In some parallel universe, the RSS is tightly integrated with browser and main way to "subscribe" to site's updates

In that parallel universe web sites stop publishing RSS because they are overwhelmed with the polling traffic. The world really does need something like ActivityPub as much as Dave Winer denies it.

RSS has two polling speeds: too fast and too slow and it can even be both at the same time.

There are rss aggregators that poll every feed occasionally, then combine them into a single feed for each person to consume.

Nostr works on a similar basis but you push to the aggregator instead of them pulling.

Those Planets are great, I wish we had more of them!

It costs me 10 cents/(month*feed) for superfeedr to ingest an RSS feed; it fills up an SQS queue with content items which is a model I find easy because I can drain that queue at my leisure.

It would be very cost effective for me to add a Planet and have my recommender system discover articles I like from it, that is, I'm happy to ingest 1000s of items a day to my reader.

CDNs exist, as well as http cache headers, though.

Is WebSub not a reasonable solution to this?

Yes. Superfeedr makes RSS feeds look more like WebSub feeds and it's a great product. 10 cents/(month*feed) is a bargain if you want to subscribe to 100 high volume feeds but prohibitive for the model of "I want to subscribe to 10,000 indy blogs"

Firefox and Safari used to have RSS built in, but removed.

In some parallel universe, they may still have it built in.

Wasn't that parallel universe our 15 years ago?

Browsers were bad interfaces for RSS feeds anyways.

What was good was when they had discovery. The RSS button would show up in the menu bar and you could click it and add to your RSS reader.

I understand your database of feeds is curated but you might be interested in another database listing feeds: Wikidata. See https://w.wiki/MpjN for a simple example of twenty (random) feeds (obviously, you can add filters for e.g. a country, language, etc.). (There are currently 18598 feeds there in total.)

Something I read here on HN some time ago and that can be of your interest: you can get specific Atom feeds from Youtube channels (only long format videos, only shorts, only live streams, only popular, etc)

>https://stackoverflow.com/questions/71192605/how-do-i-get-yo...

I have to admit my interest rapidly waned when I realized I'd have to inspect a SQL database to look at this. Is it so enormous that a regular text file won't work?

There’s non-corpo reasons you might not want to expose all the content into an RSS feed. https://ciechanow.ski/ can only really be appreciated from the browser. In the salad days of blogging, it was as much about what kind of fun shit you could do with HTML or CSS as it was about writing and distributing your thoughts. The Yamauchi family office website could have just been a templates squarespace page, but they opted to make it look like this instead: https://www.y-n10.com/

This is exactly why I built https://apps.apple.com/us/app/wire-rss-news-reader/id1438331...

Why is it that I discover the first decent baby feeding tracker right as my last baby has aged out of needing one? SMH

Agree to disagree.

I think many platform did support RSS, or API, but at some time it was dropped. It is not that hard to provide RSS. It is not like it has to be implemented from scratch. It is just dropped by platforms. There must be reasons for dropping. One may argue it is not worth it, but even to drop functionality there needs to be a decision from management. The management always think about money. Is RSS or API monetized correctly? Free? No? Then we drop it, because data from algorithms serving user contents can be easily monetized. Just follow the money, the oldest truth.

Not just no money in, but actual money out. RSS is a hole in user funnel inflow to platform.

Also, scraping was a problem for publishers: if your site was popular, copy-cats would show up republishing your work with minimal credit.

That said, this seems quaint in the modern era where trillion-dollar tech companies are doing that to publishers now, too.

No. That wasn’t a concern with RSS for publishers. Most publishers have RSS still or feeds. Scraping index pages continues to be trivial.

Yes, it was a concern. You can say that some people made their peace with it but if you’d been there at the time, it was absolutely mentioned.

Both ideas can be true. It’s not on their radar because despite their popularity in consumer space they can’t find a business purpose that aligns with their self interests that require such user information. If I’m running a free podcast, in contrast, I might be happy anyone’s even bothering to visit and listen to what I have to say compared to who they are and whether I can assign a monetary value to their attention because spending money on something without clear, intentional, measurable ROI is anathema to our predominant modus operandi in business

Same difference really.

Though some seem to go out of their way to make RSS impossible.