Hacker News new | past | comments | ask | show | jobs | submit login
Show HN: The Brutalist Report – A rolling snapshot of the day’s headlines (brutalist.report)
232 points by cylo on Feb 22, 2022 | hide | past | favorite | 87 comments
Hi HN. I was inspired by so many other folks also longing for a return to the old web that I put together a service to scratch my own itch: An extremely fast headline aggregator done in 1990s style HTML.

Sharing it with you all for those of you that also would enjoy this now esoteric style.




The Drudge Report's headlines look hilarious when rendered in a consistent, normal font size, in a plain unordered list of links. I'm sure the same is true of sensationalist outlets across the political spectrum, but this is the one that stood out to me.

I like it. It won't replace my RSS reader, but it's well done. One thought: could you link to articles on text.npr.org rather than the full-fat NPR site?


I like it, but it needs an option to be less. Most obvious way would be a simple filter so you could say "show me the last X minutes / hours" or just the top X headlines. It'd be nice if the layout was a bit more forgiving too - better line spacing, nicer font - nothing too much, just... hurts my eyes a bit...


Could there be a view where I see all the news from different outlets ("all", per topic) in one "thread" with reverse chronological order?

Also, for categories, may it be possible to categorize articles based on tags rather than the source outlet? What happens if, say, NYT publishes a business story?

Otherwise, this is really cool site. Bravo!


One takeaway from feedback I've seen is I'll be implementing support for filtering by keyword at the URL level.

I'm inclined to keep this completely javascript-free.

This should be able to get you close to what you're asking for once implemented.


Second this. Just having a massive chronological list with the source mentioned after the date is a lot more interesting to me.


Interesting project and appreciate the simple / brutalist layout.

A couple questions:

1) That's a lot of news! Just browsing the headlines would take a long time. Any ideas on how to make it more compact? I'd definitely be interested in a site that could give me 15-30 high quality links per day across a diverse spectrum of content and sources.

2) I personally appreciate the discussion aspect of HN and wouldn't enjoy just visiting the links alone. Any thoughts around including a link back to the original HN discussion?

Thanks for sharing!


1.) It is a lot! The frontpage in particular is almost overwhelming. Browsing by individual topics makes it a bit more digestible. One element I could support is a parameter you can supply in the URL to limit the amount of articles returned per source. The challenge with providing 15-30 high quality links is the "human editing" element required to do that effectively. I'd like to avoid that if at all possible, as well as not get into the business of harvesting visitors' personal information. There's no tracking or harvesting going on here whatsoever and I intend to keep it that way.

2.) Great idea, I could perhaps add an [hn] tag to those stories to preserve the link to the discussion.


I've implemented a 'limit' query parameter you can optionally supply to control how much content is returned back from search source.

Example: https://brutalist.report/?limit=10 or https://brutalist.report/topic/business?limit=10


I like the compact headlines using

https://brutalist.report/?limit=5


Love the no tracking mojo, but it would be great if I could be able to "collapse" (equivalent to HN [-]) a site (hide all that site's stories) and have that setting remembered when I come back.

Putting it another way, I'm fine with manually parsing the long list, as long as I can manually exclude whole sites from hitting my eyeballs and have that work remembered next time I visit from the same device.


Re 1): I agree. The source in small writing, then the top 3 stories only? With a drop down if you wanted more?


I built something similar a long time ago using deno and deploy.

https://github.com/searchableguy/burger

Although, after trying out different feeds and approaches. I must admit, I don't care about every item that is posted on a news site or someone's blog.

Anything interesting will get posted here.

So I figured I need a way to clean up HN stories that I don't find useful. I built an API server which does this using sentiment analysis.

Another problem I noticed is I tend to click on comments of stories I know won't be any useful (web3?) but it's hard to stop the urge so I check if comments are overly negative and return a score which is used to hide or show the see discussion button.

I also return all the urls from the discussion separately because those are usually good resources.

Meet cabbage news - https://github.com/searchableguy/cabbage_news

I haven't updated the repo so it is probably outdated and that's my first time writing a python application.

Oh here's the block list of keywords I recommend to everyone using HN

https://github.com/searchableguy/cabbage_news/blob/main/app/...


It would be great if you could expand the aggregator to non-US / non-Western sources - one place where we could compare how the same story is written from different sides, would be great. The Rashomon on news aggregators


"... done in 1990's style HTML."

Aside: I would have called it "original" style HTML as there was no www before the 1990s.1 IMO, calling it "early" style HTML would not be apropos because the internet, including the www, is still in its infancy.

1. According to Wikipedia there was SGML as early as 1986 but Berners-Lee did not write the the HTML specification until 1990.


sgml is the precursor to XML, with dsssl as the precursor of XSLT and DTD as the precursor to XML schema and relax-ng.

The first "final" sgml standard document was published in 1986, the initial draft was available in 1980.

http://www.sgmlsource.com/history/sgmlhist.htm

The first html spec was thus accompanied with a DTD.

CERN was heavily involved in the sgml standardization, thus it was natural for Tim Berners-Lee at cern to go the sgml route with html.


So basically a web-based RSS reader hardcoded to a few feeds?


Some tips to improve the readability:

- Give more line-height between each bullet point. - You use white for time and blue for unvisit links. The contrast for the time is stronger than the link, resulting in more reading effort. Change up the contrast to focus where you want the reader to look at easily


There's also Legible News, which is similar, but aggregates news sources from Wikipedia articles instead of just pulling different site feeds: https://legiblenews.com/


First impression. I like it. But some sources have 3 or 4 headlines and others have dozens. Perhaps the sources with few headlines could be on a row together (or the next source in a column moved up) to eliminate huge chunks of whitespace.

What if I want different sources, e.g. Babylon Bee instead of (or in addition to) The Onion? Would be great if this is doable without login, just give the custom feed its own URL. Yeah, it would become my own echo chamber but otherwise everyone is stuck with whatever echo chamber you choose.

OK this one isn't old web but it would be sweet if I could select a headline (or word/words within a headline) and see all articles matching that topic. e.g. volcano, or FCC.


A HN user posted a project of theirs that seems to do most of these things:

https://sumi.news/


This one is awesome!


Sure they say: > The day's headlines delivered to you without bullshit.

But there is a whole lot of •shit.


Fascinating. What did you do for scraping? I built something similar but to do sentiment analysis on news (it is shamefully slow https://maudlin.standingwater.io/ ). I used scrapy to get the articles and post them to a postgres database that flask reads.

I think I'm going to take some design queues from you and simplify my system. Word clouds are cool but too intensive on the $5 server I pay for.


The whole thing is written in Go on my end. Ingesting new headlines is handled in a goroutine that spawns within the process every 30 mins using a combo of the wonderful gofeed (https://github.com/mmcdole/gofeed) and colly (https://github.com/gocolly/colly) libraries.

When loading the front page, you're loading a 1-minute-cached HTML page of it that was constructed out of headlines already in my PostgreSQL database that were put there by the ingestion goroutine.

I like the idea of word clouds actually, I think you're on to something there. I think you just need to pre-generate them rather than doing it adhoc (if that's what you're doing here) for speed. Additionally, perhaps consider using sentiment in a way that orients stories based on positive and negative sentiment. Right now I am not seeing how I as a visitor/user can act on the sentiment analysis as it is presented now.

It would be neat to see a collection of uplifting stories grouped together through the sentiment analysis.

Anyway, food for thought. I hope you keep hacking away on it as it's just good fun to build things.


Great! Thanks for sharing! I needed something similar as a first stage for scraping buzzwords: https://ofia.info/

Furthermore https://brutalist.report/ is perfect "product" for me as I used to read headlines only ;)


Something else along these lines is newshound: https://newshound.co/ I use it as my daily basic newspaper.

I think it was also introduced on HN.

By the way, is “brutalist” a good description of this style? I don’t feel I have a great handle on how to use this term, but doesn’t it mean something other than simply simple?


I think the only thing that is different from what the generally accepted brutalist web design[1] tenets is not having the links underscored. Personally I understand this choice because if the majority of your content is largely made out of links the screen gets pretty crowded. So I think this passes as a brutalist page.

[1] https://brutalist-web.design/


> Brutalist buildings are characterised by minimalist constructions that showcase the bare building materials and structural elements over decorative design.[4][5]

This website certainly does seem to be minimalist, showcase its structural elements, and eschew decorative design.


I see the minimalism, and the lack of decoration, but not the showcasing of structural elements. I guess, for a website, this would have to be something like actually exposing some of the markup or other infrastructure. Otherwise it’s just minimalist, without the brutalist element. See what I’m getting at?


Meta: I thought HN submissions get either a link or explanatory text, but this item has both. Is this something a mod does?


I'm a fan of no-frills layouts and I do miss the web of old, but this could be easier to read than it is. I like

  body {
    font-family: monospace;
  }

  li {
    color: grey;
    padding-bottom: 0.5rem;
  }

  li a {
    color: blue;
  }


This is pretty great. Thank you.

I am newshound and I do keep several subscriptions.

I was thinking just the other day how I wished I could click on a link to get a list of "all stories in todays paper".

Just a plain list of links that can be read fast and leave more time for reading.

So this fits in with it.


This is great! May I suggest some small improvements for scannability?

ul {line-height:1.3;} ul li {padding-left:1ch;text-indent:-1ch} ul li + li {margin-top:.2em;} a:hover, a:focus {text-decoration:revert;}

Also, using <ol> lists instead of <ul> could help too.


How about a view that weights the time headlines have spent on that page per day? In other words, as I understand the page, it aggregates the current headlines, but I'd like a view that shows what that page has shown to be the news that day (or for the interval you choose as appropriate).

That may be something different than what you are trying to create though!

I was thinking of it as maybe three "brutalist" views (in descending level of abstraction): 1) what the news has been; 2) what the news is trending to; and 3) what each site is currently reporting.


I like this, thank you.

Could you make an "api" (maybe just a csv or static json file) available? Alternatively, is it ok if I scrape your page occasionally?


Thanks. Absolutely, please feel free to scrape. Though that's always a brittle solution. I've added a JSON API to my todo list for it.


I've now implemented a JSON API for you to use at /api.

Example to get the tech topic via API: https://brutalist.report/api/tech

Or front page: https://brutalist.report/api/

Enjoy!


If you run out of things to do...

https://blog.sequin.io/events-not-webhooks

PS: Well done!


Thank you!


This [1] is my attempt at tech news but with tag cloud. Feedback most welcome.

[1] https://embit.ca


This reminds me of the popular-in-the-early-Aughts RSS-based site Freshnews. Which still exists:

https://www.freshnews.org/

Pulls from 33 sites, including HN, leading with Slashdot, CNET, Pinboard, Reddit, Gizmodo, Linux Today, Github trending, Engadget, and Yahoo technology news. User-customisable if you want AFAIR.


There was something similar posted here a couple (few?) years ago, but I think it was curated by hand - can't find it now.

Also worth mentioning https://text.npr.org/ and https://lite.cnn.com/en


Maybe you're referring to http://readspike.com? I don't think it's curated, but also aggregates links.


Nope. It was much more lo-fi. Basically just a plain-HTML list of headlines/URLs.


It immediately reminded me of http://68k.news/

Warning about HTTP; no HTTPS if you're concerned about that sort of thing.


Very useful. What's your monetization plan? I'd rather pay a small fee than see any ads!


No monetization plan. Nor is there the temptation to. This isn't really something I built to make money. As mentioned, I built it mostly for myself. Happy to share it with others along the way. :)

It's interesting to see how polarizing it is to some folks.


Done the similair project in the past, main discovery was that without breaking news headlines, images, font size, color flickering, all the senzationalism was gone from the portal.

Maybe i will next do the opposite, in plain text insert senzationalism, but that's called marketing.


Ha! Thanks for including ElReg :)


Looks to be a common itch to scratch. Alongside some others that have been shared, here's one I bookmarked a while a go that does similar:

https://skimfeed.com/


I love it when the younguns use a tiny font and then when I enlarge them, the forced column sizing cuts off the text.


Had a quick look on Github for this, but didn't find any repo. Is it open-source? I would love to contribute to and self host this. Specifying my own news sources and layout on my own service would be ideal for me :)


Looks really similar to https://sumi.news/.

I wonder why these aggregators don't do a weekly version. I don't want to read the news every day.


How do you decide what to include in a weekly digest?


Great work. Will bookmark. I built something similarly no-nonsense and text-based for pro wrestling news: https://www.wrestle.buzz.


This is nice.

Not criticism, more a confession. My eyes aren't what they used to be. They somewhat struggle to focus if there is much white screen appearing with colored fonts if the font isn't sufficiently dark. My tired eyes want subdued effects now. As I said, lovely site. It's my aging eyes that are the bother.


I wrote an RSS app to aggregate headlines: https://github.com/Dotnaught/rssputin


Great name


I appreciate look and approach ( and not only because I toyed with a similar idea but failed due to editorial work I thought it would need ).

I think this is what people need. I favorited. Keep it up!


Very nice!

Now it needs an algorithm to detect related stories across different sites so they can have their own “popular stories” section or ignored after you have browsed them.


Really cool idea. It would be nice to be able to roughly sort by something like importance. Easier said than done in an automated fashion, I know.


This is a boon for blind and low vision people.


Politics section is very US-centric and the timestamp defaults to PT.

Would be cool to make it more global in coverage and presentation.


Like it - reminds me of a cleaner popurls.com. As someone else suggested, would love to custom add sites (via url).


This is amazing. I really hope you don't cave to the lamers asking for 'features'!

If you do, I'll re-release your original.

Cheers!


Can there be a little bit more space between the list items? Hard to read when they are too close to each other.


Thanks, added as a bookmark

I do like the basic style, not a massive fan of the color, needs better contrast for the text - imho


Very well done, I like this quite a bit


really cool site, using it over HN top now for the general side of things, is this going to be open sourced or self hostable to allow people to customize it? Wouldn't mind tossing in a few qol life fixes myself if it was hosted somewhere like sourcehut or github?


It reminds me of http://hackurls.com

Well done


So, is this mostly using RSS/Atom feeds from the various websites in question?


I like this a lot. I set it to open every morning at 8am using an extension :)


This is pretty sweet. What tools did you use to build it?


This reminds me of the right-wing news site https://www.drudgereport.com/ (not a news site I recommend at all, it just reminds me of what it looks like).


No longer right wing. Drudge sold it. Left leaning now.


Also see: text.npr.org and lite.cnn.com


Nice this is pretty decent


Great!


idk when brutalist became synonim of ugly and minimum effort, but I assure you it's not.


Brutalism means making the form support the function, and being true to the materials you're using. I'd say this qualifies.


Absolutely not, besides, even assuming that brutalism also include function over form, the site isn't true to materials, as links and fonts are obviously fancied up, and the page structure is absolutely hidden in a sea of white. Moreover, the structure is slaved to the content and not the other way around.

lol who donvoted this? you don't need more than two minutes of image search to utterly demolish the notion that brutalism is function over form.


Ugly? I don’t agree. I think it’s a clean and fresh layout. I don’t think it’s incorrect to call it brutalist.


> fresh [...] brutalist

Lol


I’d say, to the average person the Brutalist aesthetic is certainly synonymous with ugly minimalism.


Maybe the backend was written in concrete.


Yes, and brutalist refers to French word for raw concrete, Brut, not in relation to the adjective Brutal.


I agree. Just look through some of the listed sites on brutalistwebsites.com[0] and you'll quickly realize most of these designs are NOT easy to pull off

[0] https://brutalistwebsites.com/




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: