4 ms·

Hot take: we should revive ATOM instead. RSS and ATOM are both nearly always machine generated and machine read so this may seem trivial to argue, but for thos

Hot take: we should revive ATOM instead. RSS and ATOM are both nearly always machine generated and machine read so this may seem trivial to argue, but for those who are interesting in seeing the difference, this is a good article http://www.intertwingly.net/wiki/pie/Rss20AndAtom10Compared http://www.intertwingly.net/wiki/pie/Rss20AndAtom10Compared -------- When I wrote a forum software a while back, it was trivial to not only add a feed for new threads but also for new comments (per thread and board-wide). Made it very easy to monitor updates on all my devices. Most wikis have RSS/ATOM feeds for recent changes which is a great power feature for admins. Mastodon is atom-based, I believe. Early twitter supported RSS / ATOM feeds which arguably made it easier to follow hashtags or news accounts. Another benefit of RSS/ATOM is that you can plug them into chatbots or use them to mirror content. I would love to see syndication come back to the spotlight!

avhon1 9y ago
That comparison article is great! Is there anything that also compares JSON Feed too? https://jsonfeed.org/ https://jsonfeed.org/ As far as Twitter goes, I use TwitRSS to generate RSS feeds from of a couple of users. https://twitrss.me/ https://twitrss.me/
pasbesoin 9y ago
That server's not responding, at the moment. https://web.archive.org/web/20170914094626/http://www.intertwingly.net:80/wiki/pie/Rss20AndAtom10Compared https://web.archive.org/web/20170914094626/http://www.intert...
dannysu 9y ago
I wrote tweets2rss[1] and use it to turn my private lists into RSS feeds. For example, I have a feed in my RSS reader for interesting people (Bill Gates, Elon Musk, etc) and a list for interesting companies. [1]: https://github.com/dannysu/tweets2rss
- mxuribe 9y ago
  Hey, this seems neat; I think I'll check it out. Thanks for contributing this!
grzm 9y ago
intertwingly.net! Now that brings back memories! Sam Ruby, Dave Winer, Jeffrey Zeldman, Eric Meyer, Tim Bray, Mark Pilgrim, Joe Clark. XML, DTDs, XML Schema, Relax NG, XSL, XSLT, FOP, CSS and XHTML! WaSP, micro formats (and XML-RPC, SOAP, and WSDL (pre-WS-Death Star). Oh, heady days! Apologies for the pedantry, but a nit: It's Atom, not ATOM. It's not an acronym like RSS.
- ubernostrum 9y ago
  Once upon a time in the days of my youth, Sam linked to something I wrote, and seemed to approve or at least not completely think I was wrong. Made my entire month. I miss those crazy idealistic days.
sedachv 9y ago
Very much agreed. Chris Wellons has a good description of why Atom supersedes RSS: http://nullprogram.com/blog/2013/09/23/ http://nullprogram.com/blog/2013/09/23/ Adding Atom support to anything is trivial even without using a library, and there is a good schema for validating feeds: http://cweiske.de/tagebuch/atom-validation.htm http://cweiske.de/tagebuch/atom-validation.htm
ubernostrum 9y ago
This brings back memories of the syndication-format wars. And not good ones, either. RSS 0.9x versus RSS 1.0 versus RSS 2.0 versus Atom versus CDF versus whatever else people came up with, and which formats are relatively self-contained versus which ones had to go borrow elements from other XML namespaces and whether you should "helpfully" supplement your feed (or even your XHTML!) with stuff pulled from Dublin Core, which formats have associated publishing protocols that actually work, the mess of trying to fit arbitrary content formats into feeds and bumping into XML arcana along the way and encouraging people to put autodiscovery links in their sites and... Thankfully, for many years now I've published only an Atom feed, and it works, and that's that. And for reading I use Reeder, with a paid service that tracks my subscriptions and read/unread status, and it seems to just work. That wasn't always the case.

avhon1 9y ago

That comparison article is great! Is there anything that also compares JSON Feed too? https://jsonfeed.org/ https://jsonfeed.org/ As far as Twitter goes, I use TwitRSS to generate RSS feeds from of a couple of users. https://twitrss.me/ https://twitrss.me/

pasbesoin 9y ago

That server's not responding, at the moment. https://web.archive.org/web/20170914094626/http://www.intertwingly.net:80/wiki/pie/Rss20AndAtom10Compared https://web.archive.org/web/20170914094626/http://www.intert...

dannysu 9y ago

I wrote tweets2rss[1] and use it to turn my private lists into RSS feeds. For example, I have a feed in my RSS reader for interesting people (Bill Gates, Elon Musk, etc) and a list for interesting companies. [1]: https://github.com/dannysu/tweets2rss

mxuribe 9y ago
Hey, this seems neat; I think I'll check it out. Thanks for contributing this!

grzm 9y ago

intertwingly.net! Now that brings back memories! Sam Ruby, Dave Winer, Jeffrey Zeldman, Eric Meyer, Tim Bray, Mark Pilgrim, Joe Clark. XML, DTDs, XML Schema, Relax NG, XSL, XSLT, FOP, CSS and XHTML! WaSP, micro formats (and XML-RPC, SOAP, and WSDL (pre-WS-Death Star). Oh, heady days! Apologies for the pedantry, but a nit: It's Atom, not ATOM. It's not an acronym like RSS.

ubernostrum 9y ago
Once upon a time in the days of my youth, Sam linked to something I wrote, and seemed to approve or at least not completely think I was wrong. Made my entire month. I miss those crazy idealistic days.

sedachv 9y ago

Very much agreed. Chris Wellons has a good description of why Atom supersedes RSS: http://nullprogram.com/blog/2013/09/23/ http://nullprogram.com/blog/2013/09/23/ Adding Atom support to anything is trivial even without using a library, and there is a good schema for validating feeds: http://cweiske.de/tagebuch/atom-validation.htm http://cweiske.de/tagebuch/atom-validation.htm

ubernostrum 9y ago

This brings back memories of the syndication-format wars. And not good ones, either. RSS 0.9x versus RSS 1.0 versus RSS 2.0 versus Atom versus CDF versus whatever else people came up with, and which formats are relatively self-contained versus which ones had to go borrow elements from other XML namespaces and whether you should "helpfully" supplement your feed (or even your XHTML!) with stuff pulled from Dublin Core, which formats have associated publishing protocols that actually work, the mess of trying to fit arbitrary content formats into feeds and bumping into XML arcana along the way and encouraging people to put autodiscovery links in their sites and... Thankfully, for many years now I've published only an Atom feed, and it works, and that's that. And for reading I use Reeder, with a paid service that tracks my subscriptions and read/unread status, and it seems to just work. That wasn't always the case.