> I pushed through to the bitter end of Ghosts n' Goblins. And damned if I wasn't rewarded with the message, "This was all an illusion created by Satan."
Weeell...here's the thing. Erm. You didn't push through to the end. You just got the "bad" ending.
The game does very much have a proper ending, and reaching it is surprisingly straightforward.
Sure, I get that. I was just aiming for comic effect in pointing out that the player's initial reward for having made it through all the levels of a very difficult game is "lol, do it again".
I did work my way through it a second time to see the "proper ending", and what's interesting is that I remember nothing about it. There's a moral in there, somewhere.
In Austria you don't need an Austrian passport/Personalausweis for a Digital ID registration. Your original passport (or equivalent) in combination with a certificate of residence, student permit or similar is fine.
Apologies that I'm latching onto your post for visibility, but for the sake of discussion - the European Identity Digital Wallet project specification and standardisation process is in the open and lives on github (yeah, the irony isn't lost on me :) ):
Everything's very much WIP, but it aims to provide a detailed Archictecture and Reference Framework/Technical Specifications and a reference implementation as a guideline for national implementations:
Nah, that's not a "sacrifice", but the only sane way. In the ideal case, clearly document the constructor with a warning that it's not ISO conformant and offer a ISO conformant alternative.
In my (unfortunate) experience, DateTime/Timezone handling is one of the things most prone to introduce sneaky, but far-reaching bugs as it is. Introducing such a behaviour change that (usually) won't fail-fast, will often seemingly continue working as before until it doesn't and is deceptively tricky to debug/pinpoint/fix ist just asking for a fast lane into chaos.
And even with JS going the extra mile on backwards compatibility, I don't think most other languages would introduce that kind of breaking change in that way either.
Huh, fascinating.
The Patagonia structure is actually strikingly similar to the Bosch model - non-profit owning the shares, but no voting rights, trust having voting rights but no shares - just taking it to the logical 100% conclusion without the residual influence of the Bosch family (having retained a few percent in both).
The model has worked well for many decades for a 100 billion$ revenue company like Bosch, good to see others taking a cue from them.
(Also goes to show that even constructs like these are not safe from corporate fuckups - see the emissions scandal...)
I really like C#, but I wouldn't go that far - unions are at least on the horizon, but I've sometimes come to miss the power and flexibility of TS's structural typing...(And so has Hejlsberg, apparently, seeing his reasoning for choosing go over C# for tsc :) )
>And so has Hejlsberg, apparently, seeing his reasoning for choosing go over C# for tsc
It was more related to the fact that the existing TS code was more easily ported to Go, and also .NET AOT wasn't mature enough at that time. Structural typing has its own problems. I'm personally not a big fan of it.
Might be my own taste, but except a few of the common and easy to understand structural typing code, I find it sometimes actually make things needlessly complex.
I also write lots of Typescript, and the furthest I go is to use 'Omit' and other utility types, but already feel like it's too much.
I've come to really appreciate Typescripts structural typing, because it reduces some of the overhead & prevents the unnecessarily tight coupling that has often annoyed me in other languages.
The overhead argument seems fairly objective to me - clean code with low coupling in C# et al. requires separate definitions of interfaces and implementations, explicit conversion methods between compatible interfaces etc. This adds up over time and makes refactoring pretty annoying.
The tight coupling happens when people don't bother to define interfaces. Suddenly I have to couple class hierarchy to classes from unrelated modules, all so the compiler is happy when I pass structurally equivalent data. To keep my own modules clean I have to add yet more interfaces, conversion methods etc!
Weeell...here's the thing. Erm. You didn't push through to the end. You just got the "bad" ending.
The game does very much have a proper ending, and reaching it is surprisingly straightforward.
You just have to beat it twice.
reply