Which I find intensely funny, because it’s forced on all EU IP addresses as an ostensible downgrade, but since they’ve added a decent CSS snippet to limit the line width I actually find it superior to the “full” version.
Not sure of the premise behind this, so maybe I’m missing the point, but why not curate quality news sources and access them via an rss news reader like NetNewsWire?
No, people just assume that they don't have one due to no RSS icon. `/feed` slug is standard for finding feeds these days.
Equally, why not just use one of the many FOSS tools to scrape sites that don't offer RSS feeds and DIY your own? This is a long-solved problem; RSS remains wonderful.
I love RSS. I actually believe that it may become more relevant in the future, now that AI is so good at producing noise.
It will be more and more difficult to make a Google/DDG/Kagi search and spot the autogenerated crap. So maybe it will be time to manually select blogs/websites we like and trust. And RSS is amazing for that.
100% my thoughts, mate. I definitely see it becoming the go-to tool for many in the future. It's got a lot going for it, including authenticated access to premium feeds, ergo people can monetize and allow their content to be consumed in a way that makes sense to the end user.
Still one of my favourite things to emerge on the net. Especially being able to filter through feeds!
I've been using many of the linked sites from the EWW browser in Emacs. I also use RSS for many things but I don't want that model for general news items. For that I prefer to have the snapshot view that the web pages provide.
Check out https://markets.sh news. They are actually text only, from many different sources, clustered and summarized. It is really good to get a gist of what’s currently important without having to “read news” with the known nuisances that come with it.
Edit, thank you for the feedback. Some clarifications:
- we launched the news feature literally yesterday, it is MVP level so expect inaccuracies especially in the summaries. We are using our own models and are in the process of tuning and refining them.
Clicking on the cluster will give you the actual headline, titles and sources for each cluster.
- the ordering of the feed is super simple right now. We will improve the weighting based on recency, magnitude of the story, coverage, parties involved etc.
- This is not text-only like text.npr.org but in the sense of not being stuffed with ads, autoplay videos and images. Both a real text-only statically rendered page and RSS feed are in the works.
Thanks! Yeah, I’m one of the founders. It’s Nextjs on Supabase. We launched the news clustering just yesterday so I thought it would be nice to get some stealth HN feedback for it.
We are planning to launch an API (rss style) for the news next week where you can get temporal clusters and everything. Imagine having full access to all of Twitter trending but across all news sources.
We are trying to filter clickbait, SEO spam as effectively as possible while keeping nuance to the dataset.
Please sign up for an account or follow us on twitter if you would like to stay in the loop.
Genuine question: I wonder why you didn't think to disclose your relationship with the product? I notice this a lot, and it ruins my perspective of the product. Haven't you seen people start their comments with "full disclosure, I made this" type sentences.
Was it a relevant suggestion for you? If yes, then what benefit does the disclosure provide? If not, then downvote and say so?
In this particular case, I think halotrope just didn't think the dislosure was necessary. They have a link to the website in their profile and aren't trying to hide the affiliation.
Full disclosure: I'm not related to halotrope in any way nor to the website in question.
Since you asked for some, here's my immediate 2 cents:
Please change the slide out on the left to be visible permanently.
I have 1920 pixels of horizontal screen real estate, and the page is quite literally half empty whitespace anyway. For what conceivable reason is that menu on the left a hidden-by-default slide out? It even hides pertinent stock information.
Also concerning the stock information, please add some whitespace margin on the bottom. Chrome displays URL information on the bottom left of the window when hovering links and it overlaps and obstructs visibility of whatever stock ends up displayed at the bottom left of the window.
I like this, is your quote data live or delayed? Sparklines in your watchlist would be most welcome, plus better charts in general; candlesticks, range % between two points etc.
Thank you! We have realtime data but IEX only, sorting the contract work out for at the moment to get tick-level realtime for US equities
The sparklines and better charts are being worked on. We hope to get a new version out early April, right now focussing on the API and Chat functionality.
It would be nice if your interface sorted the sources by bias and let you side between different summaries. In the upper right corner of Modo you can swap between timeline view (seeing past incarnations of a story as it progresses, or lean view that let’s you see various bias.)
That sounds very nice, I've been looking for a good service like this. Is there a way to subscribe for updates so I know when new features are released?
The icons are super small but I get your point. Text-only in a way of no autoplay video, ads, big pictures etc right now. We are planning to launch true (like text.npr.org) text version and RSS as well.
The summaries for several headlines are completely bogus, which undermines any credibility that this was a reliable site. For example, the Lufthansa Strike blames a technical issue, not the strike that all linked articles discuss. Next, the Khazahstan election summary mentions voters in Turkey and Turkmenistan. There’s also duplicated stories, and other obvious issues any human editor would catch.
I suspect this is secretly powered by some AI that doesn’t actually work all that well.
Hi, thank you for the feedback. I should have added that we launched it yesterday so it there is still some quirks. Yes, we use language models to do the summaries, they are not perfect yet and the feed ordering also needs to be improved.
Since you get the actual news headlines and teaser on the detail, we assume some inaccuracies are ok for now. The primary focus right now is clustering and a high-level overview over the news landscape. Especially to get perspectives about an event from many different sources.
We are iterating quickly and expect this to somewhat stable and reliable by mid April.
Would appreciate if you had a look back again then to share your feedback.
Perhaps I'm missing something but for a text-only interface, why do you need javescript? I always disable it in my browser and your site shows very little without it. Ran it in a VM to allow JS see what it did and got "Application error: a client-side exception has occurred (see the browser console for more information)."
No worries, it is actually just a cdn-cached next view with a barebones layout on the feed.
We really care about internet culture, bandwidth and consumption options and try to facilitate them wherever feasible.
Me too! Luckily the open web now gets competition from LLMs. Hope this will clean up things. Seeing all these sloppy pages full of popups, ads and shit typography makes me so sad
Alright, I still can't find any button though. By the way, newsletter notifications were automatically checked, I think most users would prefer to have it disabled by default.
Until a few days ago the best solution for text only + no ads + offline reading was a kindle subscription to whichever periodical you fancy. This is gone now.
Nook still has a newstand store (for now?) but I haven't used it so can't comment on the formatting and UX.
I still subscribe to text (and images, but no video), no ads, offline reading news sites on my Kindle. Not through the Kindle subscription but through a local downloader (Calibre).
One issue is that news sites article content is written to maximise ad impressions. Ads inject between paragraphs typically so you need lots of paragraphs. You start with perhaps two paragraphs of actual information, then break it down into 4 or more and add more information about related topics. Add some opinions, maybe weave in links to related articles on your site, and you end up with 10 or so paragraphs insterspersed with ads and pictures. You can remove the ads and pictures but you can't remove the bullshit. We need a good AI tldr machine.
I've counted upwards of ten advertising insertions in WaPo articles. Between paragraphs.
On desktop, I'll nuke any interstitial element (including "related stories" and the like) on sites. The calming factor between post- and pre-edited sites is ... somewhat nuts. One of my faves was old-school Buzzfeed, where I nuked anything but the actual headline and feature story, which I'd called "Unbuzzed".
(And yes, "Buzzfeed" itself is mostly trash, "Buzzfeed News" was/is actually somewhat respectable. I generally didn't seek out the sites, but occasionally clicked through on links from elswewhere.)
> Ideal if you’re the kind of person who just quickly needs to check the feed and go away again. There’s no javascript so it feels (and definitely is) faster and less bloated. The design is nicely old fashioned.
I've been using this site for years but it's increasingly bit rotting and regularly serves broken links.
For sports, ESPN does have http://www.espn.com/nba/lite/scoreboard, but most of the links go back to the normal website. It may have been been more robust in the past, but I don't imagine anyone works on it anymore. It seems like the developer listed in the credits (http://www.espn.com/espn/lite/credits) last worked there in 2002!!
So if you're looking for a sports alternative, https://plaintextsports.com (which I made) works great! All the scores, play-by-play, box scores, standings, and schedules, but just no news stories. Blazing fast.
(No, it's not technically "Content-Type: text/plain", it uses HTML and CSS. Yes, I know it's not necessarily easier to read; it's an aesthetic. Yes, this is shameless self-promotion.)
I get most of my daily news from ... Teletext [1]. I'm lucky to be in a country that still has it.
I much prefer that format over news sites on the web: Headlines are often laid out in a list one entry below the other, which is easy to read through without having to scroll.
Because pages are limited to 40 columns × 24 rows, every article is short and to the point.
I don't usually read it on TV though but on a web-site [2] which has transformed page numbers into hyperlinks, and given multi-pages a tab-like interface.
There are still no images, no ads ... and especially no auto-playing videos. Perfect!
Recently though, I've spent a lot of time in a hospital bed and it has been easier for me then to use the TV remote with one hand to check teletext than to use the tablet or smartphone.
We have teletext in Spain too, (several of them) but as a "Computer" alternative I use EFE's (state news agency) RSS under sfeed+uxterm with the Unifont font.
VIEWER:application/postscript:fbgs %s
VIEWER:image/gif:sxiv -a %s
VIEWER:image/x-xbm:sxiv -a %s
VIEWER:image/png:sxiv -a %s
VIEWER:image/tiff:sxiv -a %s
VIEWER:image/jpeg:sxiv -a %s
VIEWER:video/mpeg:mpv %s
VIEWER:video/mp4:mpv %s
I too was using text-only versions of sites like CNN, Reuters, or Christian Science Monitor[1], and they were fine. But what I really wanted was to turn any news website into a text-only website.
* Automatically builds a list of news stores, separate from the navigational hyperlinks.
* Detects RSS/Atom feeds to provide a more accurate list of news stories.
* Uses Readability to show only article content on article pages.
* Uses meta data like OpenGraph or Twitter cards to provide richer formatting, and to determine page type.
It regularly converts 900 KB home pages or 1.2 MB news articles into into 3KB for links to news stories and 5K of text
It does this by:
* Using semantic tags like <header>, <footer>, and <nav> to determines which hyperlinks are navigational and which ones are likely links to news articles.
* OpenGraph meta data to determine page type news stories and extra metadata.
* A Aggressive HTML parser that strips out a ton of tags, CSS, JS, etc
* Readability library to extract out the text of news articles
I built this as a service in Gemini, so if you have a gemini browser you can try it. Otherwise, here is a HTTP-to-gemini proxy showing you what a NYT article looks like:
Clicking on the "more" links which take you to the news articles also works properly as well.
(you can get to raw mode by clicking "Force article view" and then "raw mode." I should probably expose that in other places)
NewsWaffle tries to determine the type of page. Articles get displayed with content run through readability, and then the HTML is stripped down. If its a "links" page, like the home or section page on a news site, it using HTML elements to try and find links to news stories vs navigational links to other parts of the site. Part of that is looking for links with longer text, since link text to news stories tend to be a few words. This helps sort "About Us" from "New Fusion Experiment a Success"). I'll check into why aldaily isn't working properly
Sorry I can't seem to reproduce the Techmeme issue. It works for me:
Lagrange is sort of the Netscape of Gemini. It works on all the major desktop and mobile OSes. Personally prefer Elaho (iOS) or Buran (Android) for mobile
Text-based news pages are better because they allow readers to quickly scan through and absorb information, without being distracted by flashy graphics or autoplay videos.
133 comments
[ 6.9 ms ] story [ 297 ms ] threadLots of great sources in the link below. I already had my own list of the usual players but this link covers many more I didn't know existed:
https://greycoder.com/a-list-of-text-only-new-sites/
The only site that I rarely see mentioned (or included in the link above) is : http://www.dailymail.co.uk/textbased/channel-561/index.html
- He sees no point in adding another layer of abstraction and complexity to something simple.
- he's into retro computing, and old machines can't handle NNW.
- Fewer and fewer web sites support RSS, so using NNW limits his options.
- Sometimes you want to read the news from a particular web site, and not graze at a buffet.
Equally, why not just use one of the many FOSS tools to scrape sites that don't offer RSS feeds and DIY your own? This is a long-solved problem; RSS remains wonderful.
It will be more and more difficult to make a Google/DDG/Kagi search and spot the autogenerated crap. So maybe it will be time to manually select blogs/websites we like and trust. And RSS is amazing for that.
Still one of my favourite things to emerge on the net. Especially being able to filter through feeds!
According to the people complaining on HN, yes. It's been a common topic of conversation on this web site for the last five years or so.
No RSS here: https://www.rfi.fr
The New York Times does RSS, but among major news organizations, it is one of the exceptions.
All blog frameworks have support by default...
Good for blogs.
Edit, thank you for the feedback. Some clarifications:
- we launched the news feature literally yesterday, it is MVP level so expect inaccuracies especially in the summaries. We are using our own models and are in the process of tuning and refining them. Clicking on the cluster will give you the actual headline, titles and sources for each cluster.
- the ordering of the feed is super simple right now. We will improve the weighting based on recency, magnitude of the story, coverage, parties involved etc.
- This is not text-only like text.npr.org but in the sense of not being stuffed with ads, autoplay videos and images. Both a real text-only statically rendered page and RSS feed are in the works.
I like the interface - compact and fast.
We are planning to launch an API (rss style) for the news next week where you can get temporal clusters and everything. Imagine having full access to all of Twitter trending but across all news sources.
We are trying to filter clickbait, SEO spam as effectively as possible while keeping nuance to the dataset.
Please sign up for an account or follow us on twitter if you would like to stay in the loop.
In this particular case, I think halotrope just didn't think the dislosure was necessary. They have a link to the website in their profile and aren't trying to hide the affiliation.
Full disclosure: I'm not related to halotrope in any way nor to the website in question.
Please change the slide out on the left to be visible permanently.
I have 1920 pixels of horizontal screen real estate, and the page is quite literally half empty whitespace anyway. For what conceivable reason is that menu on the left a hidden-by-default slide out? It even hides pertinent stock information.
Also concerning the stock information, please add some whitespace margin on the bottom. Chrome displays URL information on the bottom left of the window when hovering links and it overlaps and obstructs visibility of whatever stock ends up displayed at the bottom left of the window.
The sparklines and better charts are being worked on. We hope to get a new version out early April, right now focussing on the API and Chat functionality.
But dumb question, what do I follow ? :)
It would be nice if your interface sorted the sources by bias and let you side between different summaries. In the upper right corner of Modo you can swap between timeline view (seeing past incarnations of a story as it progresses, or lean view that let’s you see various bias.)
Interesting claim considering there are 276 image tags in the HTML: https://imgur.com/a/x9ODk0r
The summaries for several headlines are completely bogus, which undermines any credibility that this was a reliable site. For example, the Lufthansa Strike blames a technical issue, not the strike that all linked articles discuss. Next, the Khazahstan election summary mentions voters in Turkey and Turkmenistan. There’s also duplicated stories, and other obvious issues any human editor would catch.
I suspect this is secretly powered by some AI that doesn’t actually work all that well.
Since you get the actual news headlines and teaser on the detail, we assume some inaccuracies are ok for now. The primary focus right now is clustering and a high-level overview over the news landscape. Especially to get perspectives about an event from many different sources.
We are iterating quickly and expect this to somewhat stable and reliable by mid April.
Would appreciate if you had a look back again then to share your feedback.
The site is focussed on financial data, charts and realtime updates. This is why it is javascript heavy despite visually simple.
I wish 99% of sites gave a damn about these things. Well, I approve, and good luck!
If the latter, isn't that illegal or against their terms of service in some way?
Ah. That explains why in Lynx all I see is a dozen instances of:
Doesn't work on all articles: https://txtify.it
.. and already dated wrt Thomson Reuters Foundation Trust link ( news.trust.org ) which is now archived and replaced by
https://www.context.news/
Btw, Igor Chubin is the author of great CLI services such as: cheat.sh[1], late.nz[2], QRenco.de[3],rate.sx[4]
[0] https://github.com/chubin/awesome-console-services
[1] https://github.com/chubin/cheat.sh
[2] https://github.com/chubin/late.nz
[3] https://github.com/chubin/qrenco.de
[4] https://github.com/chubin/rate.sx
Nook still has a newstand store (for now?) but I haven't used it so can't comment on the formatting and UX.
On desktop, I'll nuke any interstitial element (including "related stories" and the like) on sites. The calming factor between post- and pre-edited sites is ... somewhat nuts. One of my faves was old-school Buzzfeed, where I nuked anything but the actual headline and feature story, which I'd called "Unbuzzed".
(And yes, "Buzzfeed" itself is mostly trash, "Buzzfeed News" was/is actually somewhat respectable. I generally didn't seek out the sites, but occasionally clicked through on links from elswewhere.)
> mbasic.facebook.com
> Ideal if you’re the kind of person who just quickly needs to check the feed and go away again. There’s no javascript so it feels (and definitely is) faster and less bloated. The design is nicely old fashioned.
I've been using this site for years but it's increasingly bit rotting and regularly serves broken links.
So if you're looking for a sports alternative, https://plaintextsports.com (which I made) works great! All the scores, play-by-play, box scores, standings, and schedules, but just no news stories. Blazing fast.
(No, it's not technically "Content-Type: text/plain", it uses HTML and CSS. Yes, I know it's not necessarily easier to read; it's an aesthetic. Yes, this is shameless self-promotion.)
I much prefer that format over news sites on the web: Headlines are often laid out in a list one entry below the other, which is easy to read through without having to scroll. Because pages are limited to 40 columns × 24 rows, every article is short and to the point.
I don't usually read it on TV though but on a web-site [2] which has transformed page numbers into hyperlinks, and given multi-pages a tab-like interface. There are still no images, no ads ... and especially no auto-playing videos. Perfect!
Recently though, I've spent a lot of time in a hospital bed and it has been easier for me then to use the TV remote with one hand to check teletext than to use the tablet or smartphone.
[1] <https://en.wikipedia.org/wiki/Teletext>
[2] <https://texttv.nu/> (Swedish SVT Text)
Pretty close to our different Teletext services.
I also use teddit.net which is essentially a plaintext reddit https://teddit.net/r/worldnews
There's a GitHub repository floating around somewhere with code for an rss feed of it which works great for getting it on mobile.
VIEWER:application/postscript:fbgs %s VIEWER:image/gif:sxiv -a %s VIEWER:image/x-xbm:sxiv -a %s VIEWER:image/png:sxiv -a %s VIEWER:image/tiff:sxiv -a %s VIEWER:image/jpeg:sxiv -a %s VIEWER:video/mpeg:mpv %s VIEWER:video/mp4:mpv %s
Install sxiv and mpv, of course.
So I build NewsWaffle, which for any website:
https://github.com/acidus99/NewsWaffle
* Automatically builds a list of news stores, separate from the navigational hyperlinks.
* Detects RSS/Atom feeds to provide a more accurate list of news stories.
* Uses Readability to show only article content on article pages.
* Uses meta data like OpenGraph or Twitter cards to provide richer formatting, and to determine page type.
It regularly converts 900 KB home pages or 1.2 MB news articles into into 3KB for links to news stories and 5K of text
It does this by:
* Using semantic tags like <header>, <footer>, and <nav> to determines which hyperlinks are navigational and which ones are likely links to news articles.
* OpenGraph meta data to determine page type news stories and extra metadata.
* A Aggressive HTML parser that strips out a ton of tags, CSS, JS, etc
* Readability library to extract out the text of news articles
I built this as a service in Gemini, so if you have a gemini browser you can try it. Otherwise, here is a HTTP-to-gemini proxy showing you what a NYT article looks like:
Gemini link: gemini://gemi.dev/cgi-bin/waffle.cgi/
NYT Homepage: https://portal.mozz.us/gemini/gemi.dev/cgi-bin/waffle.cgi/li...
NYT Article: https://portal.mozz.us/gemini/gemi.dev/cgi-bin/waffle.cgi/ar...
[1] https://www.csmonitor.com/text_edition
I tested aldaily.com and had trouble navigating to get to the articles. Allsides.com worked. Techmeme.com did not work.
gemini://gemi.dev/cgi-bin/waffle.cgi/links?https%3A%2F%2Fallsides.com%2F
https://portal.mozz.us/gemini/gemi.dev/cgi-bin/waffle.cgi/li...
gemini://gemi.dev/cgi-bin/waffle.cgi/raw?https%3A%2F%2Fwww.aldaily.com%2F
Clicking on the "more" links which take you to the news articles also works properly as well.
(you can get to raw mode by clicking "Force article view" and then "raw mode." I should probably expose that in other places)
NewsWaffle tries to determine the type of page. Articles get displayed with content run through readability, and then the HTML is stripped down. If its a "links" page, like the home or section page on a news site, it using HTML elements to try and find links to news stories vs navigational links to other parts of the site. Part of that is looking for links with longer text, since link text to news stories tend to be a few words. This helps sort "About Us" from "New Fusion Experiment a Success"). I'll check into why aldaily isn't working properly
Sorry I can't seem to reproduce the Techmeme issue. It works for me:
gemini://gemi.dev/cgi-bin/waffle.cgi/view?https%3A%2F%2Fwww.techmeme.com
Any suggestions?
For a GUI, I use Lagrange: https://github.com/skyjake/lagrange
Lagrange is sort of the Netscape of Gemini. It works on all the major desktop and mobile OSes. Personally prefer Elaho (iOS) or Buran (Android) for mobile
A request: In the linked NY Times front page, more formatting for the article list, maybe blank lines between articles. Visually, it's a challenge.