12 ms·
Why Podcasting Still Needs RSS
- chrisrhoden 10y agoAuthor of the post here, happy to provide any clarification or answer questions!
- _kyran 10y agoHey Chris! I run a podcast hosting/publishing service (https://zencast.fm https://zencast.fm) and am curious to hear more about what other metadata you'd like to include in a feed aside from greatest hits?
- chrisrhoden 10y agoHi there! That's awesome. The short answer is, we're still figuring it out! Part of the approach we're taking is about letting us test things with real users against live feeds without bugging people every time we change our minds (about the content of the extensions or about the format of them), and we're still at the very early phases. The somewhat longer answer is, we know we would like to get better promotional / onboarding data from the feeds - including things like a 140 character description for the feed that references similar content from other media (like tv), or the best gateway episodes to present new users with. Ordering is a big one, for shows that are e.g. serialized, or even for shows with short, strictly ordered series embedded (think 2 part specials in a daily news feed). The folks at https://syndicated.media https://syndicated.media (including me) have been keeping track of ideas on Github (https://github.com/syndicated-media/sn-spec/issues https://github.com/syndicated-media/sn-spec/issues), and you would be most welcome in the slack group for syndicated media, where things are just starting to pick up. Shoot me an email if you would like an invite!
- tracker1 10y agoFor that matter, I think RSS is still needed for blog aggregation and reading... Since google killed reader and iGoogle, I find myself only reading a handful of sites regularly.
- chrisrhoden 10y agoI agree! RSS is kind of a magical thing, that's only really happened a few times - the widespread implementation of a standard that provides a better user experience at the cost of somewhat less control for content owners. My headline is in response to certain voices in the industry which seem to be rooting for the end of RSS in favor of proprietary APIs - essentially, the YouTube-ization of podcasts. There's lots to unpack there, and to a certain extent it seems likely that more things will move away from open standards, but I'd prefer to hold the line to the best of our ability.
- jagermo 10y agoAgreed. RSS is an amazing technology, but I miss a real good feedreader. I loved FeedDemon, but since its development stopped it got so sluggish when handling my RSS feeds.
- Symbiote 10y agoI use Mozilla Thunderbird. It doesn't synchronise between devices (as far as I know), but for work-related feeds I use my work computer 99% of the time anyway.
- hackuser 10y agoI second Thunderbird. It's lightning fast for browsing through large numbers of feed items.
- SirFatty 10y agoI've been using netvibes for 9 years without issue...
- gman83 10y agoI use Inoreader, I quite like it: https://www.inoreader.com/ https://www.inoreader.com/
- bandrami 10y agoEmacs has several, and since I tend to live in Emacs I never really noticed the whole Reader saga. I would definitely be sad if I couldn't get my RSS feeds next to my NNTP and my IMAP in gnus.
- PretzelFisch 10y agoWhy would you need a GUID? When you have a URL?
- miiiiiike 10y agoBecause URLs can change and if they do you'd want to update the item without adding a new one.
- _kyran 10y agoExactly, currently if a podcast feed changes URL they have to hope that the client on the other end notices a 301 redirect and caches/looks at the new feed url in the future.
- chrisrhoden 10y agoSo, in the RSS spec, the <guid> tag is used to identify individual posts. Without some sort of identifier, it's hard to know if a user has e.g. already heard something in a feed (because new things can be added and old things can be removed). GUIDs, as defined in the RSS spec, are also not required to be globally unique - they're only supposed to be unique to the feed that includes them. This, combined with the fact that a feed that doesn't include them at all is still completely valid, is usually a pretty big stumbling block for developers looking to consume RSS. This is the biggest issue that got talked about during the Atom vs RSS discussion of a decade ago. And if you're referring to the UUID I'm automatically adding to every feed, it's so that there is an unchanging canonical way to refer to a feed that follows it across URL changes etc. The sibling comments get it perfectly.
- PretzelFisch 10y agoI see. I had to review the spec, I was under the impression that item's source element and URL required and used to locate the item. But I see that is not the case.
- Animats 10y agoMy experience with processing RSS feeds is that none of the uniqueness indicators in feeds really work. The only thing that works is hashing the content and discarding duplicates. Etags are especially useless. There are sites with a load balancer in front of multiple servers, each with its own etag. As I've pointed out before, RSS feeds remain in widespread use for hard news. Almost all the hard news sites, including AP, NYT, Reuters, VOA, the BBC, etc. have useful RSS feeds. There are RSS feeds from NASA, feeds which tell you what Congress is doing right now, National Weather Service alerts, missing kid alerts, and California fire incidents. There's even a Donald Trump RSS feed. If it matters, it's on RSS. Social has gone in a different direction, but that's just the social media industry.
- firasd 10y agoI'm making a podcast discovery app (info in profile). My plan is, if I (hopefully!) get enough critical mass to get podcast publishers' attention, to switch over to preferring Atom feeds. Just because Apple made a decision in 2005 doesn't mean we have to be stuck with it. I hope, given the combined efforts of people in this space, we can break podcasts out of the grip of the iTunes store (e.g. Why do podcasts need reviews? Have you ever seen a Youtube video where someone urged you to leave a review? It's because reviews catch the attention of iTunes store curators, but Youtube does more automated personalization.)
- _kyran 10y agoReviews are iTunes' form of social proof and help enable some amount of discoverability. Youtube's social proof equivalent is likes, comments & subscriptions. Of which most youtubers request at the end of a video in the same way.
- firasd 10y agoFair enough. It's true that whatever metric a platform uses for either manual or automated curation, publishers will know about it and optimize for it.
- dublinben 10y agoI really wonder why Apple and iTunes still have so much influence in the podcast space. They have less than 40% market share in the US, their strongest market. In Europe, Android has greater than 75% market share. Is the podcasting audience so different than the mobile market in general? If not, why do so many publishers act as if iTunes is the only place that people find podcasts these days?
- chrisrhoden 10y ago> Is the podcasting audience so different than the mobile market in general? They seem to be, for now. Not sure whether that's because it's just easier on iOS or habit. In any event, we're pretty sure that won't last forever.
- kgwxd 10y agoEverything still needs RSS! There is no better way of getting notifications. I don't have to create an account, I don't have to give out my email and I don't have to re-search through everything I've already read or decided not to read. I canceled my YouTube account a while ago, realized I sort of missed a few channels and found myself just trying to remember to check their pages every now and then. Then I realized YouTube channels actually have hidden RSS feeds. The experience is better than what you get with a YouTube account. The number you see next to the channel actually indicates the number of videos uploaded to that channel that you haven't seen yet. I still have no clue what the numbers next to subscribed channels meant when I had an account.
- robobro 10y agoI would argue that ATOM is better: http://www.intertwingly.net/wiki/pie/Rss20AndAtom10Compared#samples http://www.intertwingly.net/wiki/pie/Rss20AndAtom10Compared#... Sure, they're both automatically generated / automatically read, but ATOM seems like an improvement syntactically ... http://nullprogram.com/blog/2013/09/23/ http://nullprogram.com/blog/2013/09/23/
- onli 10y agoI worked on both Atom and RSS Feeds for blogging engines. Frankly, Atom had for me no real advantage. RSS is easy to generate, and it is supported properly by so many clients that Atom can be comfortably ignored. Stuff like the guid not being a guid work in practice just fine.
- redwall_hp 10y agoTo be honest, they're basically interchangeable. Most parsing libraries will handle both, and it's entirely down to preference.
- narrowrail 10y agoFrom my [not extensive] research, ATOM open source libraries have better support for authenticated feeds.
- smagch 10y agoI'm using feedly.com for reading tech blogs. For long articles, I use getpocket.com to read it later instead of Feedly's 'Read it later' feature. Both service give me the hints of popularity of the article. Pocket adds 'Best Of' label for popular articles. Feedly adds a number of 'Read it later'. I'm currently using Feedly for subscribing podcasts. But it's not handy to use. If RadioPublic give me the same feature that feedly and pocket provide for the sake of podcast listening, I'll definitely use the service.
- monksy 10y agoThis seems like an easy problem to resolve. Revive the standards body for this protocol. Create a new group that will define the spec and schema, convince others to use this and you're good to go. Unless you need to remove elements, it should be pretty simple to expand on what's already there. To me it doesn't make any sense to bemoan that it is in XML.. so what. It includes a schema and can be validated. That's a good thing. Don't throw json into this.
- initself 10y agoI don't think he bemoans it. He describes how Apple themselves extended the protocol through XML itself towards the end of the article. "Remember that the RSS spec is frozen, meaning that no improvements can be made to it, full stop. That said, RSS documents are written in XML, and XML has a feature called namespaces which can be leveraged to include data from multiple specifications in a single document. This allows one to create new specifications that could be used, not instead of, but in concert with RSS. iTunes did this when they initially added support for podcasts back in 2005, but little movement has happened in the intervening decade."
- firasd 10y agoThat's very naive given the specification in question. Atom was an effort to do this. The process created years of strife. And after all this drama, Facebook took over and RSS/Atom/et al became irrelevant.
- chrisrhoden 10y agoI have to be honest, I find it a little frustrating that you would not spend the estimated 6 minutes it would take you to read a post I spent hours writing before you put me down in a comment.
- gkya 10y agoRSS/Atom are second to e-mail in being called obsolete every other day with a million contender technologies (Flipboard, Google Currents, whatever else came out this morning...) yet still surviving and indeed being as relevant as what you're gonna have for the dinner today. So logical it's natural.
- Dowwie 10y agoOf course podcasting still needs RSS. I don't use itunes, stitcher, nor any app for a specific platform (soundcloud).
- erikrothoff 10y agoI'm one of the thousand who run an RSS service (feeder.co), and we're still growing! From speaking with our users, it's evident that the uses for RSS and similar open standards are endless. It can make you feel really small at times when the product you are creating is standing on the shoulder of not-so-benevolent giants. For example, Craigslist apparently hates when somebody actually uses their feeds. They IP ban anyone doing a pretty regular amount of requests per hour, and are completely impossible to get ahold of. (have you seen their forum?) When our spiders are blocked by Craigslist, our users suffer. And then they come to us asking why feeder doesn't work...
- lkrubner 10y agoThose of you who want the history of RSS (and why Atom seemed necessary) might be interested by a long article I wrote in 2006: RSS has been damaged by in-fighting among those who advocate for it http://www.smashcompany.com/technology/rss-has-been-damaged-by-in-fighting-among-those-who-advocate-for-it http://www.smashcompany.com/technology/rss-has-been-damaged-...
- thomersch_ 10y agoReally staggered to see work on Podcasting and not even mentioning or considering Podlove http://podlove.org/ http://podlove.org/ They are already working for several years on improvements and standards for publishing and consuming podcasts. E.g. Podlove defined how to deliver chapter marks inside a feed and implemented it in their Wordpress based Podcast Publishing Software.
- chrisrhoden 10y agoI am familiar with podlove, and the work on chapters in particular are interesting, but they face the exact problems I talked about in the post - they are mostly being designed in a vacuum. Apps are slow to add support for nonexistent data, and producers will never add data that will not be used (it's hard enough to correctly fill in the stuff that will be used!)
- cJ0th 10y agoOn a related note: I find it sad that Mozilla hide the button for rss subscriptions in recent Firefox versions. You can restore it when you customize your toolbar (i.e. leftclick in the tab area -> customize ...) but that prevents people who don't use rss yet to "discover" it so to speak.
- type0 10y agoI really don't understand this move by Mozilla, why did they hide it and put Pocket instead!?
- benzimmer 10y agoHey Chris, thank you for your article and the ideas in it! I'm co-founder of Podigee [www.podigee.com], a podcast hosting service. We're active contributors to the Podlove Project [www.podlove.org], which seems to have kind of the same "mission" as syndicated.media. We also try to push existing standards and try to establish new ones, some of which you can find here: http://podlove.org/specifications/ http://podlove.org/specifications/ Additionally the initiative creates open-source software like the Podlove Publisher [http://podlove.org/podlove-podcast-publisher/ http://podlove.org/podlove-podcast-publisher/] and Subscribe Button [http://podlove.org/podlove-subscribe-button/ http://podlove.org/podlove-subscribe-button/], that make podcaster's lives a little easier. Perhaps we can setup a Skype session with some contributors of both syndicated.media and Podlove. I bet that would be an interesting conversation :) Cheers, Ben
- chrisrhoden 10y agoHi Ben, that would be great! I am indeed familiar with Podlove and I think there is a little bit of overlap. I don't personally control syndicated.media but can forward an email to the person who does - can you reach out to me on a private channel? My email is in the post.
- WhitneyLand 10y agoNone of the good things you mention seem to require new investment in XML, I'm not sure what benefits you see in it. Why not just maintain existing functionality in RSS, and at the same time design a new format than is modern, simple, and extensible? The modern format could be linked to or embedded within the legacy format as long as needed. Newer clients could continue reading the legacy format while adding support for the modern format. Older clients would continue to work indefinitely. This would allow a gentle transition and allow new investment in XML code to cease. I'm not religiously against XML, but there are many practical benefits to not using it when possible.
- chrisrhoden 10y agoIt is an interesting thought. I think it's hard for me to imagine the benefits of moving on from XML overcoming the pain of supporting both formats for literally everyone involved (producers and apps), and RSS is, despite its shortcomings, quite mature. If you are proposing hanging additional data off of RSS with a link but continuing to fetch the RSS for basic episode data, I think it is self evident why that is fraught. And if you are suggesting having an alternative representation, I guess I worry about reliving the pain of the early days, with lots of format fighting and incomplete / incorrect implementations. How do you see this working?
- WhitneyLand 10y agoChris, it's only intuition, i'd have to research RSS more to respond precisely. However I'm impressed by the fact that you are thoughtfully considering the comments here and taking the time to reply to them. That's always a sign of a good architect, and I look forward to seeing the progress your team makes.
- iand675 10y agoLast time I looked, RSS and Atom didn’t have pagination support and many RSS feeds only show the most recent N items. I'd say solving that issue to make feeds more browseable would be a significant improvement for the way that I want to use them.