Aral Balkan @aral

Reclaiming RSS

“Before Twitter, before algorithmic timelines filtered our reality for us, before surveillance capitalism, there was RSS: Really Simple Syndication … As we move away from the centralised web to the peer web, it’s time to rediscover, re-embrace, and reclaim RSS.”

ar.al/2018/06/29/reclaiming-rs

· Web · 271 · 221

@aral “The more ways people have of consuming your content, the more resilient that content becomes and the more freedom people have. Duplicate content? Yes, please. The more the better!”

But exactly. I've never stopped using it, and have always made a point of providing full RSS (vs. partial-as-bait-to-make-people-come-to-the-site).

<3

@aral RSS didn't go anywhere

2008 internet still exists, people just let themselves be blinded by a shiny birdsite. reader alternatives cropped up almost immediately after Google pulled the plug.

the real problem is that foss is really bad at marketing imo

@xj @aral Right. And that FLOSS doesn't even remotely compete here anymore: People using Twitter or Facebook don't want to choose a certain software - they use a turn-key service instead that of course in some way relies on bits and bytes but has its advantage in *just* being available without any learning or administration effort. From that point of view, FLOSS that requires you to think about hosting, backup, availability, ... is just set to lose. :(

@aral

For anyone using #Friendica, you can follow any RSS feed as if it was another user, just add the RSS feed's address as a new contact.

@switchingsocial PS. Love what you’re doing with the site :)

@aral Should we take on reclaiming #RSS, or should we rather try focussing on technologies such as #activitypup to replace #RSS? For what I remember, one of the total drawbacks of #RSS was that it wasn't in any way interactive whereas Twitter and related platforms allowed for comments and discussion starting with day one.

@z428 ActivityPub isn’t a replacement for RSS (it is orders of magnitude more difficult to implement). We can embrace both (and more – like datproject.org).

@aral I know. :) Just wondered whether, in an #activitypub world, #RSS is still required - except for being a "simpler" standard, does it provide any features that can't be done with #activitypub as well? I wonder because right now I see a lot of blogs moving away from dynamic web hostings and quite some of them end up offering #RSS for syndication - and embedding platform such as #disqus because comments.. ;)

@z428 Yeah surveillance devices like Disqus are a problem. I see RSS existing alongside. Doesn’t have to be either/or. Well suited for news.

@aral The techie in me agrees wholeheartedly. Looking at end users however (the crowd that still mostly remains on Twitter and seems quite hesitant when it comes to changing tools), I wonder whether too many alternative technical approaches at hand are a benefit or more of a hazard. #fediverse seems a viable, agreeable solution so why not fully focussing on that at least for a while? ;)

@aral @z428

The advantage with RSS is that's it's just a file and can trivially be done with a site generator and a static web host. AP requires a running application that distributes the posts to the followers, the equivalent of RSS+WebSub.

I would love for there to be implementations of the RSS+WebSub-equivalent subset of AP for a whole bunch of languages.
@aral @z428 But I would also love for our OStatus-compatible microblogs to never stop supporting OStatus, and start supporting Atom without WebSub, like Friendica does.

@xj

@notclacke @aral Sounds valid. Well, maybe at the heart of things, I'd like to see a *meaningful* mix of protocols providing an end-user experience as good and well-integrated as somewhat possible (Twitter and Facebook are the benchmark here) while reducing the need to make decisions between different implementations of the same "use case" to a bare minimum. There's "always more than one way to do it", but there shouldn't be too many... ;)

@z428 @aral With Disqus, Google+ and spot.im using an URL as the key for a discussion, all we need would be an RSS reader, that implements these discussion sites. And these sites allowing access to the comments from multiple accounts (e.g. if the website owner also implemented Disqus already - don't know if you could still see the conversations from a different account).

@aral yes, #Twitter , in terms of functionality ... it's just " Commenting  #RSS with asteroids " ?

@aral

Yes! Starting to see more people talking about the usefulness of RSS.

@aral Thanks for your post! I think it speaks to the best part of web culture how RSS is still around depsite the best efforts of large social media platforms to kill it.

Isn't it still good practice to put a <link> in your HTML too, to let RSS clients know where your feed is?

@edsu Yes, I’ll update the post to mention that; thanks :)

@aral Whao ! You exactly wrote the post I was thinking about. Do you mind if I translate (part of) it (into French) (with due credits of course :-) )

@Bromind Of course, I’d be honoured. All my content is under Creative Commons ShareAlike :)

@aral I've spent a bit too much time over the past few days writing a CSS sanitiser that reads from rsstail (using my newsboat url list)andfeeds xrootconsole, the xscreensaver phosphor hack, or a terminal, with ANSI highlighting and other bits.

Very hackish awk.

@aral RSS should stay gone - bring back Atom instead!

@aral

We didn't rss. Publisher did.

They want stats, ads, targeted ads, private data & engagement. All this stuff required their users to move from rss to social networks.

They first put only part of the content in the RSS then completely removed it after they could push content to their user smartphones.

Websites without ads or paywall have not stop providing full RSS. Why don't do read these ?

@aral Really enjoyed your article. It's dawning on me email newsletters are like a really crappy implementation of #rss.

@aral Firefox still has the RSS feed button, it's just not in the default UI.

Also some (mostly commercial) sites do deliver tracking pixels linked from their RSS feeds, which kinda defeats the idea of not being tracked...

Some still publish their feeds through FeedBurner, providing usage information for the Google surveillance machine. Fat RSS feeds do create quite a bit of traffic on popular sites, as each client reloads the full file on each access... No real incremental updates...

@galaxis @aral There are ways around it, though. Leaner feeds (don’t go back far in time), add pagination via tools.ietf.org/html/rfc5005#ap and make sure client and server use HTTP caching (simple use of if-modified-since headers and 304 responses). These already help a lot. The next step is caching the RSS to a file and letting the web server handle it. I had to do all this for Emacs Wiki, years ago.

@starbreaker @aral are they? 90% of the RSS feeds I subscribe to are for commits to repositories, notifications from issue trackers, and website service outage alerts.

@starbreaker @aral I'm loving the links you shared though - and I'll admit I'm fairly unique among folk I know who still bother with RSS.

@starbreaker @aral I was prepared to ignore a post with the title “How the Blog Broke the Web” but it turned out to be an interesting read. I don’t agree with the main point, but it was interesting none the less. Thanks for the link.

@starbreaker @aral

I don't like the second one. The issue in the headline isn't really explored, and instead the text is about a second Eternal September and is upset that somebody let the riff-raff in.

The first one I like better. At least it has a point, that hand-crafted sites mean that more of the soul leaks into the page. Not that it has become too easy for the plebs to publish, but that even the craftsmen aren't spending as much time on their texts, because the tempo is different.

The long texts are now in places like Wikipedia, with often high quality information, but with the soul edited away as imperfections.

It wasn't the blogs. It was centralization, corporatization (including large non-profits) and link aggregation with the associated clickbait.

Blogs to a large degree are still thoughtful long-form material, at least the posts I read. What's ruining the internet is low-effort shitposting. So I guess it was the microblogs. Oops, our bad.
@starbreaker @aral

I'm actually hoping that forms like #plume interacting with the Fediverse will bring back more long-form and commentary. I like the long things I write here the best, and I only write them here because (hey, I'm low effort too) I don't want to bother setting up a place to put them.

I quite like the idea of a single AP backend (Pleroma or whatever) supporting several front-ends, especially if that would involve metadata about the preferred front-end for viewing a particular post.

Now people have argued above, Aral for starters, that feeds should be full text and none of that luring people back to intended sites business, but I actually think in the context I just described that the way PeerTube does it is a pretty good idea. The AP version is a summary that fits the microblog feed, and it has a link to the full content with its intended layout. You could still have a fat Atom feed too, for non-microbloggy consumers.

I'm looking forward to high-effort writing, as in the creative process, distributed via low-effort posting, as in the distributing of the pieces.
@starbreaker @aral I have been wondering why blikis* never quite took off (but many Hugo sites are actually along those lines, with both chronological and non-chronological content, and a specific feed for the chrono content), and would love to see one that offered high customizability for the canonical view, and also an AP backend where you could just run a Plero or Masto FE for handling the comments, but also showing comments below the article, possibly simply with an iframe for the AP frontend (which would have to be adapted to that scenario).

* blog + wiki, your own personal wiki where you categorize your thought and iterate on articles across timespans of years, but also with wiki that relates the process of writing the wiki, and/or just random time-sensitive commentary, but preferably something that makes it make sense to host the blog and wiki in the same place.
@nonlinear @aral @starbreaker Over a decade old cluster of pages discussing the idea here:

http://wiki.c2.com/?BloggedWiki

Martin Fowler coined the word bliki, and still uses it
@nonlinear @aral @starbreaker I wrote http://wiki.c2.com/?MultiverseBliki 12 years ago. I did *not* "whip up an implementation" later that summer. :-D

@starbreaker @aral Yeah same with structured coding and linked libraries. The culture was so much better when we wrote everything in machine code. Kept all the riff-raff that couldn't grok the m68k ISA out, too. #goodolddays

@aral I find it mildly ironic that I am reading this blog post on my seed of your #dat site instead of using my #RSS feed reader, which I use for every other blog I follow.

@aral Great post! Although Firefox allows you to extract the rss link from any page which has it. On newer FFs, just click on Hamburger menu -> Customize and pull the "Subscribe" button to the toolbar.

@aral Feels ironic you posted this right after I took my first steps in supporting RSS in Odysseus.

Good irony though.

@aral I use RSS for years with Firefox and do search blogs for it. It is so easy to read the stuff I want.

@aral have you considered using a clearer term than "RSS"? For me, I know when you say "RSS" you're really talking about a group of related syndication technologies that include Atom, but I wonder if others reading this without the historical context might come to the mistaken conclusion that deploying more non-Atom feed setups is a good idea.

@aral More like others reclaiming it. I never left RSS. 😉

@aral The things that are happening now like ActivityPub, DAT etc reminds me of the semantic web days with PubSubHubbub, Microformats, Atom/RSS, FOAF(+SSL) etc but with newer technologies. Seems that the silo monopoly that replaced the semantic web is being demolished. Interesting times!

@aral @Gargron With a bit of Bayes-filtering and Mastodon, an RSS-reader could be a killer app for news reading.

@aral RSS is solid.
Rediscovering RSS is like when people are done falling over each other to build Slack integration for everything and someone is going to mention that everything could have hooked into XMPP just fine

@mjjzf @aral That's a good point actually: It could have been done for XMPP. But it unfortunately hasn't. Same about the desktop UI or the usability for, say, ad hoc group chats or sharing and finding inline media. A load about Slack is about trivial, obvious usability stuff. It all could have done with XMPP too, but Slack *did* it. And so a load of people use Slack (where a load of these features are just there) rather than XMPP (where a load of these features *could* be there). :(