Exposed RSS | Hacker News Reader

Exposed RSS

chriscoyier.net

116 pointsalexzeitler30 comments

I always like to put a link to the RSS in the actual HTML (i.e. so a human can see it)

However, clicking it is horrible UX: a wall of text. Or even worse (shaking fist at Firefox), a download "Save as..." prompt.

So... <shameless self promotion> I made Feed.Style[0] to help people make it be a semi-presentable page by adding an XSLT stylesheet.</shame>

I have a couple of examples (of HN and high-ranking HNers natch ;) and found a fair number of bugs^H^H^H^H non-optimal choices in their feeds, probably because XML is not all that human-readable, even for devs.

Source [1] is MIT-licensed and is completely static, except for the examples, which are just an optional hack running on Cloudflare pages.

Feedback welcome!

[0] https://www.feed.style/

[1] https://github.com/fileformat/feed.style

I've seen RSS styling before but it does feel like an oddity. I think I'm so used to the code view that when I see a styled feed page, it makes me wonder if I'm on the real URL. But at least the page does explain itself.

As a side note, it seems like a number of your "after" examples do not load. HN hug, or something else?

> I've seen RSS styling before but it does feel like an oddity. I think I'm so used to the code view that when I see a styled feed page, it makes me wonder if I'm on the real URL. But at least the page does explain itself.

Indeed, that was my greatest concern for adding a stylesheet to the RSS on my website. I tried to solve it by including a screenshot of the source tree :)

On my website, the RSS is announced through metadata only on the news page. Per the “Exposed RSS” statement, it should be linked from all pages? I wonder whether it really makes sense for someone to subscribe to a page without even checking it out on such a cursory level as to find the news page, though?

Hmm... The examples are running on the free tier of Cloudflare Pages Functions. I should be dramatically under any CPU/networking/transfer limits. And it should scale up as needed.

I just tried all of them now, and they did load. What is the exact error you are seeing? Just a blank page or ???

If you mean that the input box for "copy to clipboard" is empty, that's because the feed doesn't have a "self" link, which the stylesheet needs.

I'm not sure then. The first two "after" examples were blank for me. I tried again now and the first loaded, but second did not. After another few refreshes, it did load, though as you said the "copy and paste" link was blank.

I tried it in Firefox and got an XML parsing error[1] instead. Again refreshed and it loaded fine. So it seems intermittent at best.

[1] Error follows:

  XML Parsing Error: mismatched tag. Expected: </link>.
  Location: https://www.feed.style/example.xml?feedurl=https%3A%2F%2Fnews.ycombinator.com%2Fshowrss
  Line Number 6, Column 3:

Probably needs something like how openrss.org does it. Shows a UX friendly view for RSS but gets picked up by RSS readers as well.

/me shakes fist at general direction of Firefox for removing RSS button, as the browser should be the one informing the user about additional features that website provides.

<https://addons.mozilla.org/en-US/firefox/addon/want-my-rss/>

This is what I use as well. I hope it gets support for Firefox Android some day.

For Firefox Android, there’s <https://addons.mozilla.org/en-US/android/addon/awesome-rss/>

It seems to lack a customizable destination which is a non-starter for me.

https://addons.mozilla.org/en-US/firefox/addon/drop-feeds/ is probably the most feature-rich RSS extension for Firefox (desktop)

totally agree that this should be a built-in browser feature... in safari there's this paid ($1) extension: https://rssdiscovery.app

Even worse than this are sites that have RSS feeds and Cloudflare in front. Because my RSS reader doesn't look much different than a bot, and won't complete Cloudflare's CAPTCHA.

My feed reader is a swlf-hosted in Linode. I get blocked by Captcha too - but Cloudflare loves my home IP. Wireguard to the rescue.

I'm wondering if you could use something like FlareSolverr to bypass the Cloudflare CAPTCHAs.

I'm a little surprised an RSS feed itself would be rate limited/protected though.

I was running into something similar when I was accessing the API of someone's MediaWiki instance and adding User-Agent was enough to solve the problem

My favorite Firefox feature was a button that copied a site's RSS feed to your clipboard. Sadly, they removed it a while back.

https://addons.mozilla.org/en-US/firefox/addon/want-my-rss/

Thank you! But the reason I used that feature was because it was Mozilla official code and not some third-party add-on that might get sold behind my back three months later.

I sometimes extract a browser extension’s source and manually install it, if I find it useful, so that my copy will never turn into spyware or be removed from the store.

Sometime I wonder if their main revenue source (Google) asked them to remove features like this one...

I just recently found out that you can add /feed after a Substack publication's URL and subscribe via RSS. (e.g., https://[publication].substack.com/feed)

You can do something similar with Buttondown: https://buttondown.email/<publisher>/rss (for example, https://buttondown.email/j2kun/rss)

My feed reader just tries a variety of URLs if the page doesn't have autodiscovery. It works surprisingly well:

https://github.com/fazalmajid/temboz/blob/master/tembozapp/a...

Of course, my hand-rolled site (https://joeldueck.com) uses feed.atom (don’t ask me why, it was a long time ago), which happens not to be in your list. I do have it exposed in the HTML though.

I add them as I find new ones. Note that your site throws the following exception in my locked-down browser where cookies and other persistent storage is disabled unless explicitly allowlisted:

failed to read the 'localstorage' property from 'window': access is denied for this document.

...presumably using your splitflap project in Racket, which I use for my own site! Thank you for your work on it.

Thanks! I have zero idea how many people use it, so running into someone who does is very encouraging.

You can also have a discoverable OPML with many interesting feeds in it. I'm not aware of any clients that can find it but I think it useful enough for someone to one day implement it. Seeing just one feed appear in your aggregator just for the page you are looking at right now (while very useful later on) isn't as exciting as 10 or 100 new sub worthy subjects.

Reply on news.ycombinator.com