As an Akamai user I already serve all my DASH traffic (video) over http3. Akamai itself return to origin only supports http 1.1 LL-HLS forces me use HTTP2.
The problem here is Akamai really in only supporting HTTP1.1 to the origin.
Cloudfare I think only supports HTTP2 to origin.
Does Fastly yet support QUIC to origin? Does Cloudfront, I could only find information about it supporting QUIC the last mile.
Maybe more CDN support will drive web server support.
I maintain auto-generated Rust and Zig bindings for my C libraries (along with Odin-, Nim-, C3-, D- and Jai-bindings), and it's a difference like night and day (with Zig being near-perfect and Rust being near-worst-case - at least among the listed languages).
> Do you find zig easier than the ffi interface in Rust?
Yes, but it's mostly cultural.
Rust folks have a nasty habit of trying to "Rust-ify" bindings. And then proceed to only do the easy 80% of the job. So now you wind up debugging an incomplete set of bindings with strange abstractions and the wrapped library.
Zig folks suck in the header file and deal with the library as-is. That's less pretty, but it's also less complicated.
I've somehow avoided Rust, so I can only comment on what I see in the documentation.
In Zig, you can just import a C header. And as long as you have configured the source location in your `build.zig` file, off you go. Zig automatically generates bindings for you. Import the header and start coding.
This is all thanks to Zig's `translate-c` utility that is used under the hood.
Rust by contrast has a lot more steps required, including hand writing the function bindings.
You only hand-write function bindings in simple or well-constrained cases.
In general, the expectation is that you will use bindgen [0].
It's a very easy process:
1. Create a `build.rs` file in your Rust project, which defines pre-build actions. Use it to call bindgen on whatever headers you want to import, and optionally to define library linkage. This file is very simple and mainly boilerplate. [1]
2. Import your bindgen-generated Rust module... just use it. [2]
You can also skip step 1: bindgen is also a CLI tool, so if your C target is stable, you can just run bindgen once to generate the Rust interface module and move that right into your crate.
Exactly.
gpu's have become too profitable and of strategic importance, to not see several deep pocketed existing technology companies invest more and try and acquire market share.
there is a mini moat here with cuda and existing work, but some the start of commodification must be on the <10 year horizon
I has to convert a bitmask to svg and was wishing to skip the intermediatary step so looked around for papers about segmentation models outputting svg and found this one https://arxiv.org/abs/2311.05276
more tricky frequently because you need to measure the moisture for each plant as maintainance is difficult without, but this is generally the most efficient low cost method in very arid regions from what I have seen (dad is prof in the field, so exposure is years of unpaid labour as a child and student)
you don't need anything fancy for drip, a small hole in the pipe and a timer on your pump is is generally enough. If you really want to go fancy you can isolate the system and use moisture sensors, which are cheap.
Moisture sensors that measure conductivity are pretty useless unless frequently recalibrated but time domain reflectometry sensors are much better and more accurate.
As I have aged (almost 50 now) I have realised that greed will turn any good thing dark without vigilance, and then who watches the watchers. Eternal spring perpetually. A culture which prioritises individual profits over the commons will destroy anything not regulated.
As I have said before on this forum, we need a way to price in the whole lifecycle of manufacturing through to waste and punishments for behaviour like Purdue and the opioid crisis should be the loss of all wealth generated by such dark behaviour.
of course this is also unworkable, but I am not personally certain what is workable without a secular moral revolution.
Or we could move away from pricing everything and the neoliberal obsession to turn everything into a market, and restructure our economy to provide everyone with their basic material needs, regardless of how much income they have.
even if we restructure everything to meet human basic needs, pricing will still be part of the equation. finite resources means value has to be attached to materials.
Value and price are not the same thing. Money is a form of access control which limits the availability of scarce resources to those people and organisations which have enough funds. It also collapses the value of all things to a single dimension. There are other ways to decide how much of something is produced and how it is distributed, for example participatory economics.
Money isn't just a form of rationing, it creates the scarcity it claims to solve.
A neoliberal economy wastes talent and skill in much the same way an ICE wastes most of the energy from the gas it burns. Vested interests clog up the engine and keep it from running cleanly and efficiently.
This doesn't just create pollution of all kinds - physical, social, political, and ecological - which makes the environment a very unpleasant space for most humans.
It also puts a hard cap on the maximum speed, which is nowhere close to what's possible.
pricing is the exercise of determining value, even in a non monetary economy (think barter or contribution) you still need to price the value of materials and time. obviously money creates secondary effects which are not related to value due to arbitrages and other effects, but I do not believe you can have finite resources without determining value, which is what I meant by pricing.
We still haven't found a better resource allocation model than pricing.
We have tried central planning, and it resulted in horrendous living standards (as compared to the western world), queues all-night-long that you had to wait in if you wanted to buy bread in the morning, "if you're not stealing from your employer, you're stealing from your family" being adopted as a common proverb, and the whole system basically running (for some definition of running) on bribes, favors and theft. Communism finally fell around '89 in most of Eastern Europe, and we're still recovering.
Perhaps you could solve some of these points with computer-aided optimization and dystopian AI-powered mass surveillance, but is that really what we want?
In my view, the problem isn't capitalism, the problem is the government trying to fix capitalism, but instead making it much harder for small competitors to emerge, effectively causing almost-government-mandated monopolies.
Think about what industries are complained about most in America, and how regulated those industries are. You can't just lay fiber, make medications or help patients without going through a regulatory minefield, mostly for good reasons, but this is why the big providers of these services aren't outcompeted by smaller ones. There's a reason why the mostly-unregulated big tech is considered to be one of the most trustworthy industries among most (non ideologically motivated) consumers, far surpassing any political party.
Capitalism is sometimes bad, central planning is worse, but heavily regulated capitalism is the worst of them all.
A key problem with unfettered capitalism is the tragedy of the commons. If left alone rogue/selfish actors will destroy that which belongs to all of us and is required to live (see nature). How do you propose to solve this without "benign" interference?
If your property generates air pollution, noise pollution, smells, unclean water, radio interference etc, you get taxed and/or have to offset the effects (e.g. by planting trees).
You can do this with very simple, straightforward regulation, in a way that is very easy to understand, doesn't require an army of lawyers to follow and doesn't advantage or disadvantage large companies.
The temptation of exceptions, exceptions to exceptions, and exceptions to exceptions to exceptions might be too much for governments to stomach, though.
The problem here is Akamai really in only supporting HTTP1.1 to the origin. Cloudfare I think only supports HTTP2 to origin. Does Fastly yet support QUIC to origin? Does Cloudfront, I could only find information about it supporting QUIC the last mile.
Maybe more CDN support will drive web server support.