search

scripting / Scripting-News

I'm starting to use GitHub for work on my blog. Why not? It's got good communication and collaboration tools. Why not hook it up to a blog?
115 stars 10 forks source link

An upgrade to the open RSS market #113

Open scripting opened 5 years ago

scripting commented 5 years ago

A proposal to developers of feed readers and podcatchers, to create a more powerful open ecosystem.

Here's a screen shot of the post.

image

If you have questions or comments this is the place to post them. :rocket:

OleEichhorn commented 5 years ago

If there was a one-click-link way to subscribe to feeds, that would be awesome. IMHO this more than anything has kept RSS from taking off. Auto-discovery was/is good but not universal and URL-copy-and-paste was always crappy.

scripting commented 5 years ago

@OleEichhorn -- this is the first step toward one-click subscription. Let's see if we can get any feed reader people to work with me on this. ;-)

marsanyi commented 5 years ago

Reminds me of Mog, an old music aggregation system. It provided a tool to create a playlist from pages containing links to, eg, .mp3 files (intended for freely-released soundfiles), and a way to follow another user's playlists. This last feature meant that, once you found someone who had similar taste to yours and spent hours on the net curating playlists, you could find new music just by asking what they'd been listening to. Much better way to navigate the space than the "this is like what you listened to before", or "this is what got released this week".

rosskarchner commented 5 years ago

I like this a lot, and see how this improves life once I've subscribed to your OPML in my compatible feed reader-- but establishing that subscription still seems to require the same-old copying-and-pasting-urls-to-XML-files, right? Or am I just missing something?

scripting commented 5 years ago

One step at a time. If that becomes burdensome, there are ways to make it easier, the trick is to get people to work together for the good of the market. It's rarely happened in this space. So the victory would be taking the first step.

On Tue, Mar 26, 2019 at 10:11 AM Ross M Karchner notifications@github.com wrote:

I like this a lot, and see how this improves life once I've subscribed to your OPML in my compatible feed reader-- but establishing that subscription still seems to require the same-old copying-and-pasting-urls-to-XML-files, right? Or am I just missing something?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/scripting/Scripting-News/issues/113#issuecomment-476662233, or mute the thread https://github.com/notifications/unsubscribe-auth/ABm9O7pPAtMssrkfTZC3BFATTNJtzsOZks5vaiqigaJpZM4cHH6h .

funkyboy commented 5 years ago

I am intrigued by the idea but I don't have clear the difference between importing OPML and subscribing to OPML. That said, I'd be all in with anything that allows one click subscriptions. At the moment you really have to be motivated to subscribe to a feed. Usually it's:

Services could enhance this by shipping browser extensions. Feed readers for macOS/iOS could include an extension that simplifies this. That would be an alternative/complement to supporting feed://-like URLs.

I agree with the "one step at a time". I just don't know which is the simplest first step :)

sillygwailo commented 5 years ago

@funkyboy My understanding of subscribing to an OPML file is that it's sort of like an import, except if the maintainer removes or adds a feed or updates its URL, that feed is removed or added updated as you see it in your reader. You'd be subscribing to a curated list of feeds, not a one-time import. A reading list, as it were.

A fictional example: Say you were an expert in birdwatching, and you maintained an evolving list of birdwatching blogs and other news feeds, but instead of it being a list on a sidebar (what we at one time called a blogroll), it'd be an OPML file, and people could just subscribe to that and get all the feeds. Maybe some bloggers stop talking about birdwatching, or a new birdwatching blog emerges. You as the maintainer of the birdwatching OPML file would do the subscribing and unsubscribing, and those interested in birdwatching could subscribe to one feed and rely on you to maintain their list for them.

A one-time import would get stale after a while. A maintained OPML file would be fresh, assuming the maintainer keeps it up to date and hands it off to someone else when they move on.

scripting commented 5 years ago

@sillygwailo -- that's exactly right.

jsavin commented 5 years ago

Maybe I should have posted this here instead of on the thread about the feed: protocol...

It looks like Inoreader might support OPML subscriptions (still). They announced it in a blog post from 2014, but I'm not an Inoreader user and haven't looked to see if it's still there or not: https://blog.inoreader.com/2014/05/opml-subscriptions.html

funkyboy commented 5 years ago

Thanks for the clarification @sillygwailo

Then how you’d implement states like “read” or “bookmarked” for entries?

I think it’s a fascinating problem to solve. I am mulling over it, but so far no ideas :(

sillygwailo commented 5 years ago

Then how you’d implement states like “read” or “bookmarked” for entries?

@funkyboy Same way RSS readers do now, however that is. Individual feeds would just be individual feeds. No change there.

daveajones commented 5 years ago

@sillygwailo The main decision to make when implementing this is around whether or not you will remove feeds from the aggregator when the OPML file(s) you're following removes them. There's a line where following somebody else's interests becomes them literally controlling your news river if deletion is supported.

scripting commented 5 years ago

@daveajones -- maybe we've had this discussion before, it sounds familiar.

It's not a problem. Keep a reference count for each feed. Only stop reading it when the refcount goes down to zero.

So if you subscribe to my feed on your own, and it also happens to be in a reading list, if it disappeared from the reading list, your aggregator wouldn't unsub because it would have a refcount of 1.

This is well-understood tech, many garbage collection systems use this technique to know if a block of memory can be released. When the count goes to zero, no one cares, so you can deallocate it, the equvialent of unsubbing.

funkyboy commented 5 years ago

@sillygwailo I guess that depends on the client implementation. I was thinking the service that offers OPML subscription could take care also of synching the state of entries. But it could also be a separate service or a client side functionality. Or both :)

eamercadoq commented 5 years ago

Maybe I should have posted this here instead of on the thread about the feed: protocol...

It looks like Inoreader might support OPML subscriptions (still). They announced it in a blog post from 2014, but I'm not an Inoreader user and haven't looked to see if it's still there or not: https://blog.inoreader.com/2014/05/opml-subscriptions.html

image @jsavin It´s already live,

cisene commented 5 years ago

I build nightly OPMLs for podcast feeds, this is a small flow in a small project, this is focused on audio podcasts only. http://b19.se/data/