Back to the Future with RSS
ncase.me
ncase.me
1. To detect RSS feeds and show an icon in the UI
2. Render RSS files, so they look better than raw text
3. Link that UI icon to a subscribe function: Show a configurable list of RSS Readers and send the feed to the subscribe URL of the selected one.
This can already be added with extensions, but it would boost RSS adoption if it were a default feature. And thus support the official mission of Mozilla.
Would mozilla block it if a RSS advocate group would provide the development work to get this back into the browser? Is there a technical issue, like that there is no fitting RSS/XML parser that can be used?
And it would be a huge marketing win for Mozilla in tech circles.
While thunderbird is a good reader, it doesn’t synchroniZe “already seen” across machines, and many feeds only have the latest 10 items in their XML feed, so unless you poll continuously, you are likely to miss updates.
I like to occasionally take a week off the world (without the laptop), and many of my feeds have more than 10 updates a week (some more than 10 a day).
I’m a 4-year happy paying customer of NewsBlur, which I wholeheartedly recommend.
Not true; maybe I wasn't clear enough in my last comment.
As an example, for platforms like WordPress that support pagination, if you miss some updates that were posted to alice.example.net/feed/, then you can add alice.example.net/feed/?paged=2 (and ?paged=3 and so on) to Thunderbird until it bridges the gap between items you had before going dark and what you have now.
It works for accessing a blog's backlog of content, too; it's possible to discover a blog today with regular (let's say weekly) updates stretching back e.g. 8 months and still get at the old ones. In fact, I've done this to go back and ingest years of posts after first discovering a feed only recently and then realizing that I've been missing out by not having it on my radar since the beginning.
https://addons.mozilla.org/en-US/firefox/addon/brief/
Source: https://github.com/brief-rss/brief
Also, for those not aware, Youtube still supports RSS feeds (for those without Google accounts) in the format of:
https://www.youtube.com/feeds/videos.xml?channel_id={Channel Id}
probably not... what did is Google having popups promoting Chrome every time you did a Google search.
Website owners could easily do that by themselves; it would be a trivial change.
FeedBurner:
<?xml-stylesheet type="text/xsl" media="screen" href="/~d/styles/rss2full.xsl"?>
Full-Text RSS: <?xml-stylesheet type="text/xsl" href="css/feed-free.xsl"?>
For anyone attempting this, note that there's an escaping issue on some browsers that requires Javascript to fix[3].[1] http://feeds.feedburner.com/johnpilger?format=xml
[2] http://ftr.fivefilters.net/makefulltextfeed.php?url=http%3A%...
Firefox and chrome had them for a decade, it did nothing. Beating a dead horse will not magicaly revive it. RSS is not dying because people have forgotten about. It always was an obscure feature for a low number of special people, and this did not change. And today it's even worse, because world has moved on, yet the RSS-crowd has barely adapted to it.
I'm using Newsfeeds for 15+ years now, and with every year I'm moving ,pre and more away from them, because that's not were modern live is anymore. We should start adapting and raise a new school of tooling for the purpose that RSS feeds, instead of playing around with yet another bunch of low effort-solutions in hope to find the magic potion.
If you wanna bring in new blood, you should deliver them tools where they are, to raise them up in their own world. But today tooling in RSS-space is so aweful bad and outside of peoples daily live...
RSS can become very messy super quickly, and I think we need clients that don't look like the ones we know from 15 years ago, but ones that help users to keep their subscriptions neat and clean.
To me RSS and the reader are and should be separate.
RSS is the protocol to query and retrieve updated articles. The reader can be anything you like.
Personally I use rss2email for retrieval and my UI is my email app.
There's a Reeder 5 and the two improvements (for me) that it has over 3 is that Twitter probably won't say "outdated browser" and refuse to show tweets and support for adding newsfeeds directly. I think MacOS added support for this after Reeder 3 was released.
Mixing the two readers is fine because Feedly is keeping track for them. If Feedly were to start charging, I'd probably start using NNW exclusively since they provide the option of using iCloud to sync.
This is quite what RSS needs: Killer usecases for the Z generation. And podcasts are only halfway fashionable. Instagram, OnlyFans… it seems RSS needs a payment model to be fashionable ;)
Entirely replaced by link-voting sites now, like HN.
They had some other RSS feeds than for channels and playlists?
https://github.com/kickscondor/fraidycat/wiki/RSS-Atom-Exten...
It feels like there is still a lot of room to extend these formats - part of the advantage to them is how easily they can be extended. The media enclosure extension is the reason they've been so useful for podcast subscriptions.
> Now you know how to escape the attention-draining, empathy-killing, critical-thought-suffocating siren song of the algorithms.
The only way I got my sanity back to cut out everything that's not RSS. I still use HN's algo, but everything else goes to NewsBlur. I'm trying to quit the HN algo, but can't do it yet... mainly because the content is actually interesting.
I don’t think it’s an either or thing so much as a matter of intervals. Come back from time to time to top up with new sources you missed before.
As an FYI in case you’re not funneling HN through your reader, this is pretty handy:
* Hacker News RSS https://hnrss.github.io/
You can subscribe to the front page, the firehouse, comments, search queries, and more. I really like it!
It is worth noting, for the pedantic among us, that this feed is technically Atom, not RSS. But as some other commenters have posted out in this very post, most clients seem to treat those interchangeably these days and honestly I don't really understand the difference myself.
Is JSONfeed actually used by anyone on HN?
[0] - https://0chan.vip
So yes, someone on HN is using it.
e.g. I've got an economist sub - a reputable source of news. Checked out the feed...picked one...week in review. Title on every post is week in review. Content of post is a link. No context, no summary, no single sentence...nothing.i.e. Would need to click through to find out topic on each one.
Might as well not use RSS...
Love the concept but have concluded in practice it is a non-starter. The usefulness of it depends entirely on what providers stick in the feed. And all their focus is on pushing their app/twitter/insta etc. Not hard to monetize RSS.
https://www.economist.com/feeds/print-sections/69/leaders.xm...
Business this week
Politics this week
KAL’s cartoon
Politics this week
Business this week
etc
Was more about the wider point than bitching about economist: RSS readers are in a uniquely vulnerable position. There is damn near zero incentive to do this right from providers PoV - if anything the opposite. If you're trying to consume news/info better that seems like a bad starting point to me...even with superior tools.
I've got a self-hosted freshrss install floating around on my home server...so might give it another try at some point. Not massively optimistic given the dynamics though. Everything is about dark patterns these days, not open ad free convenient hard to monetize stuff like RSS.
Run it through an RSS full article extractor:
https://github.com/AboutRSS/ALL-about-RSS#full-article-extra...
Many feeds are useless without one, but tremendously useful with one.
And if you don't like a website's RSS feed, make one yourself, formatted as you see fit:
I work on a service called Full-Text RSS[2] that used a PHP port of Readability, coupled with site-specific extraction rules[3] to identify and extract article content from each feed item. It then produces a full-text version of the given feed. The idea is you subscribe to the full-text version in whichever feed reader you use and it will give you full-text articles where you had partial content before.
[1] https://github.com/mozilla/readability
I use https://sumi.news which has both of these views, and can follow RSS, Twitter, and newsletters. Another reader with the per-feed view is https://fraidyc.at, but it’s a browser extension with no newsfeed view.
I'd like a good news feed that only has really important stories. No filler, no press release stuff.
newsblur has a 'trainer' feature where you can train it to hide certain topics or authors and highlight others. it doesn't work with every feed though so you'll have to try it out yourself or maybe there are other readers that do something similar
Iirc Samuel Mentioned they are working on something better that trains statistics/ML on examples. But at present it’s just a “highlight/stop” list.
It's not great nor complete, but is very simple and does the basic thing, it has no ads and there is no risk somebody will turn it off or push it in commercial ways.
I built it about a month ago over the weekend and haven't looked back to other popular services.
If you're a developer, making an RSS reader you like seems like a very nice side project to try out new tools, frameworks, etc... more useful than a TODO list and also very simple to build.
Mine is here: https://github.com/fallaciousreasoning/progrssive
This article mentions Atom once:
> (Note: there's something very similar to RSS called "Atom", but all modern apps work equally with both.)
Here’s what I say about it:
If you’re doing podcasts, use RSS, because almost nothing supports Atom there, because for all practical purposes Apple took over and froze the ecosystem at a certain point in time that was just before Atom became popular and fixed up the mess that was RSS.
If you’re doing any other type of feed, use Atom, because it’s technically substantially superior to RSS and supported just as well.
Where Atom and RSS are both supported, you should always prefer Atom, because RSS has the potential to mess things up because it leaves the client to guess whether certain fields are text or HTML. RSS is simply a hopelessly incorrect format, whereas Atom models content types properly. RSS is even harder to write correctly than Atom, due to things like using a weird date format.
(Look through older comments of mine for more explanation and reasons: https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu....)
Can we please stop talking about RSS and talk about feeds? A bit like we finally mostly stopped talking about SSL in favour of TLS.
RSS is worse than Atom in real, practical ways. Unless there’s a concrete reason for you to use RSS (which largely means “it’s a podcast”), you should use Atom, because it will protect your content against being mangled when you have things like less than signs and ampersands in it.
Returning back to the matter of the naming: I think there are two sound reasons to stop calling it “RSS” and start calling it “feeds”:
① Using one specific, inferior technology to describe it is factually wrong, and perpetuates the wrongness and inferiority, because people hear of RSS and so implement RSS, rather than Atom.
② When someone isn’t familiar with the stuff, “RSS” is just a meaningless name (or could remind you of other organisations that use the acronym, e.g. in India it means commonly-extremist Hindu nationalism, https://en.wikipedia.org/wiki/Rashtriya_Swayamsevak_Sangh), but “feeds” invokes a concept that is more likely to be familiar and immediately understood.
So please talk about Atom and let RSS rest in peace. Atom is technically superior if only because of the reliable standard.
- One post per week, showcasing the sponsor - Embedding a static image into every feed entry for their current sponsor - Adding a text-based „this post is sponsored by“ sentence/paragraph to every post
For me personally they’re fair tradeoffs
That's... how I prefer to read them.
Those of us in the FOSS community now need to also move away from GitHub. They are taking our code and putting it into their ML-based copilot product to allow proprietary developers to use FOSS code without having to conform to a FOSS license. This practice makes GitHub is hostile to the FOSS community. Decentralization is key to preserving our software rights.