Architeuthis

Here's a quick and dirty vanilla js script that highlights all posts in a thread according to how recent they are, the brighter the newer, and alse separately highlights new posts, to make long running threads easier to follow. I'm posting it in the stubsack because it's the thread I had in mind when writing it.

Pasting it in the browser's console and pressing enter should be enough for the page you have open, not that I've cross tested it any... Worst case scenario it does nothing or it colors the posts wrong and you just reload the page, I swear it won't steal your crypto, or mine any new.

In Firefox you can find the console by pressing F12 and selecting the console tab.

edit: Also if you prepend javascript: to the code and store it as a bookmark you can just invoke it by calling the bookmark, like a macro, see https://awful.systems/comment/4173451

Note: longer threads don't load all comments at once, so you'll have to rerun the script if you scroll down far enough.

edit: fixed for Edge, because why wouldn't it show dates differently there.

edit: updated it to check if there's a (xx New) notice in the post count in the OP and use the number to highlight the latest xx posts, i.e. all post made since the last time you were here. Change the value of variable newPostColor if you don't like the lovely shade of lavender I picked. Depending on if edited posts are counted as new or not the count might be off, and like, what if there's a new post that's also been edited? Solving that seems to mean moving away from the warmth and comfort of the quick and dirty territory, and also is there a public philthy repository somewhere?

edit: here's how it looks in the SAP thread:

edit: NEW: added some legibility changes and also consecutive executions now toggle old post highlights.

Code now in spoiler:

(() => {
    function getHighlightedColor(min, max, value) {
        const percentage = (value - min) / (max - min);
        return `rgba(0,0,255,${percentage})`
    }

    const newPostCount = (() => {
        const text = document.querySelector("span.fst-italic").textContent;
        return text.includes("New") ? parseInt(text.match(/\d+/)[0]) : 0;
    })();
    const newPostColor = "#783AFF";

    const timestampNodes = [...document.querySelectorAll("span.moment-time")]
        .map(x => {
            return {
                Node: x,
                PostBox: x.closest('.ms-2'),
                Date: Date.parse(
                    x.dataset.tippyContent
                        .split('\n').at(-1)
                        .replace(/Modified |at /g, "")
                        .replace(/(?<=\d+)(st|nd|rd|th)/g, ""))
            };
        })
        .filter(x => x.PostBox != null)
        .sort((x1, x2) => x2.Date - x1.Date);

    const minDate = timestampNodes.at(-1).Date;
    const maxDate = timestampNodes.at(0).Date;
    const hl = (dt) => getHighlightedColor(minDate, maxDate, dt);

    timestampNodes
        .forEach((x, i) => {
            if (i < newPostCount) {
                x.PostBox.style.backgroundColor = newPostColor;
                x.PostBox.querySelector('.person-listing').style.textShadow = '1px 1px 0.75px #FFFFFF';
                x.PostBox.querySelector('.comment-content').style.paddingLeft = ".5em";
            }
            else if (x.PostBox.style.backgroundColor == "") {
                x.PostBox.style.backgroundColor = hl(x.Date);
                x.PostBox.querySelector('.comment-content').style.paddingLeft = ".5em";
            } else {
                x.PostBox.style.backgroundColor = "";
                x.PostBox.querySelector('.comment-content').style.paddingLeft = "";
            }
        });
})()

The whole point of using these things (besides helping summon the Acausal Robot God) is for non-technical people to get immediate results without doing any of the hard stuff, such as, I don't know, personally maintaining and optimizing an LLM server on their llinux gaming(!) rig. And that's before you realize how slow inference gets as the context window fills up or how complicated summarizing stuff gets past a threshold of length, and so on and so forth.

The engineers can generally also do other things

What's the job posting for that going to look like, LLM stack maintainer wanted, must also be accomplished front end developer in case things get slow?

That just sounds incredibly British to me.

When I clicked the video I fully expected Rocko with a c to mean Rosko.

I'm not a native english speaker, how else was he supposed to pronounce it?

Conceded.

If he becomes president he's selling off everything that isn't bolted down, isn't he? The US's own Boris Yeltsin.

I see, thank you.

(update: disproven by Crowdstrike’s blog post).

How do you mean? The current top post on the blog seems to mention .sys files as part of the problem very prominently.

Channel file "C-00000291*.sys" with timestamp of 0527 UTC or later is the reverted (good) version. Channel file "C-00000291*.sys" with timestamp of 0409 UTC is the problematic version.

Basically

Companies probably actually need to curate down their documents so that simpler thinks work, then it doesn’t cost ever increasing infrastructure to overcome the problems that previous investment actually literally caused

Definitely, but the current narrative is that you don't need to do any of that, as long as you add three spoonfulls of AI into the mix you'll be as good as.

Then you find out what you actually signed up for is to do all the manual preparation of building an on-premise search engine to query unstructured data, and you still might end up with a tool that's only slightly better than trying to grep a bunch of pdfs at the same time.

I mean, that was definitely a thing when I was at school, only it was mostly about teaching undergrads graph search algorithms and the least math possible in order to understand backpropagation.

As an aside, weird that we don't hear much about genetic algorithms anymore, but it's probably just me.