have clients that render Markdown, and servers where users can create their own access-controlled groups.

Have a group where only the owner may post: That's a feed.

Have a group where only the owner may post top-level articles, but anyone may respond: That's a blog.


I had the same ideas about feeds and blogs as you have, and I agree.

I have seen the idea about Markdown before, and have considered that too; it is not a bad idea. However, clients should only render Markdown if the article contains a "Content-type: text/markdown" header, and assume plain text otherwise. Clients that don't understand Markdown you can still read the article, since Markdown is meant as readable even when not rendered, so that is good, too.

I even wrote a simple NNTP server software and NNTP client software. Both programs are based on SQLite.

I have not considered your idea of servers where users can create their own access-controlled groups, although that doesn't seem so bad; some people may wish to run such servers, and may have some way of users controlling them.

Also note, you could have multiple interfaces if wanted: HTML, RSS, email, and NNTP, with NNTP as the primary and preferred interface. (To these, Gopher can also be added for read-only access, if wanted.)


I have not considered your idea of servers where users can create their own access-controlled groups [...]

One of the properties of traditional Usenet I don't like is that creating groups isn't a matter of minutes, but a matter of days or weeks. If we want an NNTP-based alternative to blogs, twitter and reddit, we can't have that. Instead, a user should just get control of a namespace managed by the server he's registered with.

Also note, you could have multiple interfaces if wanted: HTML, RSS, email [...]

That's a very important part of the equation: The server should auto-generate HTML pages and RSS feeds for the messages to allow seemless integration into the WWW. Let the power user use their news reader, but don't deny access to the 'layperson'.


Namespaces is an idea that I have considered, and I have defined a kind of namespacing called "Unusenet", which uses "un" followed by the number of components of the domain name that the namespace belongs to, a reverse domain name, and then the rest of the components. The domain name is normally the domain name of the server where the newsgroup originated; if it is then federated with other servers, they use the original domain name for that newsgroup and not their own. (For example, a newsgroup originating on "example.org" named "misc.xyzzy" will have the full name "un2.org.example.misc.xyzzy".) This is what I have on my own NNTP server. (You do not have to use the Unusenet convention, but I use it in order to avoid namespace collision.) However, I currently have no "feeds" or "blogs", but only "forums"; if I do implement a blog then I would probably implement in that way, though (and as I already mentioned, this was even my idea before you mentioned it; we just happen to have the same idea). But I run my own server rather than using those of others. Of course, not all users will want that, but that is why they can have a server where users can create their own access-controlled groups in a namespace in a few minutes easily like you mentioned.


It would be nice to see RSS come back and replace some of the centralized web. In theory social networks could be replaced by event feeds your friends publish. If you use something like IPFS you could, in theory, have a fully decentralized social network.

There's still some business benefit to hosting, scraping, providing search etc. The business models are closer to the open web.

I'm sure there's flaws and challenges in the idea. Even if private posts were encrypted you'd still leak some info publicly. Still I think its a neat idea.


That seems like the idea behind the ActivityPub protocol that Mastodon and other similar social media platforms are based around.


I wrote this just yesterday in another thread but I'll say it here as well:

I think RSS should be better built in to browsers. Make it as simple as a follow button is on centralised websites.

e.g.: A little RSS icon pops up when RSS is available on a page, press it and you're now following that feed. Feeds window in the browser shows your feeds, and a small alert icon shows up somewhere when there's unread content. Your subscriptions are saved to your account. If you want to read the article, you click the link and read it directly on the source website. It doesn't need to be any more complex than that.

Firefox had "Live Bookmarks" for RSS but it was relatively terrible, and eventually got removed.


I never stopped using RSS.

But man -- what a testament to how much I loved Google Reader that it still feels like a fresh wound to have it brought up again.

Damn you, Google!


Feedly seems nice but it's a total non-starter due to the lack of an email-based signup. I have none of the social media accounts it wants you to log in with and it'll be a cold day in hell before I share that much personal information with an RSS reader anyway. But a brief reading of the privacy policy will tell you that Feedly is a data hoarder and has every intention of selling that information:

https://feedly.com/i/legal/privacy

$6-$18/mo should be fucking enough for a service to keep my goddamn information to themselves.

I'm using FeedReader[0], but I'm not especially happy with it. Would love to hear some more recommendations.

https://jangernert.github.io/FeedReader/


Given that one of the options for Feedly is “Continue with Feedly”, which offers a “sign up” option where you provide a name, email, and password, I’m not sure what you mean about forcing social media access.


Thanks, somehow I missed it. But with a privacy policy like that, I'm still a hard pass.


I use, and recommend Feedbin[1].

The interface is pretty good, privacy is respected as far as I can see, service is hosted on their own hardware, not in the cloud, the author is very responsive when issues arrive, and the code is available for self-hosting[2].

It is also how I keep up with people on Twitter, and YouTube, without having to access Twitter or YouTube.

[1] https://feedbin.com/

[2] https://github.com/feedbin


This is Petr from Feedly, please let us know where in the Privacy Policy you read we're a data hoarder and selling information? We're doing the exact opposite and we can afford it with Feedly Pro, Pro+ and Business as they feed our business so we can stay profitable without the need to sell any customer data.


> We may share some or all of your personal information in connection with or during negotiation of any merger, financing, acquisition or dissolution, transaction or proceeding involving sale, transfer, divestiture, or disclosure of all or a portion of our business or assets. In the event of an insolvency, bankruptcy, or receivership, personal information may also be transferred as a business asset. If another company acquires our company, business, or assets, that company will possess the personal information collected by us and will assume the rights and obligations regarding your personal information as described in this Privacy Policy.

You collect data, use it for marketing, then add it to your valuation. Says everything I need to know about your company.


Good point, this is valid only if a merger or acquisition happens so we're not selling anything or using your data in any way to give/sell it to 3rd parties.


Feedly has email sign up. The button is labelled "login with Feedly" instead of Login with Google ect.


The business model of RSS is harder than just slapping ads on your homepage and encouraging more page clicks. I think that's the real reason you don't see a strong RSS ecosystem. The content producers don't have a strong financial incentive to support it well.


In my experience RSS is at its best when dealing with infrequent publishers or niche audiences. People who don't expect to be paid for their writing because they want a wider audience, like think tanks or university departments.


That's definitely where it shines for me. I follow hundreds of infrequently-updated sites using Newsblur. It's sort of the same value proposition for me as Twitter. I'd never follow the New York Times in either place; what I want is to get beyond what's commercially viable to publish.


The nice thing about RSS is a lot of sites offer filtered feeds. I only follow a few feeds from the NYT, for instance.


If we get back to RSS, can we at least all stick to Atom 1.0 :) ? It's way stricter, simpler and easier to use, and I never found a parser that was not accepting it.


I strongly agree: Atom is technically substantially superior to RSS, and in regular feed readers is supported universally (with fewer issues, e.g. fields like title and description are definitely plain text or HTML, as specified, rather than the RSS approach which leads to clients guessing all sorts of different things so that you can’t safely use things like angle brackets in titles).

However, the podcasting industry seems to have ignored Atom, which is stupid. Podcast feeds are exclusively RSS, and from information found, most major clients probably don’t support Atom. But the area is a mess with no good documentation anywhere on what works or doesn’t (and all of the even-halfway-decent content of this sort is from 2006–2010).

————

People colloquially refer to feeds as RSS, even when they’re mostly Atom. Reminds me a bit of the SSL/TLS situation (where I think the name “TLS” is finally more popular than the name “SSL”).


When people don't use extra namespaces and everything is consistent, Atom 1.0 works great. And I understand why namespaces exist, but they create this system where each site speaks its own language and that fundamentally undermines the whole point of RSS as this shared way of communicating.


Atom has one annoying characteristic: It requires a self link to be valid. For static website generators it means you configure the domain for no reason except Atom validity.


Haven't seen newsboat be mentioned so far. I've written a bunch of scripts that help subscribe to feeds (e.g. search YouTube for keyword and add RSS feed for channel of the top hit), scripts to curate (e.g. extract 'topics' from BBC articles that are only available on the page and not in the RSS feed) and consume (watch videos with mpv, open images in feh, add long videos to a backlog).

It's one of the best news experiences I've had and is an improvement over what I was used to with Google Reader and Feedly. I feel much more in control of my content consumption.


Care to share your scripts? Avid newsboat user here also!


I used commafeed with various categories and have a mixture of websites and YouTube channels. It was a game-changer for me when I discovered that you can “subscribe” to a YouTube channel’s videos as an RSS feed by pasting the channel URL into a feed reader. All of a sudden YT became usable again.


They're very messy - they've built up slowly over the last year or so. Will add some comments over the next few days!

https://github.com/daharka/newsboat_scripts


I'm moving from Feedly to self-hosted Tiny Tiny RSS soon: https://tt-rss.org

Would be fun if others did the same and posted about their experiences!


I have been using self-hosted RSS aggregators for some time now (currently [FreshRSS](https://freshrss.org/ )) and I have been delighted ! I feel it complements HN or Reddit really well in giving me the ability to follow specific people or projects.


I moved to self-hosted tt-rss when Google Reader shut down and the 50 or so other clients/services I tried were not good enough. Been using tt-rss ever since and very happy with it.


I used tt-rss for a long time, but about 6 months ago I moved over to FreshRSS.

But either way, self-hosted RSS is the way to go. Nobody can take it away from you...


Why did you move? What's your experience with FreshRSS been compared to TT-RSS?


Honestly, it's hard to articulate or even pinpoint one thing. I like the layout and options of FreshRSS, and a few little things; for example, I use categories, and when I hit 'mark as read' on a category in tt-rss, it would indeed mark everything as read, including items that had refreshed behind the scenes since the last time I opened the category, effectively marking items as read that I hadn't seen yet. FreshRSS seems smart enough to realize that those items haven't been displayed.

Also, tt-rss' developer, fox, is rather abrasive, akin to Theo de Raadt from OpenBSD (which I love and will always use, despite Theo's abrasiveness). I will admit, the projects do generally benefit from this unyielding approach, but it makes it a chore to request help, knowing that you'll likely be upbraided for asking questions.

I also like that FreshRSS has better 3rd party app support (at least on Android).


Interesting, thanks for the link. I just noticed it is also packaged into Debian.


RSS is the only way I consume content regularly -- if Innoreader can't see a feed at a site that I may come across, I won't be visiting regularly. Too much of a PITA otherwise. No Twitter account either -- I don't want a "push" model, and the S/N ratio is way too low...


Totally agree. If I can't add an RSS feed for a site to Inoreader then I simply don't visit the site. RSS is the single thing that makes sites like CNN usable to me.


Same. A site having a feed is much more likely to get consistent recurring traffic from me than a site I can only consume via social network posts.


I turn those into feeds too.


If self-hosted is your thing I highly recommend tiny-rss and the iOS app accompanying it. https://tt-rss.org/


I came here via ttRSS, with the Android app. I host it via the hosting service I've used for years, it only requires PHP and a Cron job. The Android app lets me keep ttRSS and its authentication behind a separate HTTP basic auth, which I think keeps the attack surface low and security high-ish.

I think it's fantastic.


You mean Android app? There's no official iOS app. A decent iOS alternative seems tiny Reader.

https://apps.apple.com/us/app/tiny-reader-rss/id689519762


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: