mary-ext

I've noticed that there's a decent amount of people who had benefitted having access to computer and internet really early on that seemed to be pro on banning teen access to social media, is there a reason why? the social media of today don't seem all that much different from the internet forums of back in the day

if algorithmic amplification is the reason then I'm not sure why social media as a whole has to be banned over it.

Solid.js and Vue uses these sorts of tricks to shave the amount of bytes from the HTML templates they generate

https://github.com/vuejs/core/commit/30dbdc101a48d747b56bcdd...

are we not reading the same post and comments?

the "industry hearsay" from two replies above mine is about deliberate data duplication to account for the spinning platters in HDD (which isn't entirely correct, as the team on Helldivers 2 have realized)

the industry hearsay is about concern of HDD load times tho

not `__proto__` but likely `constructor`, if you access `({}).constructor` you'd get the Object constructor, then if you access `.constructor` on that you'd get the Function constructor

the one problem I haven't understood is how it manages to perform a second call afterwards, as only being able to call Function constructor doesn't really amount to much (still a serious problem though)

also depends on the codebase, if you use frameworks with deep reactivity like Vue, you can't do structuredClone without toRaw (which only works if the object is shallow) as it'd throw on proxy objects.

Svelte has `$state.snapshot()` for this reason I believe.

that said, gorhill has made a decent effort on making most uBlock/Adguard filter rules work within dNR.

the only problem is that you just don't have any choice for custom filters, it relies on prebaked resources.

closing open source plugins? which?

Zod's source code is a bit unwieldy to explain what's going on I feel, so here's a small validation library I wrote that uses the same principle

https://github.com/mary-ext/atcute/blob/596e023bcb490b16d09a...

the solution is to avoid TypeScript eagerly validating the object's value type. eager validation breaks cyclic references.

it seems worthwhile to only mention it as a sidenote rather than a negative score

I do wonder if it might have fixed the type complexity issue that has prevented me from exporting Zod schemas from a library, but I suppose I could try finding this out later

This is awesome, bundle size has been my number one issue preventing me from trying out Zod as v3 shipped with validations that were unnecessary for frontend use but weren't able to be treeshaken.

I've commented this before, but does this grant an exemption from the automatic purging that ITP does? the wording makes it seem like it does but that just makes their justification for ITP all the more baffling

Never go for ccTLDs for anything critical, since you're practically at the whims of the government controlling it (see: .af ccTLD that the Taliban took over)

That's pretty much an issue with every LQIP solutions though, including BlurHash and ThumbHash. The only thing that matters is that it's close enough to the real thing, since they're meant to serve as placeholders.

The animations makes it feel slow, the lack of any throttling/debouncing means that my input feels slow too.

does this grant an exemption from the automatic purging that ITP does? the wording makes it seem like it does but that just makes their justification for ITP all the more baffling

A vulnerability can be high risk, a vague disclosure means that people can only assume the worst.

Maybe don't vaguepost about vuln disclosures

I don't really understand what this is for then if it doesn't spawn in a minimal frame. This is somehow worse than just having a shortcut to open the app in a regular browser tab.

I don't really understand this viewpoint, Chrome does show your extensions

gorhill spoke against these sorts of testing sites back in Oct 2022

https://twitter.com/gorhill/status/1583581072197312512

I've decided that if I ever had a need to write React apps I would stick to react-router 6.0.0 specifically.

I think they did good with v6 despite drastically changing it, but the v7+ smells like trouble.

right, then the claim of this being caused by "AI companies" feels a lot more dubious because at least this time the perpetrator has just been this one customer on Alibaba Cloud, and we don't actually know what they're up to.

honestly didn't know hr actually stood for that, huh

I do wonder if it's a customer on Alibaba Cloud that's doing all this or if it's entirely Alibaba's own doing that's been wrecking havoc on everyone's sites (mine included)

I don't really like blocking an entire ASN, especially since I don't mind (responsible) crawling to begin with, but I was left with no choice

To be fair, Bluesky hasn't done a great job advertising the fact that you can buy domains from them.

It's not even promoted in the app, they only promoted it in a blog post and then never again.

https://bsky.social/about/blog/7-05-2023-namecheap

rather worried that it's going to go the same fate as ShareDrop (https://github.com/ShareDropio/sharedrop) and Snapdrop (https://github.com/SnapDrop/snapdrop) where they recently got taken over by LimeWire the crypto/AI company.

See this issue talking about WireGuard and F-Droid. https://gitlab.com/fdroid/fdroiddata/-/issues/3110

OpenAPK here seems to be linking directly to official binary downloads, whereas F-Droid insists on building the APKs for the apps.

the decision to stick with WebKit for Linux is interesting, I haven't had a good time with WebKitGTK as it provides a subpar experience, so I wonder if Orion is going to be different in this regard