I'll summarize my thoughts mentioned in the Github issue[1]:
> I think this comes down to who W3C considers the authority for RSS2 today.
> While from a technical standpoint [both Dave's and RSSBoard's spec] are identical, they do have some differences .... and if rssboard comes up with a new donations attribute for <item> and the Harvard site doesn't list it, which copy should W3C run their validator against?
The drama around it is awkward, but I think Winer has a point about the success of RSS being at least partially attributable to it not changing. The immediate analogy that springs to mind is email, also archaic, also widely adopted, you don't have to fear that there will be breaking changes or some company will decide it's going away tomorrow. You just use it and it works. (RSS actually had that apocalyptic moment when Google Reader shut down, did it die? Nope.)
The net result is that RSS is low maintenance, the implementations out there are stable and have been around for years, very little work needs to be done on code which deals with RSS, millions of websites continue using it and no one has to lose any sleep over it.
Not that I wouldn't like to see any improvements at all. It would be nice if these guys could set aside their differences and say look, we'll do an update of some kind and increment the major version by 1, say, every 10-15 years. Yes that's practically glacial in Internet time, maybe that's actually good.
Yep. Use RSS if you’re dealing with podcasting, because most of those platforms don’t support Atom (and it’s basically all Apple’s fault—they took over and froze the space at a certain point in time just before Atom became popular and fixed up the mess that was RSS), but use Atom for all other feeds, because it’s just hands-down better than RSS, in ways that occasionally even matter (as in, will cause some software to mangle your content, most commonly your titles if you use stuff that could look like HTML).
AFAICT Google Podcasts is doing the same as Apple and only using RSS 2.0.
Nothing is stopping people from providing both RSS 2.0 and Atom, and given that most feeds are probably auto-generated, probably not hard to do so, but it is disappointing.
Apple and Google did this because (1) alternative feed formats like Atom or JSON Feed weren't being used, and (2) a few years ago, the podcasting industry and standards bodies chose to require RSS for extensions/enhancements.
Atom wasn’t being used in podcasting because it wasn’t supported by key podcasting platforms! I know from my research of multiple entities that tried to use Atom (since it’s obviously and unambiguously superior), particularly because of some things that claimed Atom was supported, but gave up because it turned out that even most of the platform things that claimed support didn’t actually support it, either at all or fully. As for things like JSON Feed, that’s never been a serious contender.
This Apple Podcasts update saying “we no longer support Atom” genuinely surprises me, because when I looked a couple of years ago, as far as I could tell as an outsider with no access to any of their software, they didn’t support Atom, and never had.
I’m not certain what you’re referring to as “the podcasting industry and standards bodies”, since I haven’t been paying attention to the space (a couple of years back it seemed a bleak wilderness). If it’s things like https://podstandards.org/, well, “Apple has created some thoughtful vendor-specific extensions to the RSS standard”… bleh. I cannot understand how anyone could have anything but scorn for the mess that is podcasting RSS. It’s shamefully bad, and the paragraphs there are rather baffling to me in their untempered praise for Apple and condemnation of (unspecified) others. As for their standard, they’re explicitly standards-washing certain Apple interests (though I will admit that most of the iTunes namespace stuff is generally useful), adding stuff that has no place in a “required standard” (funding!? And what happened to the <link rel> mechanism for some of that stuff, which would obviously be appropriate? There really looks to be a lot of poorly-thought out stuff in this).
> Atom wasn’t being used in podcasting because it wasn’t supported by key podcasting platforms!
But Apple did for many years, no? If Atom was obviously and unambiguously better for the use case, Atom would've been successful for that use case.
> I cannot understand how anyone could have anything but scorn for the mess that is podcasting RSS.
I built a reasonably popular podcast feed validator and don't understand which part is a mess. If you wouldn't mind sharing, I know the Podcast Standards Project folks are considering what their next project should be.
> …the paragraphs there are rather baffling to me in their untempered praise for Apple and condemnation of (unspecified) others.
Spotify's embrace/extend/extinguish approach successfully killed the original "open medium" meaning of the word "podcasting". Whatever your other critiques of Apple, they worked within the RSS standard.
> I think about half of the RSS feeds I subscribe to are Atom-based.
Meaning, they have an `atom` namespace declaration? Because that's very common in RSS feeds, and is required for Podcast Standards Project-compliant feeds. Apple supports RSS feeds that use the `atom` namespace.
Podcasting uses RSS because alternatives like Atom and JSON Feed don't solve a problem in this space. New podcasting standards (Podcast Standards Project, Podcasting 2.0) are supported by extending RSS, which is now explicitly required. (Apple is just a consumer of RSS — they don't use RSS between their back-end and their podcast app.)
RSS is an awful mess. Atom was invented in significant part to fix the problems of RSS (things like the content encoding disaster), and it succeeded. Except that, while everyone else accepted it (most things use Atom rather than RSS, admittedly with a few rather notable and massive holdouts), the podcasting industry ignored it.
The way RSS and especially podcasting RSS just pulls in bits and pieces from everywhere haphazardly, regularly ending up expressing the same thing three or four times, is an indictment against it. And if making <atom:link> mandatory doesn’t make you boggle (pulling in one element from the competing and superior standard, in what is at the very least a very mild abuse of the XML namespace—not that RSS cares about XML conventions on namespaces), nothing will.
I suggest publishing with Atom. However if you are consuming feeds you will probably want to support RSS 2.0 as well as various buggy implementations and some earlier RSS versions.
It's still not the easiest to come up with an attack of consequence against what is basically a static site that is worth an attacker's time. You can draw out scenarios, certainly, but most of them are some combination of "doesn't matter" and "if that's the goal, there's a better and easier way to do it".
I ran on that philosophy for a long time too, but Let's Encrypt tipped the balance for me. When TLS certs cost real money it was easy to decide that the super-marginal security benefits for my minimal readers weren't worth hundreds of my dollars a year. Now it's more on the order of "incidental noise in what it took to set the website up anyhow", so I go for it.
Confidence that Verizon (my internet service provider) would not know what article I was reading (although they would still know what servers I was connecting to) and would not be selling that information to anyone.
The entire Github thread felt unpleasant to read due to (what I perceived as) the passive aggressive responses. In a way it solidifies my choice to go with ATOM for my org rather than RSS, there's too much confusion for end-implementers in this space.
I've recently converted to it and can't believe I didn't sooner. It's a very nice way to get news from different sources. I started to use it after getting off reddit, and now have it to receive the latest news from Hackernews, pl discourses, lemmy communities and more.
Have you ever used it? It used to be a really nice way to consume content... instead of having to check every blog/webcomic/whatever individually, you could have a folder of RSS feeds in your browser and sweep over them to see at a glance if anything new had been posted.
Or you could have a dedicated RSS reader that would do stuff for you such as archiving content for offline use, search indexing, etc.
I am still deeply annoyed that RSS has fallen out of popular use.
> I think this comes down to who W3C considers the authority for RSS2 today.
> While from a technical standpoint [both Dave's and RSSBoard's spec] are identical, they do have some differences .... and if rssboard comes up with a new donations attribute for <item> and the Harvard site doesn't list it, which copy should W3C run their validator against?
[1] https://github.com/w3c/feedvalidator/issues/106