5 ms·
I recently started to consume content via RSS again. It turns out, most of the blogs I care about have a working RSS feed. I'm using newsblur (no affiliation
by ndr 6y ago
I recently started to consume content via RSS again.
It turns out, most of the blogs I care about have a working RSS feed.
I'm using newsblur (no affiliation) and it's working quite well, I no longer need to poll the websites, or wait for posts to appear here or on Twitter.
The nicest side effect is that checking newsblur before HN/Twitter limited my mindless scrolling, although this comment proves it didn't eliminate it.
- vanderZwan 6y agoNewsblur also lets you subscribe to YT channels, which is much nicer experience than YT's own way of handling subscriptions
- addicted 6y ago{RSS Reader} also lets you subscribe to {Feed based Service} which is a much nicer experience than {Feed Based Service}'s own way of handling subscriptions. ----- I've found that the above statement works for a very large set of values for {Feed Based Service}. One really great example was when I setup Redmine (open source Jira alternative) for my company's project management needs. Instead of having it email me notifications, having them in the Reader was amazing.
- waterhouse 6y agoHmm, interesting. Is it based on Youtube's raw RSS features? Because you can use RSS for an entire channel, or for a playlist, but the latter doesn't work if there are more than 15 (IIRC) videos in the playlist. Does Newsblur do something specific to Youtube to work around that? Also, does any RSS reader have some kind of regex filtering option? "I want all the items from this RSS feed that match the following regex". While I'm at it, how about getting the output of an arbitrary program? (Use case: websites that have a page with what I want—e.g. every article published by a certain author—but have no built-in RSS functionality. Ideally I would specify the program as "curl [URL] | egrep -o [stuff]", and it would generate new RSS entries anytime that output changed.) I know I can kludge together a cron job that runs such a program and generates a file, and point my RSS reader at the URL "file://path/to/output.xml", but is there anything with built-in functionality like that?
- alex_duf 6y agoI think youtube had killed every RSS endpoint, I remember channels disappearing from my RSS app. So either youtube has done the right thing or that specific reader is scrapping the HTML of the channel pages.
- waterhouse 6y agoYou can get a Youtube channel's Atom feed as something like this: https://www.youtube.com/feeds/videos.xml?channel_id=UC0Xe2bvq_2uucE2IRALSR_Q https://www.youtube.com/feeds/videos.xml?channel_id=UC0Xe2bv... From the channel's URL like this: https://www.youtube.com/channel/UC0Xe2bvq_2uucE2IRALSR_Q https://www.youtube.com/channel/UC0Xe2bvq_2uucE2IRALSR_Q
- tomjen3 6y agoI am subscribing to youtube channels through thunderbird. It works well for right now and it makes it easy to either open the url or copy it to youtube-dl or mpv for out of browser watching, but getting the rss url for a named channel is so janky now, and it feels like it works only because nobody has turned it of, like it is one design tweak from being disabled entirely. Ah well, I will continue subscribing in Thunderbird until whomever keeps RSS alive at youtube (thank you, if you are reading this) moves on to better things. Then I will watch it sunset just as I did with google music, G+, Google reader and all the rest Google has killed over the years.
- ecliptik 6y agoI'm not entirely sure how the Newsblur youtube feeds work, but here's an older blog post on switching to API v2, https://blog.newsblur.com/post/117767869761/a-real-solution-to-the-deprecated-youtube-api https://blog.newsblur.com/post/117767869761/a-real-solution-... I have about 20 youtube channels in Newsblur and watch them almost always within Newsblur itself, so it works well enough. For regex filtering, Newsblur has training features. You can curate a feed based off specific attributes, like author, topic, keywords, etc.
- waterhouse 6y ago