CodexArcanum

Ooh, I’d been looking at wasmer but wasmtime looks easier and more appropriate. Thanks for the suggestion!

Also wow, a D programmer in the wild! I used to really like that language before I got into Rust (my beloved).

Literally have a dozen other tabs open about how to embed a WASM engine into my Rust game. At least I’m not (currently, at this time, right now) writing my own language or trying to embed a prolog engine.

Yeah, you can just use Google’s VM, Google’s renderer, Google’s sponsored image library, write it in Google’s language, and you can probably borrow some of the WebGL code from the Servo project.

Obviously the problem is so trivial when you’re just bolting together premade components that the fairies delivered to you. Good thing none of those components are hard to write, hard to integrate, or written with corporate interests in mind.

Hey where’s your snap-together browser project? It’s so easy, all this free code just laying around.

Oh or were you arguing that I said you had to write everything from scratch? Because I didn’t say that. I also didn’t say that you needed all those things for communication, kind of the exact opposite. What exactly was the point of your “well acktshually” comment?

The modern internet, driven by corporate mandates, is rather complex. A browser needs to (at least):

• transmit and receive several protocols, usually HTTP/IP (ipv4, ipv6, http2, SSL/tsl for security, sometimes also ftp, gopher, torrents, etc)
• parse HTML documents and render them correctly to screen
• parse user input to forms
• parse CSS, correctly apply and cascade all the rules to the HTML element
• include a full, performant VM to run ECMAScript in, manipulate the rendered document tree according to those scripts
• include APIs required by ECMAScript covering the DOM, windowing, network requests, screen readers and other accessibility options
• render and work on at least 3 major desktop OS flavors and 2 mobile ones, across countless versions
• play media: images, audio, and video
• DRM handling for media
• Handle rendering of text in any world language

And that’s just the tip of the iceberg! I’ve come around to the idea that the modern internet is actually, to be technical here, totally fucked. Big Tech is going to keep pushing users deeper into walled proprietary gardens. They’ve already made the open internet so complex and heavy that it requires a multimillion dollar company (dependent on Google’s allowance or massive ad dollars) to create a browser for it.

I think the only solution is to throw it all away and start over. Twitter and reddit aren’t being “saved” by the “resistance” users. The concept of free and open exchange of ideas on the net is being saved by new protocols and services that are built to resist corporate ownership, like Gemini and the Fediverse.

It’s going to be hard weening off the flashy, ad driven Web, but its the only way. Go download Lagrange and start browsing Gemini space. If you weren’t around for the 90s era of GameFAQs and the mostly-text web driven by individual writers and hosts, then here’s your chance to go back to a better way of doing things.

We can’t “fix” an internet that’s owned by Big Tech, we need new spaces owned by the people using them.

I’m happy to unveil that ive fully moved off FF and onto various forks of it.

Can you imagine how great software would be if we had UBI? No one would need advertising money to stay viable, and every developer could just make whatever they wanted or what felt important to them.

import birthday;

let myAge1 = 4;
let sisterAge1 = 2;
let myAge2 = 44;

let sisterAge2 = birthday.deriveAge(myAge1, sisterAge1, myAge2);

print(sisterAge2);

Any bugs should be reported upstream. Please open a tracking issue to sync changes with eventual upstream fixes.

Oh don’t worry, I get myself involved in plenty. I prefer to make problems at the architectural or “leadership” level though.

Wildly, in C# you can do either and it has different results. I believe a bare throw doesn’t append to the stack trace, it keeps the original trace intact, while throw e updates the stack trace (stored on the exception object) with the catch and rethrow.

In C#, you can only throw objects whose class derives from Exception.

This is incorrect. The C# is valid. Throw in a catch statement simply rethrows the caught exception. Source: I’ve been writing C# for 20 years, also the docs.

I won’t act like MS absolutely didn’t steal core concepts and syntax from Java, but I’ve always thought C# was much more thoughtfully designed. Anders Hejlsberg is a good language designer, TypeScript is also a really excellent language.

PETA isn’t going to like all those für loops

At least the names are extremely self-documenting. Some of those German variable names are long enough they might even be self-aware!

It’s an untenable situation because its so much bigger than the tech world and open source. FOSS fundamentally works on a communal model: everyone needs lots of software, no one can hope to write it all themselves, so what if we distributed the labor out among the community so that everyone can work on some things important to them and the whole community benefits.

Then, capitalist businesses entered the picture and began using more and more open software as backbone for their enterprises. Government entanglements further complicate the picture, but fundamentally the capitalist mindset is incapable of building or maintaining our current technological base. It isn’t capable of maintaining or building our infrastructure either: almost all of that was built on government subsidies, socialism.

And now that vulture capitalism is the law of the land, everything is falling apart because there’s no more “slack” in the system where people can engage in personal socialism on projects like FLOSS, every bit of our time is being stolen to pad the numbers of capitalists.

This bleeds over into attitude as well. Every entitled user who thinks their personal issue is more important than any other concern is a trump or musk in miniature, believing that the the blowhard bravado of our current government is a model for forcing work to get done rather than a death spiral there’s no pulling out of.

You want FLOSS software that’s good? You want less burden on maintainers? You want a safer, saner, more human-centric technology base? You want a better tech world?

Eat. The. Rich.

I’m being sarcastic but not by much. Nordic countries do have much better digital id systems and the EU overall looks to be following their model.

https://www.mitid.dk/en-gb/get-started-with-mitid/

https://www.openbankingexcellence.org/blog/eid-and-the-eu-digital-identity-frameworka-scandinavian-perspective/

He’s complaining that a number isn’t unique and is being poorly used, but the number isn’t supposed to be unique and he’s complaining that it’s not being used in a way that experts are specifically warned not to use it in.

But on a second, stupider layer, this is the system those numbers originate from. So however they use them is how they’re supposed to be used.

But then, back above that first stupid layer, on an even more basic and surface level degree of stupid, the government definitely uses SQL databases. It uses just… so many of them.

It’s wild too. I’ve been in the hospital a lot lately and in addition to a bar-code wristband, every healthcare worker, before doing anything with me (the patient) will ask my full name and either birthday or address and then double-check it against the wrist band. This is to make sure, at every step, that they didn’t accidentally swap in some other patient with the same name. (Not so uncommon, lots of men have their father’s name.)

Meanwhile in like Iceland, everyone gets assigned a personal GPG key at birth so you can just present you public cert as identification, not to mention send private messages and secure your state-assigned crypto-wallet. Not saying such a system is without flaw but it seems a lot better than what we’re doing!

This is a good summary. I had to go pull up wikipedia on it since I roughly knew that social security was a national insurance/pension kind of system but am actually hazy on details.

The major issue with it as id (aside from DBA’s gripes about it) is that credit agencies and banks started to rely on it for credit scores and loans. You see, the US has a social scoring system (what we always accuse China of) but the only thing it tracks is how reliable you are about paying off debts. So with your home address, name, and SSN, basically anyone can take out loans or credit cards in your name. This will then damage your credit score, making it harder to get loans, buy a home, rent property, or even get a job.

That’s why Americans are always concerned about having our identity stolen: because you don’t need a lot of info to financially ruin someone’s life.

I’m hardly the king of databases, but always using a surrogate key (either an auto-incremented integet or a random uuid) has done me pretty well over the years. I had to engineer a combination of sequential timestamp with a hash extension as a key for one legacy system (keys had to be unique but mostly sequential), and an append-only log store would have been a better choice than an RDBMS, but sometimes you make it work with what you have.

Natural keys are almost always a bad idea though. SSNs aren’t natural, which is one pitfall: implicitly relying on someone else’s data practices by assuming their keys are natural. But also, nature is usually both more unique than you want (every snowflake is technically unique) and less than you’d hoped (all living things share quite a lot of DNA). Which means you end up relying on how good your taxonomy is for uniqueness. As opposed to surrogate keys, which you can assure the uniqueness of, by definition, for your needs.

I’m sure folks on here know this, but you know, there’s also that 10K a day that don’t so…

What makes this especially funny, to me, is that SSN is the literal text book example (when I was in school anyway) of a “natural” key that you absolutely should never use as a primary key. It is often the representative example of the kinds of data that seems like it’d make a good key but will absolutely fuck you over if you do.

SSN is not unique to a person. They get reused after death, and a person can have more than one in their lifetime (if your id is stolen and you arduously go about getting a new one). Edit: (See responses) It seems I’m misinformed about SSNs, apologies. I have heard from numerous sources that they are not unique to a person, but the specifics of how it happens are unknown to me.

And they’re protected information due to all the financials that rely on them, so you don’t really want to store them at all (unless you’re the SSA, who would have guessed that’d ever come up though!?)

It’s so stupid that it would be hilarious if people weren’t dying.

I didn’t realize Framasoft was contributing to so many open tools, that’s pretty neat! Reminds me of TROM who also fork and develop tools for a less commercialized world.

I can’t browse lemmy or play phone games while I read docs, but I can while I wait for the program to compile, load itself into a docker container, deploy to the test server, load my browser, and then fail to have fixed that bug I was looking at. Oh well, let me change one character and try again, it only takes about 15 minutes per attempt.