HTTP-over-QUIC to be renamed HTTP/3: No more user edition

[archive.fo/UJCMC] zdnet.com/article/http-over-quic-to-be-renamed-http3/

[archive.fo/TACI4] datatracker.ietf.org/doc/draft-ietf-quic-transport/?include_text=1
Welcome to no anonymity

Attached: http3.png (1000x425, 6.61K)

so fucking what just keep using http/1.1 like a sane person
HTTP/1.1 is the last good version of HTTP

inb4 a string of new "security" problems and browsers start putting giant warning flags on HTTP/1.1 scaring the goyim away before blocking them requiring a configuration change to bypass a few months later. a few months after that the browser will just ignore the configuration option like chromium does with self-signed certs

and by browsers i mean chrome and firefox because those are the only browsers in existence and what they do the internet must do.

I don't like gÖ0glê either, but I don't see anything bad yet.

I think OP cited well:
It means if you have Dynamic IP, take traveling mobiles, that ID is used to identify the stream.
Properly named "Connection Migration."
I see the benefits and the cons. You should still VPN/proxy your connection regardless.
The point is QoS.

if you keep an open connection then obviously you are also keeping other shit
otherwise you will also start a new connection because why not
the fact that it's possible to retain connections doesn't mean everybody will force you to use 1 and only connection forever.
pls don't be dumb.

...

So let me get this straight im a brainlit
The migration addition is supposed to keep an open connection and detect when things like your ip address change and route it back through the same connection regardless?
If so, how is that preventing anonymity?
What prevents me from closing the existing http/3 connection and then changing my ip?
All i'm seeing here is more fixes to initial connection problems. That is unless you're completely locked out and that connection isn't easily circumvented which i highly doubt is even possible

The reason i see for this is something like my router decided to change my ip suddenly and i ended up getting that socket connection error on a website i was downloading from.
I guess the idea is that this should fix that sort of thing from happening within http for some reason but i don't understand why that'd break your anonymity.
Actually, the only way i can see this breaking it would be if your using a vpn and for whatever reason the connection stops and you don't have a deadswitch enabled.

I guess that pointed above can be a major issue.
Please explain if i missunderstood.

Attached: 1530414622103.png (248x248, 128.52K)

Nothing. It just gives you the ability to have ips change without the connection breaking. You can always just close the connection like normal.

Tor Onion Router

It's not recursive Mr. 133t.
The Onion Router

YEAR OF THE GOPHER WEBTOP
NOW

Attached: GOPHER.jpeg (640x480, 161.85K)

we should literally go back to HTTP 1.0 and HTML as it was 20 years ago, and keep nothing but from HTML5

that's what they _WILL_ do. HTTP/HTML is for viewing static documents, not applications. when people get confused and think they can implement a banking application over the web, the result is CSRF,XSS,clickjacking, and whatever new meme vulns kiddos are talking about these days. there will always be new updates to patch these (when they are caused by something inate to the browser or protocols, instead of the website), but there will never be an end to them. they're literally just patching them to look good

Enjoy your slow as fuck connections. Also

...just reset your ID?
god this board is full of fucking retards

They're all larpers, user.
Just roll with the crazy and see how far the tards go.

Attached: 1542071483.jpg (620x515, 58.77K)

It can't undermine security of Tor or VPN.

Google, Cloudflare, Nginx, Apache, GitLab.
Just off the top of my head

i dont see the point of designing standards for things nobody uses

It is better to standardize on a standard earlier than later so you don't have to support all the weird implementation oddities that are in different people's existing implementations.

indeed innovation in large scale almost entirely depends on how good the open standards are in specific industry

Before long, it will be too hard for small-time programmers to implement code that uses basic internet protocols. Then you are stuck trusting third party compiled binaries.

There is no reason to continue mucking up HTTP. None whatsoever. This is an obfuscation maneuver. Reject it entirely.


HTML can't last forever.

Attached: 9dc20abfb8f8cb0dda2ffa55f0cd348e646508d5d567c5eb0ce59b61d4882e9b.jpg (600x428, 27.83K)

literally just change your connection id retard.

Good shit, been advocating for this

I don't do online banking and I wouldn't recommend it but everything you named has to do with user content or third party content on the website.
One should think twice before running those flickering ads on one's website or anything one hasn't created by oneself.
I personally think the ability to link things from other servers without even proxying them over (or mirroring them on) one's own server is the biggest problem the web has.

We seriously need a modern gopher net.

What does that mean? Gopher already exists.

Huh, what are the RFCs for those?

What do you suggest in its place? One UBP (Unified Botnet Protocol) to rule all "modern internets"?

...

JSON.

TLS 1.3 to be strict, dear satan.