Show more

Had a very interesting chat today with Sebastian from Mozilla who's working on their android components project: github.com/mozilla-mobile/andr

If you want to make a browser for android, this is the place to look. Well written, well-tested, modular and composeable components covering all of the different parts of the browser. This will make making browsers and browser-like apps much easier!

"Why should I use a Reverse Proxy if Node.js is Production-Ready?" by Thomas Hunter medium.com/intrinsic/why-shoul

Really well-written and interesting post. I appreciate it has a benchmark to back up the performance argument.

Thank god the entire industry hasn't decided to revolve around Blink/Chromium for 95% of the web's browser traffic, or else Google would have an unlimited ability to push forward whatever bullshit standard they wanted.

Wait hold on, my producer is telling me something [places finger on my ear piece]

Oh no.

Been writing up how basic Dat support is implemented in Firefox with dat-fox:

sammacbeth.eu/blog/2019/03/22/

Next post will be how dat-webext fixes all the shortcomings that dat-fox has!

“These companies are unavoidable because they control internet infrastructure, online commerce, and information flows. Many of them specialize in tracking you around the web, whether you use their products or not. These companies started out selling books, offering search results, or showcasing college hotties, but they have expanded enormously and now touch almost every online interaction. These companies look a lot like modern monopolies.”

gizmodo.com.au/2019/02/i-cut-t

DatArchive support is also included, so Dat applications like Solo also work: dat://6af095e72edd6d729bad712ff545c25030111f0c72acf569955939351879f5be/

New milestone - beta build of Cliqz browser with dat:// support built in! Dat urls load natively using the dat-webext extension.

Great news, I've got a #dat archive explorer app sketched out and working with dat-gateway.

Bad news: It only works when debugging remotely and freezes up.

Worse news: I can't debug it since attaching the debugger changes the JS engine that's being used. 😂

github.com/RangerMauve/datmobi

That feeling when an awesome idea gets shot down because of limited browser APIs. Sigh.

RT @evanpro@twitter.com
Fuck reCaptcha.

I am sick of doing unpaid labour classifying images for Google.

We need a captcha widget that contributes to the global commons instead of siphoning value into yet another proprietary lockbox.

SUMMARY OF FINDINGS

HTTP requests to appleid.apple.com are rejected with 502 Bad Gateway if the User-Agent string contains any case variation of "X11; Linux".

That said, I wanna compile a lil' list of people working on #Dat / #IPFS technologies.

I know @aral's working on it as part of Hypha and @darius plugged me to probably the best guide I've seen.

Who else is out here for the distributed Web / #dweb?

The performance has been shown not to hold for adblocking extensions (whotracks.me/blog/adblockers_p), and there are many other problematic webextension APIs w.r.t. privacy.

The solution to these issues should be institutional, not technical. Google should impose performance constraints on extensions, and audit what data extensions are sending home. However, looking at the rampant malware on all their stores, this is something they're unlikely to do, so instead they'll nerf extension's capabilities.

Disagree with this article's thesis that the Webextension webrequest API should be removed.

By moving to a declarative blocking API you are limiting the possible ways for extension to function to blocking and redirecting. This is a very narrow use-case, based on the current prevalence of blocking extensions.

But this is not the only way to solve problems such as blocking ads and improve privacy, as shown by privacy badger and Cliqz' tracking protection.

ctrl.blog/entry/removing-webre

A product's privacy claims are only as good as its default settings.

Providing options to change the default privacy-violating settings in some "Advanced" section doesn't make your product a privacy-respecting one.

#Privacy

@liaizon Mastodon would need to be able to speak over the dat protocol and make the query to confirm. so we'd have to add a dependency to Mastodons server of a library that can essentially do "curl dat://" and then use that to confirm the link

Great post by a colleague on benchmarking adblocking engines. The level of optimisations in these engines is incredible! whotracks.me/blog/adblockers_p

I made a thing. 😀

This uses the work I'm doing on dat-js to load content from
#Dat archives using either WebRTC or a websocket gateway. This is a proof of concept and I'm still playing around with performance and the API, but I'm liking it so far. 😊

ranger.mauve.moe/dat-js-exampl

#p2p #dweb

Show more
Mastodon

macbeth.cc is one server in the network