HN user
Building adblock for live TV that auto mutes and changes channels around
www.relaxoplayer.com/#android
Do people even want their culture democratized with just anyone being able to produce high entertainment? The recent popularity of "Harry Potter by Balenciaga (2026)" AI fashion parody retelling shows we might actually be stuck in this cultural rut forever with or without AI help.
Insidious perhaps. There is no middle ground of neutral advertising to be had. It's all just pressing upon the pain points of one social sect or another, undermining mental acuity to make 'em feel inadequate or falsely adequate.
I run an adblock app as well so it's hard to begrudge anyone trying to gain undeserved significance for their thing - when some invention is new who is to say what is deserved unless the right people know about it?
Whilst you might be able to block the ads for now you can't block the downslide of polite society that largely accepts the techniques into their lives and keeps them in their back pocket.
Can you point me to that chromecast tool or tell how it worked?
I'm in the process of releasing a video player with adblocker for live IPTV on android that can definitely auto-mute the whole adbreak for you if you watch the stream in the player (see my previous comment for link).
But if you are suggesting a system-wide mute triggered by a dormant app as another live broadcast plays in some official app then that's easier in some respects and harder in others because people need to trust the data is accurate. Once we have enough users sharing accurate adbreak data (powered by accurate software with human corrections) we can offer a system-wide mute for ads in any other app playing those known broadcasts.
We've built an android video player that lets you adblock live IPTV [1]. It does exactly as you describe and more. Just endlessly avoids ads going from A->B->C->A and helps you update what plays instead of ads with ease. A passive hardware solution is certainly the endgame, we got enamored with adblock at the codec level as lets us add all the extra worthwhile UI features that should be inside of an innovative linear TV player.
[1] https://relaxoplayer.com - join waitlist for link to beta apk.
If it's a social app users should worry about account take over making you look bad/illegal or tricking you to enter your password to other OAUTH accounts. Privacy implications etc. Similar to if the app owner changed hands to someone trying to milk it. As always you be suspicious of any permission asks to limit damage in these cases.
Github reserve the right to stop serving those release downloads at any time. They usually just kick you off entirely if your project gets unwanted attention. I don't see them allowing revanced (modded popular social apps) forever so we still need a better way to trust outside that touch and go easy relationship.
Finally, a no nonsense Auto-App-Updater App! if only sites would include a version number somewhere on the download page so obtainium could find it. Looking at you https://grayjay.app (it doesn't seem to work for partial file hash either so I had to turn auto updates off for this one)
We sorely need 1:1 replacement of app store trust and discovery mechanisms too without any kafka-esque approval hoops. Obtainium app config sharing and perhaps a standard for APK release webpages would be a great first step towards that.
The choice of intonement even mimics creatives which I'm sure they'll love. The vocal fry, talking through a forced smile, bumbling host is so typical. Only, no one minds demanding better from a robot so it's even more excruciating fluff with no possible parasocial angle.
Limiting choice to frivolous voices is really testing the waters for how people will respond to fully acted voice gen from them, they want that trust from the creative guild first. But for users who run into this rigid stuff it's going to be like fake generated grandma pics in your google recipe modals.
for dev there is already pg-lite server (postgres as nodejs/wasm) you can spin up, it's self-contained single-folder db to disk, 100MB RAM thingy. Highly experimental. Not sure if supabase codebase is setup for debugging the way you suggest.
For prisma/nodejs devs who just want postgres-in-a-can for local dev you are better off using the recently released serverizing of pglite, pglite-server: https://github.com/kamilogorek/pglite-server
It's faster, can persist data to fs, though less stable under heavy use than the full x86 emu e2e test server. I found pglite-server uses only 150MB ram compared to 830MB for pgmock-server. You can then use dotenv to checkout a new .env.local with updated DATABASE_URL for all your nextjs/prisma package.json run scripts
DATABASE_URL="postgresql://postgres@localhost:5432/awesomeproject"
"db:pushlocal": "dotenv -e .env.local -- pnpm prisma db push"
eye on ai had Yann LeCun on the other day and has cool ideas about present and future that the less expert host is doing great to get out of him. https://aneyeonai.libsyn.com/150-yann-lecun-on-world-models-...
the two aussie brothers who try to deep dive the latest news/thing they can test/banter about in a fun producty way. https://www.thisdayinai.com/
found via https://allainews.com/podcasts/
Many broadcasters are positioning to accepted the broadcast part of their business as the freemium entry point into their online app. The ads will never be turned off everywhere but the better experience will only be found over the jump.
I like the interface but to be truly private it'd have to be offline first like https://everywhen.me. And I don't know if I'd ever trust a score I hadn't reasoned myself. I already do a version of this style of rubberducking with LLMs and the idea score you naturally give something changes with each unexpected key question the LLM spits out. Problem is you don't know what key opens your idea or closes it and it needs to be worded just right.
Like "who is angered most if they could no longer get your product" and "under what context is the value your product provides most apparent to your target audience".
Pretty big jump for java eval, what is the reason for java being so notoriously difficult for LLMs? never mind I asked phind[1] and it said all the complexity... but do you have any tips or tricks for working with that language in your model?
[1] https://www.phind.com/search?cache=u3mnj3iwmjvgqlyf60bnbqo1
Somewhere in a book Plato himself laments that texts are not interactive enough, when one interacts with them one has the desire for advice on action. So if essays aren't cutting it in schools and books aren't the pinnacle of learning then we should respectfully move on.
I disagree that experts will be the starting point for the next "medium" after books, they still have too much cruft to blast away if you are going to benefit anywhere near as much as they did from the concepts. The new knowledge will be much more synthetic and symbiotic.
There is no explanation of need here. Broadcasters want the security feature regardless of effectiveness and consumers will be forced to "upgrade" to devices with hostile features just to receive public airway broadcasts as they do now. No one welcomes more targeted ads with mute disabled. The standard should be shunned on general computing grounds.
If it came down to it would you like the ability to automatically blackover the ad while it plays and mute your tab as a last ditch adblocker? Because that's an easy enough final solution to the ad problem but not everyone would want to endure the dead-air or plan for the time sink waiting for the real video to start.
stay active, while injured
use only your good leg
The app helps users work around the injury but I'd argue that it's generally wrong to seek ease or "Stay off it", "not put any weight on it", "limit the angles" as it risks a spiral of muscle atrophy and imbalance just like knee wraps do.
Of course the safety alignment of chatbots perfectly fit with the app's goals, they will gladly never advise you any possible immediate risk even if that means prolonging recovery times by a lot.
No one asked but my favorite introductory bodyweight exercises for meniscus tear recovery are inspired by @kneesovertoesguy: simulate SLED PUSHES by holding the roman rings in a prone position, BACKWARDS WALKING for a few tiny backward steps holding roman rings upright for resistance, and lots of short single-legged dips with foot on a wedge/decline. Perfect control of weight is possible at all angles.
Do not ask the user follow-up questions.
kagi seem to prefer the small legacy web to limp on forever rather than chatbots take off, this GPT experiment seems like a box ticking effort since a serious search engine has to have one not because they believe in the next wave of technology.
Only stick to the facts from the provided context.
the things chatbots are best at like analogies and creative output are cut down keeping in alignment with above.
now those of us with up-the-nose cams on our laptops can install an oversize thinkpad nipple there.
Small group messenger. Compared with session/signal/status you miss out on large and public groups.
The transmit to every node with fake data all the time approach quickly becomes unsustainable for large groups of >100 people so even if you had a public user ID it wouldn't be seen by that many people anyway. But if they solve this feature disparity without IDs or bloat it will be very interesting.
A versatile bunch of ideas better presented in the announcement [1]
I like the idea that simple ram storage in the program can be used as your complete durable database (since everything is durable) negating the need for even serverless databases like neon.
All I want to know is if their ram usage & pricing is competitive, considering databases get big and are already perfect at taking cold data out of ram back to disk.
Everything else is gravy: you can tickle every function in your multi-tenant actor system from a http birds eye view.
The paradigm they are going for seems less like scale-to-zero more like sure-wake or never-fail or infini-migrate workers and that bodes well for being able to be charged according to the exact amount of active ram that your worker uses moment to moment rather than per invocation on an inflexible amount of ram.
Ask the Spartans how well “winning” on quality goes in the long term.
Also there is the AI issue as all the established social networks lock down data and make it possible to flood potential rivals with toxic interactions.
The Discoverers: A History of Man’s Search to Know His World and Himself - Daniel J. Boorstin
How Innovation Works: And Why It Flourishes in Freedom - Matt Ridley
The Technology Trap: Capital, Labor, and Power in the Age of Automation - Carl Benedikt Frey
https://open.spotify.com/show/1jRACH7L8EQCYKc5uW7aPk
I'm listening in preview mode, ben says he prefers open standards for the podcast but it appears paywalled and personalized on top of an open standard.
Look at Hollywood, they spawned photo-realistic CGI and figured out their captive audience will settle for much lesser quality anyway so why bother.
Bell labs thought something similar about the first answering machine, fearing people would be held to their word resulting in fewer telephone calls.
Now big tech worries that "perfect" trust in their communication platform goes away, but the paradigm shift happens no matter what. Many governments already have the power to post as users on platforms and that should be cause to usher in this new trust calculation faster.
This just rewards bad behavior with my consideration which will only beget more of the same kind of trickery elsewhere in the videos.
Personally I BlockTube anyone I see using a soyface/arrow because it is indicative of what kind of uncritical audience they want. I can easily wait a year and get a summary of all those videos in 1 minute total from a respectable intellect. There is no FOMO.
better formatted original online link
we can't serve adults steak because babies can't chew it.