Unless you also want to enable castle building, no, I think a 3D interface would only add unnecessary complexity and cognitive burden. A 2D UI on desktop with keys for moving, flipping, looking at cards, and stacking, splitting, shuffling decks seems feasible.
On the other hand, if you spend time surveying the state-of-the-art in 3D input and real-time 3D physics engine, it will be clear how far away a functional deck of card simulation is. We simply don't have the tech now, and won't for the next 10 years.
3D games work because their focus is extremely limited (racing, shooting games), and/or low resolution (minecraft). As soon as you try to emulate the real world, you get things like Surgeon Simulator[1], where the awkwardness of the medium becomes fun in and of itself, albeit extremely unproductive.
[1] https://store.steampowered.com/app/518920/Surgeon_Simulator_...
Early on in the history of TTS it was pretty clumsy but it's gotten much nicer since they added scripting facilities to let you automate game setup.
That and most of these problems can be solved in other, better ways... for the actual parts that are worth answering (no one is asking who is everyone meeting with now, and tiny avatars don’t answer that any better than a simple list).
Deep learning was a cool but impractical idea in the 80's and 90's. In-browser payments were laughed out of the room when Netscape and Microsoft tried proposing them to banks and credit card firms in the 90's.
Part of what the author is saying is that, maybe with our current toolset, we can find that the "simpler" interactions are more physical and interactive now than tapping and clicking.
Sometimes people actually are asking questions like "Who has meetings right now? Which rooms can I bump people from? Did half my group go somewhere that I should be joining?" and a list in Outlook doesn't always fit the bill. Sometimes I need to know where activity is in a building, and just giving me a list of zones or spaces in the building is indeed simpler, but less contextual than, for example, seeing a map with avatars.
There's something good about making what's old new again. It does happen that things we've decided are out of reach or impractical have a novel solution waiting in current capabilities.
It's the core idea that is bad, so the passage of time won't really change much...
We might built real-world like 3D-UIs to view with Oculus-style goggles, but 3D UIs in the 2D monitor / desktop (outside of gaming and modeling) has been proven a bad idea time and again.
That being said there were some full 3d prototypes early on that were atrocious.
[1] http://tetramor.ph/wormwood/view.cgi?url=static/doc/doc.xan....
Wasn’t this the thought behind the addressing of pages on GeoCities?
Perhaps that is why webassembly as been create for, to leap forward helping the emerge of new ( 2.5D) interface.
If not, then A13 iphone 11 will only help the mega bytes of poor dev reactjs interface sur-comsumming giga watt of power for no reason, and brutalism will then succeed ;) ;)
Why should we have to emulate real world interactions, the whole idea of computing is to make tasks easier, less of a cognitive burden, not replicate them in a virtual world. Sure there are specific kinds of software that are in need of a more humane approach but I'm not sure skeuomorphism is the answer.
The problem is the software is _so bad_. Someone like Zoom needs to buy this and fix the software.
Caravan to Midnight interview on John & Yoko
https://www.youtube.com/watch?v=NJBwsaRmNgQ
I remember now thinking about it when talking to a friend about why he doesn't like electric cars. The acceleration lacks the intensity, mostly sonically as we discussed. And I said we could skeuomorphicly add it back in.
From my point on view, using classical old school RPG-game type interface will be one of the possible next move ;)
