There’s a new, fun aggregator out there: bubbles.town. I think the pitch is “like Hacker News, but for independent blogs instead of Silly Valley hucksters”. So it’s got upvoting and comments and all of that stuff.
Like Hacker News, it also has feeds. The feeds are nicer than on Hacker News — they have snippets of texts from the pages they link to:
But it’s only a snippet, so I wanted to make a fuller feed, like I’ve been making for Hacker News.
So here it is: Fuller Bubbles.
The source code is on Microsoft Github. I’ve just tweaked the original script a tiny bit more, and it works by going through the feed and then grabbing text from the original pages and putting that in there.
The point of all this is to be able to read the articles themselves in your favourite feed reader, i.e., in Gnus, reading from the Gwene NNTP server. (Isn’t that your favourite feed reader?! WELL I NEVER.)
See? All purty.
Of course, it doesn’t work reliably — it’s getting more difficult all the time do to web scraping, and especially from data centres (and that’s where this is running). But it works well enough for it to be worth it, I think.
Hi, Bubbles dev here. Nice addition! Just so you know, there’s a full-text feed at bubbles.town/feed/full (linked on the /rss page) that pulls in whatever authors include in their original feeds. Most blogs syndicate full text anyway, so it covers the majority of posts.
The reason I don’t scrape article bodies myself is because I’d rather let authors decide what gets syndicated. Not a judgment on what you’ve built for personal use, just the policy I’ve settled on for the official feeds.
Ah, didn’t see that feed. But it’s a feed of all new posts? Could you make one that has just the posts that get to the front page, but that includes the text? 😀 I know, so many different feed combinations…
Sure thing, I just updated all feeds to include full text (where the original author syndicates it). Still no scraping, some entries might not have full-text.
Great! Looking at the feed, it looks much better than my cobbled-together version — since web scraping is getting to be impossible, about half the articles in my version of the RSS feed had no text, while in your version, it looks like the vast majority has (more or less) the full text.
So I think I’ll just discontinue my version of the RSS feed…