Hello! author here, i just wanted to address a few things i've seen in the comments:
* i absolutely built this for fun, it's an idea i had late at night and needed to scratch that itch
* the scolling sucks, laying out arbitrarily sized articles into a given layout is a challenge, and the scrolling hides a lot of the sins. I'm going to switch to truncating articles this evening, and linking to a single article page.
* thanks for the suggestions on hyphenation, i've been playing with a few different approaches but none have worked with all the browsers i tried, i'll give a few suggestions a go this evening!
Love to see experiments like this with web articles. Have you thought about making it more print focused? It's the kind of layout I'd want to print out and read offline.
I've experimented a little with trying to create more readable, printable layouts of web articles and feeds on https://www.fivefilters.org [1],[2] but yours has a much more traditional newspaper look which could look great in print.
They have some really nice ideas, implemented well. See for example how the articles click to expand. And the horizontal scrolling there works so well with the columns!
Wow, what a great idea, and largely fantastic implementation! It really looks good.
Do RSS feeds have "just" text that you pull to get the article content, or are you parsing the webpage somehow? If so, how?
I've done something slightly (well, about 1% :-) similar for the "Popular" page on pinboard[0]. It used to have a line or two from the start of the article, using a webpage content extractor that got turned off, so I can't use it anymore.
One arguably nice thing about mine is that it's updated once a day, and it remembers what you've read yesterday, so new articles are marked, and you can flip a switch to see the new stuff first.
RSS feeds are a bit of a mess, but only due to each publisher's implementation being slightly different. the vast majority only send a headline and summary on the RSS feed, i then have to go and scrape and extract the article on the backend to populate the content, which is it's own challenge and i've not gotten it to work for a few sites yet.
This is also running in a semi-serverless container in Google Cloud Run (only costs me £1 a month!) so fetching and re-caching all of that when a new container is scheduled is painful, however it seems like state in the container is persisted longer than i initially thought, so it's good enough for now.
You might have come across this already, but we maintain a collection of article extraction rules for various sites here https://github.com/fivefilters/ftr-site-config - it was adapted from a database maintained by Instapaper in its early days and today has contributions mainly from users and developers of an open source Instapaper/Pocket alternative called Wallabag: https://github.com/wallabag/wallabag
This looks really interesting! I've been building an RSS reader for myself in my free time, and this will be really useful. I was wondering if you know anything about the legal implications around scraping full-content like this and packaging it up? I was planning to do it with some fun added-on features, but was worried it would be considered copy-right infringement (since I would basically be re-hosting other site's content without permission). And some websites outright ban this kind of usage in the TOS for their RSS feeds. For example, from the Washington Post[1]
> a. For any article, you may not display more text than we provide in the RSS feed.
> I'm going to switch to truncating articles this evening, and linking to a single article page.
Ah, much better now.
Paper newspapers do this in spades, so this behavior shouldn’t surprise most people.
I hate to say his, but you might have to turn on justification to make it look more paper-like. In doing so you can get rid of many of the vertical separator lines (but you might need to be more conservative with max columns)
Now if you could just get an AI to rewrite the content to bring it up to the old (roughly pre-"USA Today") journalistic standards, it would be perfect!
I love the scrolling! My first thought was "finally a way to read the whole rss feed without the need to click on a link". I'd keep it that way.
Also, I think that the scrolling feature is one of the most innovative things I've seen in many years on how to format articles on the web. It makes a lot of sense to be honest IMH.
* i absolutely built this for fun, it's an idea i had late at night and needed to scratch that itch
* the scolling sucks, laying out arbitrarily sized articles into a given layout is a challenge, and the scrolling hides a lot of the sins. I'm going to switch to truncating articles this evening, and linking to a single article page.
* thanks for the suggestions on hyphenation, i've been playing with a few different approaches but none have worked with all the browsers i tried, i'll give a few suggestions a go this evening!