Benjamin Franklin famously taught himself to write well by doing what you describe: Read a piece of a book, then rewrite it, then compare.
At first his copies were badly degraded. Eventually, he was considered one of the best writers of his time.
I feel like there's probably some way "the copy is better" could be quantified (at least to the point where it fools most of the people most of the time). If so, then expect LLMs to learn the same trick within a generation or two.
Assuming O_DIRECT actually blocks until the SSD has acked (this isn't actually what O_DIRECT's contract says, but what they rely on), you have to wait until each page write acks whenever you need a persistence barrier.
My guess is the preallocation + zeroing is what got them most of the win, and the O_DIRECT is actually hurting, not helping throughput. This has been the case 100% of the time I've benchmarked such things.
If you're doing this sort of stuff for real under Linux, check out sync_file_range. It's the only non-broken and performant sync API for ext4 (note that it's broken by design for many other file systems, and the API is terribly difficult to use correctly).
If you really care, it's probably just easier to use SPDK or something. Linux has historically been pretty hostile towards DBMS implementations.
That's a lot of valuable information and thanks for the input. Yes the original blog post is mainly focusing on reducing the metadata overhead due to fsync(), and I got a lot of good feedback from here and a lot of discussion is beyond our original scenario settings. We would like to incorporate all these enhancement suggestions without re-introducing fsync(), and make it work for more general environments.
Not damaging their relationship with Google as a vendor most likely. For better or worse, GrapheneOS is depend on Android which is controlled by Google.
Is there a way to just ban all these sites? Like a firefox plugin or whatever that detects this crap, and just bounces over to some place more reputable, like archive.is.
As someone that uses AI agents, this makes me want to install a browser plugin for "public windows" that just archives everything I see, and then farms out clicks of content that are missing from those sites.
The result of this would be to upload it all to a bot-friendly alternative to archive.org.
Nice, I understand it is similar to ArchiveBox + its web extension.
Now to be honest, while it's optimal to archive pages from you browser view I am not sure I want a random web extension to be in everything I see from a security point of view.
I would rather have a local proxy doing it. Maybe something like the InternetArchive warcproc [0]. Haven't tried yet.
for a short time i had warcprox sitting behind my firefox and auto feeding its output to pywb, it seemed to work but i had connections failing randomly after having warcprox running for more than a few hours~days. not sure if it's an issue with pywb or warcprox but there were some urls missing that i did browse on firefox, and many dynamic pages couldn't be replayed at all.
Unless they use a completely different transmission platform than the 1500 or Jeeps, I wouldn't bet on it.
The transmission has an integrated computer that controls shifting, cooling, etc. The lines between that and the speedometer are flaky, so every once in a while it'll decide to not cool itself, get stuck in 2nd/3rd, etc, etc.
I have an additional 15 years of cool down I consider another Stellantis product. By then, they won't be selling ICE, for one thing. For another, the current engineers will be long gone.
It used to be that the six-speed Aisin HD transmission was an option, which wasn't a Stellantis product. Also, I'd immediately disable the AFM stuff, which has always been a threat to top-end lubrication.
My grandpa used to say that every American ought to purchase a Chrysler product once every ten years (if only to get the urge to do so out of their system) and sell it immediately afterwards. He was a Dodge fan in the same way that some people are Cubs fans; i.e. inured to disappointment.
It sounds like (from the earnings report) they now have four classes of users:
- Identity verified adults
- Identity verified kids
- Anonymous adults
- Anonymous kids
The last three groups are in a “degraded” mode, I think.
I imagine a large fraction of groomers do not want to tie their activities to real world identities, so they’re mostly group 3.
A different reading is that, if a groomer can trick the face scan into “under 13” mode (putting them in group 2), then the game opens up chat for them, but just with kids.
I’m not sure which it is. I’d happily pay $50 one time for a copy of the game that includes parental controls and does not include anonymous contact with random strangers (including sketchy content creators), but they don’t offer this as a product.
This is strong evidence they are doing child safety wrong.
My kids want to play roblox, but we all know their primary customers are pedophiles and influencers.
Actual parental controls would mean I’d let kids use this service (and so would my friends). Instead, age verification partitions the user base creating a special space just for child abusers + targets.
According to California’s mandatory child protection training, the first step a successful pedophile will take is identifying a trusted, but not actually monitored environment full of kids.
Age verification laws force platform providers to create such environments online. This shouldn’t be surprising because they were strongly backed by people that directly profit from privacy violations and online abuse (like zuck).
At first his copies were badly degraded. Eventually, he was considered one of the best writers of his time.
I feel like there's probably some way "the copy is better" could be quantified (at least to the point where it fools most of the people most of the time). If so, then expect LLMs to learn the same trick within a generation or two.
reply