Hacker Newsnew | past | comments | ask | show | jobs | submit | debazel's commentslogin

My impression was that you would apply this filter after the logs have reach your log destination, so there should be no difference for your services unless you host your own log infra, in which case there might be issues on that side. At least that's how we do it with Datadog because ingestion is cheap but indexing and storing logs long term is the expensive part.

The example on their website is editable and it looks like they overlay the highlighted output on top of the textarea with `pointer-events: none` like you mentioned.

The code isn't minified so you can see how they do it by looking at the `doHighlight()` function here https://arborium.bearcove.eu/pkg/app.generated.js


Oh, you are right!

Hmm .. and the approach already shows its weaknesses when I play with it: When I search for something on the page, it gives me twice as many hits as there are. And jumps around two times to each hit when I use the "next" button.

I wonder if that is fixable.


There is a neat `inert` html attribute you can use to disable all interactions as well as hide the text from ctrl+f searches. (Sadly Safari is the weird one out, and does not exclude the content from searches.)

https://developer.mozilla.org/en-US/docs/Web/HTML/Reference/...


One simply needs the Highlight API. I held back, but now even Firefox ESR supports it.

https://developer.mozilla.org/en-US/docs/Web/API/Highlight

All the trickery vanishes and you get first-class CSS support.


And there's an open issue for that already: https://github.com/bearcove/arborium/issues/62

That works on the text inside a textarea? Is there a demo showcasing this somewhere?

GitHub had to solve the same problem when speeding up there code viewer.

https://github.blog/engineering/architecture-optimization/cr...


But adding caching to SSE is trivial compared to completely changing your transfer protocol, so why wouldn't you just do that instead?


This sounds awful, now you'll be reading some documentation or comment about llm-stream where they didn't mention the full namespace, so you have no idea which of the 50 different llm-stream tools they're talking about, and on top of that you can't even search for it online.


This paper just says that handwriting requires more cognitive load?

Which is exactly my experience with handwriting through my school years. When handwriting notes during lectures all focus goes to plotting down words, and it becomes impossible to actually focus on the meaning behind them.


The actual research doesn't back up your personal experience.


It's due to every hyperscalar building out new AI datacenters. For example you have Google recently saying things like "Google tells employees it must double capacity every 6 months to meet AI demand", and that they need to increase capacity by 1000x within 4-5 years.


Not OP, but personally I very much prefer Firefox font rendering on Windows. Text in Chromium based browser looks blurry to me, which causes eye strain. Firefox also has a much sharper and better looking image down-scaling algorithm that again looks blurry in Chromium based browsers.


Sure and that's why EU now has the weakest tech sector of any service industry and have become absolutely dependent on US and Chinese software instead.

I cannot even use my official government ID application that is mandatory almost everywhere without signing on to Google or Apple, so much for data privacy and sovereignty.


The tech sector is for people, and not people for the tech sector!

If they cannot do without exploiting people for their organizational benefits, then they can go to China and USA, they do not care about living standards over there! That is the place for ruthless, greedy, lazy fast paced groups to stomp on people! As they already do.

The EU is not the strongest and who cares?! It is strong enough, and that's what counts! Fullfills the needs. Good. Not everything is a race to infinite numbers in quarterly reports! Less inflated pushy tech idiots racing each other to oblivivion? GOOD!


This is pretty much untrue. Look at India, Africa, South America, Japan, Singapore, UK, Israel, the Arab world, Turkey, Russia, Ukraine, Norway, Switzerland, or Australia and compared to them the EU is doing just fine


You’re comparing the tech sector of the EU to that of Africa?


No


Nice edit


Bad troll!


> What works is content based advertising - so advertise a power drill on a woodworking hobbyist site. No tracking required there. Conversion can be obtained when user clicks a link via redirect. Like in the good ol times.

This still requires tracking to follow the user through the whole flow, which is required unless you want to be defrauded with fake users at the very least, but also very important to track the actual performance of each ad source.


Why do things that are important to the advertiser trump what's important to the user? I don't care how hard it is for you to track the performance of your ad sources, I just want you to stop tracking me.


Because without ads we're not profitable so there would be no service?

You can't just buy a domain, put your service out there, and expect it to gain traction. Advertising that you actually exist is essential for any service, but especially so for smaller businesses and startups.


I guess you have a point there.

I am trying to imagine a scenario where you just track the actual conversions (sales) and the only datapoint is where your customer originated from, something akin to podcasts/youtube giving affiliate links. That could work right? Or maybe I am missing something. If I am not it feels like the current model only benefits the middle man and is detrimental to everybody else.


Most are running ads and needs to track the performance of their ad spend I believe, at least that what we do. We don't care at all about tracking anything other than x amount of users came from x ad source with some basic device info like mobile/desktop/etc.

We tried to get rid of any tracking banners but have been unable to do so.


So can’t abuse people’s data without their consent is being strangled?

Is that like I’m strangled with my start up of “cheapdvds.com” because I can’t sell someone else’s data?


You have a funny definition of the word “abuse,” and “sell.”

“25% of our users that arrived from the newest ad came from Facebook and 85% of those were mobile users.”

So abusive. So much selling.


And when someone visits your website, you don't tell anyone about their visit, right?

RIGHT?!


That's an egregiously poor faith interpretation of what they said.


How do I stop you from tracking this information about me?


Do not consent when asked or, better yet, do not use websites that implement these techniques.


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

Search: