This page lists part of the section. The page links below work without JavaScript. With JavaScript, a filter searches the whole section and pages the matches.
Podcasters can now set up value for value in about five minutes. But after hearing Australian podcasters struggle to fund a listener wallet, Dave says the only people set up to pay are those already deep into Bitcoin who know how to get through the hoops. The listener side was the real problem.
you're you're sitting there with a with a, you know, functional receiver. But there's no, there's the only people out there who are, who are set up to pay are the people who already are intimately familiar with Bitcoin.
Purely idealistic standards efforts end up like the XKCD comic: 11 standards, then 12. Dave says building on interop with Bitcoin and Lightning changes that. Success is defined by working with the system and making money sustainably, so it's not just one ideal argued against another.
if it's just pure idealism, well, then your ideal is known as on his merit may not be any better than anybody else's ideal. And you're just arguing about, you know, you're just arguing about your vision.
What Dave finds beautiful about open source: everyone can have their own politics and get angry, but once you agree on a spec, all of that becomes ancillary. In developer calls nobody even knows the others' politics. It's about the shared vision and goal.
if you can agree on a standard on a spec, it eliminates all that stuff. None of that other stuff matters. It all becomes ancillary. All that matters is the goal and the project.
Like the dome-dwellers in Asimov's The Caves of Steel who panic at open sky, the podcast industry is so used to silos that it can't think clearly about open source. It assumes there must be an ulterior motive, because it has only seen money made by capturing part of the industry, not by serving it.
everybody's so accustomed to silos and walled gardens. And that's how you make money is by capturing part of the industry not by serving it that they can't decide they don't know how to think about it.
Even if a podcaster runs a private Mastodon instance just to own their comments, Dave says integration must use open ActivityPub. Otherwise all podcast comments end up having to use Mastodon, which is opinionated in its own way. What's missing is a schematic of the interaction points between the tag, the comment server and the app.
you're gonna end up with a system where all the comments in podcasts have to use mastodon.
Asked why ActivityPub comments weren't moving, Dave says nothing like it existed, so it meant maybe six months of from-scratch code. Nobody wants to commit until the pitfalls are worked out, because no one wants to discover a dead end three months in.
nobody wants to commit to it. Until you until you have all of the pitfalls worked out ahead of time because you don't want to get three months into it
For podcasters not yet monetized, or earning only a couple hundred dollars a month, Dave suggests a free LNPay wallet to get their feet wet. Shows making significant money should run their own node, as Dave Jackson did on Voltage. It's more complex and not free, but it gives the most freedom.
if you have a larger podcast that's monetized as well monetized and you're making significant money off of it, I would probably just go ahead and roll your own node. Because that really does give you the most freedom.
People insist 'it has to be an app' to be real. Dave says part of that is Apple deliberately hobbling progressive web app technology so native apps stay the only first-class citizens on its platform, which is why web-first podcast apps struggle there.
Apple has intentionally kneecapped their progressive web app technology because they they want to box in the native apps and make them make them the only first class citizens on the on the platform.
Discussing calls to make Bitcoin transactions reversible and unmasked, Dave explains public choice theory, which analyzes political behavior as if it were a market. One tactic is the 'juicer bill': threaten an industry with legislation to generate publicity and campaign donations, with no intention of passing it.
juicer bills are a tried and true long standing thing within politics, where you threaten an industry with a certain bill, and it gets public, it gets publicity and that kind of thing. But the goal is never to pass the bill.
After Podping's Hive writer silently dropped URLs, some wanted it to report write failures back to the front end. Dave prefers each piece to be completely independent and handle its own failures. An 'okay' should just mean 'I received it', and failures downstream are fought with persistence and retries.
each piece of the puzzle needs to be completely independent in eponymous where it can handle its own failures.
The internet has desensitized us into thinking art, sounds, video and even web page snippets just spontaneously generate. Every piece was created by a human being. Industries advance the same way: people doing things, an app here, a host there, interoperating, so the industry moves forward one notch.
every bit of it was created by a human being in some shape or form or fashion. And we have been desensitized to that,
Answering a claim that apps wait for the biggest players before adopting 2.0 features, Dave says that's circular reasoning. If everyone waited for everyone else, nothing would happen. In open, decentralized processes like RSS, someone has a good idea, writes the spec, builds reference implementations, then evangelizes until others spend developer time on it.
nobody asks, nobody waits for, for another person to implement something before they implement something, or nothing ever gets done.
Looking at Spotify's remaining exclusives, Dave argues that brand safety is boring, so silos need extreme shows whose fans will follow them to a new app. Hosts were told to keep repeating that the app is free. The result: cheap audiences who only come because it's free, served bargain-basement content.
They they're going to get cheap audiences that will only come if it's free. And using bargain basement content.
Filtering the Podcast Index's episodes table by audio or video required an index on media type that Dave never added. With 70 million rows, adding one later is a big lift: test on a dev copy, and possibly stop all writes and shut the aggregators down for hours.
that is the one thing I regret about the NFA items table, which is the episodes table that I did not index properly. I did not index the the type the media type properly.
Discussing El Salvador's Bitcoin law, Dave clarifies a common misunderstanding. Legal tender doesn't just mean the government allows you to accept a currency; by force of law you must accept it as payment for goods and services.
legal tender means not that you can accept payment, but you must, by force of law accept payment in this currency.
Dave compares closed 'podcasts' to Howard Stern on Sirius XM: it isn't radio. Like radio, podcasting means anyone can broadcast and anyone can receive on any device, with no approval or corporate blessing. A show can be rebroadcast on closed platforms and still be a podcast because its DNA is open. If the open original goes away and it's exclusive to one platform, it isn't a podcast anymore.
That is what podcasting is podcasting is the ability for any creator, to broadcast to any receiver, not listener, but to any receiver.
Podcasting has no universal ID, so everyone keys on the feed URL or Apple's ID. When a show moves hosts, its URL changes and nobody knows it's the same show; at worst aggregators hold duplicates. The podcast:guid tag fixes this with a UUIDv5 hash of the feed URL at the moment it's implemented, under a shared namespace, so the ID stays the same when the show moves.
if Dave's amazing podcast at RSS calm, then moves over to fireside now, the Feed URL is different. And nobody knows who nobody knows is the same.
Dave's realization about why he cares about free speech and refuses to act as an editor: reality is more complicated than we want it to be. Viewed from one side, say the Western one, you exclude an entire view of what's happening. It's impossible for a third party to make unbiased decisions about whether someone's speech is right or wrong.
it's not hard, it's impossible to sit back as a third party and make unbiased decisions about speech. And whether those things that somebody is saying is independently right or wrong. It from only your side of the fence.
When a story like El Salvador's Bitcoin law breaks, Dave works from two assumptions. The first understanding is probably wrong, and the media cobble narratives together from the most vocal parties, often NGOs and financially interested actors. Headlines won't get you there; you have to dig deeper.
number one, that our first understanding was stories, probably wrong. And number two, the media, in general, just cobbled together narratives from the people that are the most vocal about a subject.
After years of finance conferences selling blockchain as a solution looking for a problem, Dave understands the eye-rolling. But a few use cases are legitimate: cryptocurrency, public lot-tracking ledgers, and Podping. Any host sends a ping, and the URL is written to Hive where anyone can read it.
I do think wholeheartedly that we have found a legitimate use case for blockchain technology.
Dave lists WebSub's problems. Re-subscribing to 1.6 million Anchor feeds takes a dedicated server. Many big hosts don't support it because they'd have to run a hub sending hundreds of webhooks per update. The free public Google hub is overloaded and misses a lot. And it carries only feed-change notifications, nothing like live-stream alerts.
web sub is a great is a great idea. And it does democratize some of the subscription stuff. But at scale, it completely falls down.
Before going through clips of a hosting CEO and Mike Dell misunderstanding Podping, Dave says he holds nothing against them. Misunderstandings are natural and give an opportunity to correct them. After a decade of blockchain hype, he understands why people's eyes roll at the word.
it helps to see these things be misunderstood and give an opportunity to correct them.
Reading the MORE Act, in which Congress itself describes how the drug war ruined millions of lives over roughly 80 years, Dave wonders why anyone should trust official predictions, such as Janet Yellen saying inflation is only temporary.
reading the government saying how, how wrong their own actions were. In light of what they thought at any given time over this 80 year drug war. And I'm thinking, Okay, how can we trust any of their predictions at all?
As fact-checkers retracted their Wuhan lab-leak rulings, Dave pointed out that six months earlier YouTube was instantly deplatforming people for that 'misinformation', sometimes via automated transcription flags with no explanation. Somewhere after 2012 the dialogue shifted from 'it's free speech, deal with it' to believing every word could be moderated.
Go back six months ago and if you were on YouTube, and you were spreading quote unquote, Coronavirus, misinformation, man, they would hammer you. You got d platformed. Instantly, no questions asked
Debates about content controls get muddled because four things are conflated: policing content such as hate speech, licensing content, blocking where a show can be listed, and fraud. Dave says each needs its own conversation. For example, a license tag makes sense the way open-source licenses do, while a feed trying to choose its directories is silly.
it does not help the conversation to have to have them sort of all poured into one mole and like MIT, you know, mixed around, because they're really four separate conversations to be had.
On the proposed podcast:block tag, Dave will support it but ignore it if companies use it for leverage, for example Anchor blocking everyone but Spotify and Apple. robots.txt broke the same way: sites blocked everyone except Google. Do Not Track failed because once one big player ignores it, it's useless. If only small directories respect it, it just punishes them.
do not track that header didn't work, for the same reason, is because everybody, you know, as soon as you have one big player this that ignores it, well, it's useless.
Media subscriptions churn constantly because a flat fee can't reflect what a listener is getting, and that value changes daily. You binge a show for weeks, then barely listen for months. When a busy person hits play, they've committed precious time; at that moment they'll give a lot, more than they would as a subscriber caught on a random day.
can never accurately reflect what they actually you know, what the value that they're getting from the product, because the VAT, the value of media is ever changing on a on a daily basis,
Hollywood-style production brings self-importance and award-show back-patting. Dave says slick production loses the beauty of raw, community-driven audio. The unpolished nature of podcasting, not highly produced content (which was always available), is a big reason it is what it is today.
the wrongness and the the unpolished of what podcasting is and has always been, is one of the reasons why it it is what it is today.
As AT&T unwound its Time Warner purchase, Dave said the lesson isn't just mismanagement. Something inherent in media doesn't jive with tech. Even Netflix's real hits came when it wrote a check and let the media people do what they wanted. Tech trying to manage creative work into efficiency is destined to fail.
I think it's a marriage made in hell. And any attempt to for big tech to take over media in some way, is just destined to fail.
The alternate-enclosure tag had drawn input from almost no one but Dave and Alex. Dave is fine finalizing a tag nobody thinks they need yet. Months or a year later someone will find it exists and use it. If the people who care think it's good, silence shouldn't hold it back.
I don't mind putting in a tag that maybe nobody thinks they need. Because eventually, it may be three months, six months, a year from now. Somebody says, Oh, well, okay, this tag exists, I can use this
For most people, the on-ramp to open podcasting is their hosting company, but it doesn't have to be. A feed can live at one host while being on-ramped into the ecosystem from somewhere else, such as a tool where you add your feed and manage its value block and stats.
what we need in general within the podcasting open world is, is really more on ramps.
Responding to a developer who wants to give listeners full control over value splits, Dave agrees in principle. But apps guide users, and that means guardrails, the way browsers honor content security policies. Users shouldn't be able to strip out a show's charity split. Anyone who wants no limits can write that app; it's all open.
apps guide users in a certain way. And part of guiding the user is putting some sort of guardrails on that, that you can't avoid that.
Seeing shows offer club memberships, Dave likes the idea but dislikes making the ad-free feed the selling point. Listeners liked the show as it was and have a fast-forward button. Ask them to support the show as is, rather than implying your product is better without the ads.
I guess I just don't like selling your product as if it's better without the ads. Maybe it is, but I don't, something feels wrong to me about that.
A feed can carry a dozen namespaces with overlapping tags, such as title and itunes:title, so every aggregator must decide which source of truth wins. Dave says you often default to the baseline that you know works for most shows, then adjust as problems surface.
a lot of times what you do is you just default back to a sort of baseline. You say, Okay, well, alright, let's just do title. We know title works.
Dave's 'IT guy' theory on Apple's API dropping feed URLs. Announcing you're cutting a feature others depend on brings the mob with pitchforks, which Apple doesn't want while facing antitrust scrutiny. Making it less reliable, or just not caring, gets the same result: once 15 to 20 percent of URLs go missing, everyone bails, and it can be dropped without an announcement.
there is a nice benefit you get by D stabilizing something within your own within your own ecosystem that you don't really care to maintain.
Big aggregators fall down through sheer numbers and need optimization shortcuts, and every aggregator, Dave's included, has bugs. Those bugs often live in gray areas where it's unclear how to handle a situation. Developers struggle with cases where the variables don't mesh cleanly.
programmers and developers tend to really struggle when it comes to things that are gray. Like, well, what do I do when these set of variables don't really mesh out?
Podping's design puts one DNS name in front of a group of servers, run by different parties and reached round robin. Each server must service a request from start to finish on its own. With six or ten of them, any one can run everything, so there's no single point of failure, and publishers only need to know one address.
each one of the piping servers needs to be able to fully service a request from start to finish all the way through. And that way, there's no dependence on one on one single point of failure.
Worried about Apple limiting the open ecosystem, and hearing its new app is poor, Dave says the fix is easy. Hosting companies should give new accounts a list of apps that support 2.0 features, and podcasters should tell listeners on the show which apps give the better experience.
it is so easy to fix this recommend people to use different apps on your show and on your hosting platform.
Google Podcasts gets its feeds the way the Index does: by crawling, without Apple or Spotify's direct pipeline. Dave thinks that makes Google well suited to champion open RSS podcasting, because despite its faults, its core business depends on a huge network of independent websites existing.
their core business still depends on the open web existing. It depends on a humongous network of decentralized independent websites,
Not in the audio. Guest episode; the quote comes from a 9-cue Dave run. Dave adds that he's no Google fan, but that line was labelled Unknown and isn't quoted.
As people fretted about Apple's new paid-subscription rules, Dave pointed to a collective amnesia. A month earlier everyone was discussing creators leaving YouTube over monetization problems and bringing their content in-house; now that lesson seemed forgotten.
there's this collective amnesia that happens, where everybody just forgets things that that were just precious and present in your mind, like, like a month ago.