from Hacker News

RSS Server Side Reader

by Bogdanp on 6/26/25, 12:11 PM with 51 comments

  • by geoffeg on 6/26/25, 3:13 PM

    I love that we're rediscovering server side rendering. It's impressive to me how quickly the industry moved away from it as the default and is now an interesting, new technique to many.

  • by quaintdev on 6/26/25, 2:56 PM

    I read the entire article and I still have no idea what author was trying to achieve.

  • by susam on 6/26/25, 5:07 PM

    I did pretty much the same thing last weekend. I have a small Common Lisp program that generates my website. Last weekend, I wrote a new program to generate a blogroll page from a list of feed URLs. You can see the code and the resulting blogroll here:

    https://github.com/susam/susam.net/blob/main/roll.lisp

    https://susam.net/roll.html

    This program was directly inspired by @matklad's idea of using a blogroll as an RSS reader. It's only been a few days, but I already feel like I can finally stay on top of my feeds!

  • by jrm4 on 6/26/25, 5:07 PM

    Yup. I'm no programmer by any stretch, but I wrote my own crappy little thing years ago and I still use it nearly daily. One thing I realized is that everything else I used had too many features; one I really never ever wanted was a count of what I had and hadn't read. I'd rather use RSS really as "maybe some days I pick up the newspaper and some I don't"

    https://gitlab.com/jrm4/mahrss (I don't even remember if this is the last version and make no guarantees about any kind of functionality, but you can get an idea)

  • by AndyKelley on 6/26/25, 3:32 PM

    HN seems to be having trouble understanding why this is brilliant.

    Ignore the parts about JSON/XML. That's irrelevant.

    Problem: you want an RSS reader, but RSS readers are annoying because they are stateful and you have to try to sync them across devices. Or, as in the case of Google Reader they may be discontinued. Best case, you have a dependency on a third party application.

    Solution: make a web page on your personal site that aggregates links from your RSS feeds.

    This is handy because you can now simply access your own web site as an RSS reader. As a side benefit, you can share this page with your friends to help them find nice links, and help promote stuff that you like to search engines.

  • by walterbell on 6/26/25, 4:46 PM

    On iOS, lire caches RSS-syndicated articles and images for offline reading. One-time purchase, no subscription. There's also OSS NetNewsWire.

  • by rambambram on 6/26/25, 6:09 PM

    I built and use HeyHomepage.com daily for the same reasons. I love having a list of all kinds of websites (mostly 'dev blogs' from all you guys and gals!) which in it's essence is a list of bookmarks to homepages. I throw RSS techniques against it and it becomes my replacement for social media. I call RSS therefor 'Really Social Sites'. No ads and the quality of the content is completely up to me selecting the right websites to follow.

    I see only one post from one feed at a time. If I want the next, I click a button labeled 'Next random post' and it gives me the latest post from yet another feed. I only get the first two lines of a new post and then read the full article on the website if I'm interested. There's something to say for paying hommage to a fellow internet user who put work in building a website... I want to read your post on your website or homepage.

    Expanding on this list I also published (parts of) this list as a sort of blogroll, or shared list. This shared list is viewable and clickable in the browser for regular internet users, and downloadable as an OPML file so RSS users can import the websites that I like and 'endorse'.

    Also expanding on this list, I built some functionality called Newspaper which automatically (instead of manually) checks some selected feeds that I deemed extra interesting. The different articles from different sources are than presented to me in a newspaper. Every time I log in there are some newspapers waiting for me. I'm always looking forward to the one called 'Cars'. The rest is mostly work stuff. ;)

    It goes without saying that Hey Homepage is not only an RSS reader, but that it also has functions for your own timeline of posts with accompanying RSS feed.

    The open web is not dead. You neglected it for too long. It misses you. Give it some love back.

  • by rpdillon on 6/26/25, 4:56 PM

    I'm a GReader refugee that migrated to TT-RSS, and then FreshRSS. But somewhere in there, I wrote my own version of this same idea: pulling RSS feeds server side via cron every several hours and rendering a rollup. Never put the code online, but here's a paste of it if anyone is interested. I called it "Cooler" since it was designed to surface good water cooler chat topics. It's just a single Python script with a couple of dependencies.

    https://textbin.net/kdhkz0nnyx

    It tracks no state, but does color by source, and fades with age so you get a good feel of what site the post came from and how long ago. I had an idea to version control the static HTML after it was generated, so you could rewind time and see what the top stories were, but haven't gotten around to it.

    If anyone cares, I'll host the full repo and share.

  • by tolerance on 6/26/25, 3:52 PM

  • by m-localhost on 6/26/25, 7:01 PM

    Unrelated, bit it's fascinating that the Google Reader API (which is super weird I think) is still the status quo for interacting with feed readers (https://freshrss.github.io/FreshRSS/en/developers/06_GoogleR...)

  • by renegat0x0 on 6/26/25, 8:02 PM

    For some time I have been playing with RSS and web crawling myself.

    RSS is not that simple and all common properties are read into a simple JSON data. Maybe somebody will find it useful.

    https://github.com/rumca-js/crawler-buddy

  • by eviks on 6/26/25, 5:13 PM

    > If I don’t remember whether I read the article or not, I might as well re-read!

    Unless you remember after reading the first paragraph or two, in which case you've just wasted time partially re-reading them. This type of tracking is perfect for offloading to a reliable digital accountant!

  • by chazeon on 6/26/25, 5:52 PM

    The thing is, the publisher are not obligated to show all recent post, some may show 1 some may show 5, but they might publish more than what they show in your refresh interval (e.g. 1 week) In this case, using a stateless reader, you will start to miss article.

  • by rglullis on 6/26/25, 3:18 PM

    Tell me you are JavaScript developer who likes to reinvent the wheel, without telling me you are a JavaScript developer who likes to reinvent the wheel:

    - "RSS is for notifications". No, it's for content syndication. It is right there in the name.

    - "XML is complicated, JSON Feed is better". Oh, dear Lord, forgive him for he has no idea what he is saying.

    - "Lets ignore all the gazillion libraries for and tools for parsing and processing OPML, Atom and XML so that we can build a system that depends on deno a f*cking GitHub actions"

  • by sebastian_honsa on 6/26/25, 5:06 PM

    I've made https://rosselo.com exactly for being a RSS server side dashboard of news.

    Feel free to try it out, it's completely free for now and upcoming future!

  • by p0w3n3d on 6/26/25, 3:28 PM

    I remember using RSS heavily around 2007-2013 but is this still a thing?

  • by DannyPage on 6/26/25, 3:17 PM

    Interesting idea, but it mentions the idea of notifications. Is the notification when happens to be new article is at the top on the page? Or is there an external service he forgot to metnion?