“How is NYT Audio different? Great question! Most audio apps are containers for what users put into them, shaped by RSS, w/ a layer of algos on top. We believe in the combo of great audio, editorial judgement & a thoughtfully designed product will help you get listening ASAP.”

twitter.com/arainert/status/14

AWS DynamoDB down, and other services.

Time for my podcast backend to take a break, I guess.

twitter.com/kennwhite/status/1

Anything wonky with Hive today @brianoflondon ?

Starting in the last hour or so, my block watcher is reporting lots of blocks where the head (reversible) operation count is far larger than the irreversible operation count (which is zero).

This has been running for months, and never seen a discrepancy to this degree:

e.g. blocks
59778919
59778889
59778859
59778818
59778785
59778061

Just updated my podcast analytics service ranking by episode share post with November's episode data.

FeedBurner dropped quite a bit (relatively) from October to November, unclear why (it was low but steady last month)

twitter.com/johnspurlock/statu

Even after excluding the Graffiti Urban Radio duplicate feeds, they still make the new November list (much further down), thanks to their daily radio content.

Some other minor episode share changes last month mentioned in the tweet, for example: Captivate enters the top 10

twitter.com/johnspurlock/statu

@dave The problem appears to be that their CMS does not return a 404 response on bad ids, you can put whatever number you want in the url and you get that default feed!

This would be kryptonite to a simple crawler trying to find new feeds by incrementing ids.

You should probably nuke all of those, and put some sort of check/halt on that domain, or a workaround that identifies duplicate content.

Show thread

Data problem bat signal @dave

Compiling November's podcast host ranking by episodes and a new domain www.urban-radio.com appeared this month, and jumped to number 5 overall 🤨

Just downloaded the latest sqlite db, and there are about 5000 urban-radio.com feeds, most of them have duplicate content.

e.g.
podcastindex.org/podcast/44945
podcastindex.org/podcast/45009

Sql to id the dups:

select * from podcasts where url like '%urban-radio%' and title = 'Les podcasts de Graffiti Urban Radio' order by url asc

Looks like rss.com is now using hub.livewire.io for instant podping -> WebSub support. Cool!

media.rss.com/comfycouchchats/

Btw @brianoflondon, you actually don't technically have podcast namespace tags in the 3speak feeds.

The xmlns:podcast="podcastindex.org/namespace/1.0" at the top of the feed should not have the trailing slash at the end the value. See [1] for the two valid values.

Xml namespaces are compared like strings, even though they are often also urls.

[1] github.com/Podcastindex-org/po

Whoa, first time one of my projects made it to hacker news. Absolute bottom of page 2, but I'll take it! 😎

Denoflare isn't strictly podcast related, except I'm building more of my podcast stuff with it these days.

news.ycombinator.com/item?id=2

Started working on a new podcast feed validator - no rules implemented yet, just parsing..

But it gives me a way to test reading the ActivityPub comments from the podcast:socialInteract tag. The mastodon AP implementation seems to work well.

Here is the pc20 feed with the @dave post node in socialInteract

validator.livewire.io/?input=h

@dave @mitch @martin @benjaminbellamy @agates @StevenB @steven @brianoflondon Just started an ActivityPub exploration doc targeting client app dev scenarios for my own sanity, and to see if we have enough info bootstrapped from the feed to read/write comments from 3rd party apps.

Anyone can edit this, so feel free to add anything useful you find as well

docs.google.com/document/d/1eJ

Did redcircle always include <podcast:guid> in their feeds?

If not, they are now apparently...

... but using v4 uuids, not v5

Seen 5 in the last 15 minutes, examples:
feeds.redcircle.com/d7d5c8b0-b
theticketfm.com/category/tom-a

New peak in new podcast episode rate going on right now, Mondays are typically pretty high, but this is the highest I've seen so far

@dave, any updates to the PI aggregator recently?

(blue = new episodes, red = portion of new episodes that have at least one interesting podcasting 2.0 tag)

One last technical analysis of podcast new episode share, this time: which CDNs were used most often by podcast episode media in September.

Almost half of the podcast world (by new episode share), perhaps unsurprisingly, are on AWS (Cloudfront, S3, or EC2)

Also interesting to see which podcast hosting companies use which CDNs

Thanks to @jamescridland for maintaining those opawg host patterns (also you might have another pull request in your inbox)

twitter.com/johnspurlock/statu

Show older
PodcastIndex Social

Intended for all stake holders of podcasting who are interested in improving the eco system