Pardon the mess, this turned out to be quite a lot of text.

No vector DB, no paid API, no full articles. Plain Python pulling free RSS feeds, plus Ollama running on my own 2070 Super. RSS already gives you the headline and a short blurb, so there are never megabytes to sift through. Every Ollama call is one story at a time, so the context window isn’t an issue. Llama3.1:8b for summaries and scoring. Gemma4:e4b for clickbate work, nomic-embed-text for headline duplicate vectors.

Find duplicates:

Python: strip each headline down to keywords (drop words like “the”, “says”, “update”) and only compare headlines that share an uncommon word. Two headlines match if they share enough keywords.

Python: sends batches of 128 headlines to Ollama

Ollama: turns each headline into a vector using nomic-embed-text

Python: compares the vectors, and anything over the config threshold counts as the same story. The vectors just live in memory for the run, no database needed. If Ollama is down, the keyword match still works on its own.

Regen short summaries:

Ollama: Generate a 2 sentence summary from the feed text if available

Get rid of clickbait:

Ollama: Detect clickbait from the headline only. Classify this news headline. “informative”: it states the news itself (who did what, what happened, what was found). Dramatic or emotional wording is fine if the fact is stated. “teaser”: it withholds the key fact to make you click: e.g. “The Surprising Reason…”, “The Class That…”, “Why X…”, “How X became…”, “What we know about…”, “Is X about to…?”, “Then things got weird.”

Unwrap clickbait and rewrite the headline if possible, otherwise drop the story

Ollama: A clickbait headline hides a key fact. Question it leaves open: {withheld}

Does the text below explicitly answer that exact question? Mentioning the topi talking around it, or answering a different question does not count.

“answer”: the answer quoted or closely paraphrased from the text, or “” if the text doesn’t give it. “headline”: if answered, a plain, factual replacement headline (at most 90 cha states that answer, using only facts stated in the text. Keep the text’s certainty: …

Respond only with JSON: {“answer”: “”, “headline”: “”}

Headline: {title} Text: {text}

Score each headline

Ollama: return a summary and an importance score in the same call:

SCALE = ("1-3 niche, local or trivial; 4-6 notable; 7-8 major national or inte “9-10 historic events or emergencies affecting millions right now.”)

Python: rank = importance + a bonus for more outlets carrying it - points for age

Don’t put US articles in World unless they have non-US countries involved

Ollama: Which countries is this news story about? List the country where it takes place and every other country that is a real party to it (acting, targeted, negotiating, fighting, trading, sanctioned). Use common English country names, e.g. “United States”, “United Kingdom”. Ignore countries only mentioned in passing or as background. Always name at least one country: if the text doesn’t say wh give the country where it most likely takes place.

Respond only with JSON: {“countries”: [“…”]}

Headline: {title} Text: {text}

Python: if the only country is the US, it’s left out of World

Everything gets cached per story, so each one only goes through Ollama once.

Ollama container, 2070 super, 3-7b, meta and qwen

I use it to ingest tons of news rss, find dupes, summarize, and then post that output to mesh a mesh network.

I also use it for a self hosted mealie to figure out recipe->ingredient.

I also use it for receipts. I translate pdf -> txt, feed the text to qwen, ask it who is the vendor, what is the price, what it a reasonable memo for the receipt, and give a dozen categories, which is best.

Ohh fear not! Hardware didn’t plateau. Our current architectural needs plateaued. We’re still doing everything on 2d layers layers in a way that makes things organized. I’m sure they’ll shove AI in there and the next gen stuff will be unsupportable chaotic 3D tetris stuff and the memory/bus sizes will be old hat.

I would be so embarrased to send links to my config. So many hacks, so many poorly implemented kludges. Yes I inlined home-manager, i didn’t feel like updating it seperately :(

Azure is like the SharePoint of cloud services. The controls are horrible. They keep making breaking changes to how they do crap; the whole thing is unintuitive enough that you really need someone trained to use it.

I stood it up beside AWS and used them both for what they were cheaper for; I eventually gave in and migrated the last few images I had over there to AWS.

Azure works, but it’s not good. IMO, the only reason it exists is that they have good Datacenter licensing fees and people using Entre to get away from having to host their own DCs.

It’s more accurate to say somebody spent a > couple thousand dollars and a few weeks project-managing an AI.

It’s not nothing, but it’s also not coding.

Basted on the east coast parks, the solar probably brings in more money than the park :)

That sounds perfectly reasonable.

https://imgur.com/gallery/amazing-Ngmk0VN it predates AI

It does, But at the same time the ruined finish and bent up hood scan with it being left to rot.

Normally you’ll get a belly on the bottom of the flat.

Either the front plate is cracked and peeling or the numbers are AI.

There’s a lot in this picture that’s sus. But it’s also perfectly reasonable for a tree to be topped and have the sun at high noon.

I tried slackware when win 3.1 was hot. I would have went a lot further a lot sooner if I could have gotten my modem working

There’s a bunch of strange things. The logo on the VW up close looks strange, too many lines, too thin.

The Figo’s that have trunks seem to put their logos on the left.

That VW has some strange gaps around the hood.

None of it is a sure thing, but enough to raise questions.

3,4 are the default in KDE/Wayland I think

Now, now, are we talking windows 11 or XP?

awesome, I just setup traefik / le on an old sort domain of mine and make everything inside support https.

Honestly, I was hoping to add client only certs on my jellyfin and open it to friends an family, but none of their streaming devices support it.

Just don’t get too close to TV’s or streaming boxes. Wildly different levels of hell

Linux almost needs continuing education courses. New shit comes out, and if you don’t try it in the next 5-7 years, you’ll end up getting it anyway.

50:50 unsolder and transplant all the chips onto a fresh nvme pcb or try to solder jumpers to the traces/pads to the chip long enough to get a data export.