dingdingdang

Use the Brave browser and look at the inbuilt filtering (search for "Content Filters" in settings), it allows explicit removal of shorts via enabling of "YouTube Anti-Shorts" filter list. Does the job beautifully.

... but we will know from this link, thanks!

edit: to think that such short-form drivel is locked behind a paywall is just sad.

... we'll never know from this paywalled article

Basic-to-great rationality or skill may not be what is being rewarded here (although the baseline of course needs to be met) - it could well be compliance capability. Hence the string of arbitrary memorization exercises.

I don't know, does Cursor offer anything substantial beyond an extension like kilocode? (I've only used vanilla VSCodium branch with various extensions but they all seem to integrate everything from tab-completion to complete UI agentic take-over very well)

Once the UI soup around AI dev use has settled (and it's getting closer) I bet you we will see native apps with c/c++/zig/rust backends that render so much faster on all the junctions that aren't roundtrip limited (and yes, that will still matter to many people).

1. There was no point, having thought about it a bit; a lot of the patches (in essence it was at most a handful) revolved around disabling features which in turn could never have been upstreamed. An example was, as mentioned elsewhere in this comment section, the abysmal performance of the thumbnail gen feature, it never cached right, it never worked right and even when it did it would absolutely kill listings of larger folders of media - this was basically hacked out and partially replaced with much simpler gen on images alone, suddenly the file manager worked again for clients.

2. Guess that's debatable, or maybe even skill dependent (mea culpa), and also largely a question of how comfortable one is with systems that cannot be reasoned about cleanly (similar to TFA I just could not stand the bloat, it made me feel more than mildly unwell working with it). Eventually it was GDPR reqs that drove us towards the big G across multiple domains.

On another note it strikes me how the attempts at re-gen'ing folder listings online really is Sisyphus work, there should be a clean way to enfold multiuser/access-tokens into the filesystems of phones/PCs/etc. The closest pseudo example at the moment I guess is classic Google Drive but of course it would need gating and security on the OS side of things that works to a standard across multiple ecosystems (Apple, MS, Android, iPhone, Linux etc.) ... yeeeeah, better keep polishing that HTML ball of spaghetti I guess ;)

Having at some point maintained a soft fork / patch-set for Nextcloud.. yes, there is so much performance left on the table. With a few basic patches the file manager, for example, sped up by magnitudes in terms of render speed.

The issue remains that the core itself feels like layers upon layers of encrusted code that instead of being fixed have just had another layer added ... "something fundamental wrong? Just add Redis as a dependency. Does it help? Unsure. Let's add something else. Don't like having the config in a db? Let's move some of it to ini files (or vice versa)..etc..etc." it feels like that's the cycle and it ain't pretty and I don't trust the result at all. Eventually abandoned the project.

Edit: at some point I reckon some part of the ecosystem recognised some of these issues and hence Owncloud remade a large part of the fundamentals in Golang. It remains unknown to me whether this sorted things or not. All of these projects feel like they suffer badly from "overbuild".

Edit-edit: another layer to add to the mix is that the "overbuild" situation is probably largely what allows the hosting economy around these open source solutions to thrive since Nextcloud and co. are so over-engineered and badly documented that they -require- a dedicated sys-admin team to run well.

Wonder how these play out against the https://github.com/X11Libre/xserver base, would be interesting to hear from that end as to how these things are handled. My understanding is that they address any sec issues that arise on x.org but it would be fascinating if the issues are already mitigated since XLibre updated their xserver port with 1000s of issues that were never addressed on the x.org side of things.

I agree, the ghost of Stalin is right behind the curtain waiting for a 2nd, 3rd, 4th chance at making it work ("oh, no but this time we will REALLY get it right and eliminate only all the REAL enemies of the people")

Yeah, I remember having a friend over and we would take time playing Blood Money on Amiga 500 and then the C64 which fam still had at that point, the Amiga was so much better but the difficulty on both platforms was mentally deranged (as game reviews of that era also indicated)

Think this looks beautiful, any idea as to the performance penalty vs say the new wg.Go(func({doSomething()}) sugared syntax in Go 1.25?

I'm mildly confused as to the value over, say RustDesk. The latter allows remote control of external machines and has ip hole punching .. no hardware involved! Any takes here?

Exactly this - vaccinate even a small town and cardiovascular issues will start popping amongst the neighbors, there is very good reason why there has been, and is, a public backlash on the MRNA vaccine usage among those who have eyes to see and ears to hear.

Blood Money was a good game - way too hard for me as a child but it had an atmosphere and soundtrack that gave it it's own following at the time and place in my memory banks as an extremely well crafted game. Thanks to OP for the article, appreciated!

I think someone is missing a massive marketing/product-improvement opportunity by not selling a scaffold for the bike to allow it to be used as an exercise bike - this is turn would make significant savings on people's electricity budget .. then suddenly the direct-to-battery pedal arrangement would go from weakness to proper feature! #feelfreetocontactmewithmonetaryreward

Additionally it would be excellent to have a "community support estimate" score so that one might factor in how much of a support bus-factor is at play.

I started out with WinGet but found the coverage lacking and ended up sideloading stuff which then soured the whole thing for me.. Choco basically has better coverage in my experience.

>> to perform up to Apple's typical standards of polish and control.

>i no longer believe they have kept on to the standards in general.

One 100% agree with this, if I compare AI's ability to speed up the baseline for me in terms of programming Golang (hard/tricky tasks clearly still require human input - watch out for I/O ops) with Apple's lack of ability to integrate it in even the simplest of ways.. things are just peculiar on the Apple front. Bit similar to how MS seems to be gradually loosing the ability to produce a version of Windows that people want to run due to organisational infighting.

The only, as per personal exp, way to properly setup a Win11 machine these days is strictly via LTSC release and WinUtil from Chris Titus + local account - clean and no "feature" updates for almost 10 yrs. Add to that Chocolatey for package management that takes away the "every piece of software adds it's own autoupdater crap"-pain and the OS is suddenly very capable for almost all computing needs.

Even if you do ALL the techinical work you can still find yourself banned/ignored as I learned years ago the hard way.. even big providers outside MS/Goog duopoly finds themselves partially unable to deliver business emails at times.. fun times for a small shop (not).

Also better and faster though!! It's close to a Daft Punk type situation.

The legendary status is also enhanced by the absolute nightmare that pointers enable if used with indiscretion or high level proficiency - a triple pointer is a good example for me but there's many many more, and arguably worse, examples out there.

This is literally amazing research, just because we are getting closer to rejuvenating tissue does not imply that suddenly we will stop dying - it just means vastly increased health span which means less health care cost and more joy in life! Congratulations to the authors of this paper!

All centralized services will eventually present a variant of this disease and indeed almost all are already. Actual baseline improvement in technology requires fully decentralized software similar to bittorrent and it's ilk imho.

Honestly nice move on CF's part, have to say; there's more real energy moving in the open source world since Lunduke started commenting.

Not only can you configure it to never stat; it has also been on by default since PHP 5.5 afaicr

I find some of the language around FrankenPHP difficult to parse - is FrankenPHP always in loop mode (ala ReactPHP) or is it simply an option? I can't tell.. my gut reaction upon reading the intro material is that they "simply" replaced Apache/Nginx with Go's inbuilt, and much lighter, server structure and that PHP itself is left to run completely isolated as per usual?!

edit: indeed https://frankenphp.dev/docs/classic/ indicates normal runner capability similar to Apache.

The point made in the article:

* resource leaks are non-existent

Can not be emphasized enough imo. The amount of absolute ridiculous levels of pain instigated by expecting a core code loop to run -forever- and doing so well while being written for some web project or other is just too much to fathom. Just use PHP and serve the request. Then let it die. Then serve a new request.

No no, not at all, see: https://openai.com/index/why-language-models-hallucinate/ which was recently featured on the frontpage - excellent clean take on how to fix the issue (they already got a long way with gpt-5-thinking-mini). I liked this bit for clear outline of the issue:

´´´Think about it like a multiple-choice test. If you do not know the answer but take a wild guess, you might get lucky and be right. Leaving it blank guarantees a zero. In the same way, when models are graded only on accuracy, the percentage of questions they get exactly right, they are encouraged to guess rather than say “I don’t know.”

As another example, suppose a language model is asked for someone’s birthday but doesn’t know. If it guesses “September 10,” it has a 1-in-365 chance of being right. Saying “I don’t know” guarantees zero points. Over thousands of test questions, the guessing model ends up looking better on scoreboards than a careful model that admits uncertainty."´´´´