Ask HN: Is there a way to filter just tech content?

3 points by aliqot ↗ HN
Not interested in periphery and social issues. Is there a plugin or some tagging feature I'm overlooking?

7 comments

[ 1.1 ms ] story [ 31.6 ms ] thread
I do not know how to only display tech issues specifically, but someone recently posted uBlock addon filters [1] that can remove specific topics by keyword. I've modified mine a little. Adjust the keywords as needed.

After installing uBlock, go to add-ons -> uBlock Preferences -> My Filters

    # top (title / url)
    news.ycombinator.com##tr.athing span.titleline > a:has-text(/(Musk|Twitter|Mastodon|blockchain)/):upward(tr)
    # bottom (stats / comments)
    news.ycombinator.com##tr.athing span.titleline > a:has-text(/(Musk|Twitter|Mastodon|blockchain)/):upward(tr) + *
[1] - https://addons.mozilla.org/en-US/firefox/addon/ublock-origin...
Thanks! This should do the trick.
I made this a while back https://ontology2.com/essays/ClassifyingHackerNewsArticles/

I've repurposed the same code for a few different projects such as a job search (e.g. I got the second job I applied for when I told them abut it)

I am thinking about getting it out of mothballs and making it reusable and more sustainable and also adding some kind of clustering engine. What I can't stand right now is the headline "Astronomers discover the closest black hole" from the Earth that has been heavily me-tooed for the last month to the point where I'm wondering if (1) the astronomers discovered another one or (2) if some people only read the news once a month.

Wow you put a lot of work into this. What motivated you? It's interesting that the submitter based model wasn't more successful. You'd think people would center more around niches.
I have a hard time ignoring things so I want good filters. More of my motivations are written up here

https://ontology2.com/essays/HackerNewsForHackers/

but a lot of that turned out to be factually wrong. For instance that "Apple overload" was pretty temporary because Apple had made a wave of product announcements. Today people might claim about "Twitter overload" but it will be something else next month.

As for submitters I think that for what I am measuring the people who submit a lot of things and have high karma tend to be pretty average in what they submit since they've internalized some model of what does well on HN. I bet there are people with a very different profile but they would have fewer submissions so the model wouldn't be able to make strong conclusions... That logit model I used is well-calibrated and not prone to jump to conclusions if it has just a little bit of data.

With the right model and tools you might store something out. Overall I think my submissions are pretty average but I have a list of web sites that I scan and every day you'll notice I post something from The Cornell Chronicle, World Nuclear News, phys.org, arstechnica, arxiv, etc... Most superposters have an efficient strategy for posting a diversified selection of articles that have broad appeal.