Hacker Newsnew | past | comments | ask | show | jobs | submit | oliverbennett's commentslogin

We’re playing with embodied LLMs that can externalise thoughts in a virtual environment. The idea is to help facilitate knowledge work.

It’s not our main area of interest, but it’s been interesting to experiment with how human/machine and machine/machine interactions work in real-time when you limit how fast agents can move or write. It's much easier to engage in a dialogue with agents that can't create / move tens of sticky notes and graphics faster than you can create one.

You can see a short, old video of the environment at https://www.temin.net


I see the red rings in front. I tried adding some depth cues to see if I could see it both ways - https://i.imgur.com/LsPtsRr.png

It kind of works, but for me it feels more like the blue ring now has a variable depth, with parts below and parts above the red rings, kind of like a piece of fabric draped over a bar.

I wonder how it feels for people who see blue in front?


In 2015 the UN created 17 ‘Global Goals’ (https://www.globalgoals.org/) that are meant to be a "shared blueprint for peace and prosperity for people and the planet, now and into the future":

Goal 1: No poverty

Goal 2: Zero hunger (No hunger)

Goal 3: Good health and well-being

Goal 4: Quality education

Goal 5: Gender equality

Goal 6: Clean water and sanitation

Goal 7: Affordable and clean energy

Goal 8: Decent work and economic growth

Goal 9: Industry, Innovation and Infrastructure

Goal 10: Reduced inequality

Goal 11: Sustainable cities and communities

Goal 12: Responsible consumption and production

Goal 13: Climate action

Goal 14: Life below water

Goal 15: Life on land

Goal 16: Peace, justice and strong institutions

Goal 17: Partnership for the goals

Each one is broken down into subtasks and targets. Making a dent in any of them would be worthwhile - https://en.wikipedia.org/wiki/List_of_Sustainable_Developmen...


Does anyone have examples of games with a client/server architecture, where new clients have been written, but that can still connect to an original server? I think there's a valuable learning exercise in the idea, but I can't find anything that actively courts multiple clients being developed.


Old School Runescape. Pretty much everyone is using the Runelite client instead of the official one

https://runelite.net/

https://github.com/runelite/runelite


Exactly what I was after, thank you. Your other comment[0] with a link to game engine recreations is also super useful.

[0] https://news.ycombinator.com/item?id=36374297


Similar to this is there any example of someone building an entirely "new game" by changing the front-end client and assets so heavily that it literally becomes a different story/world/lure/characters? Essentially a total facade reskin?


League of Legends, Dota 2, and Heroes of Newerth were all directly inspired from a Warcraft III map mod, Defense of the Ancients. Multiplayer Online Battle Arenas, MOBAs, are now an entire genre.


Of the 10 most popular multiplayer games in 2023[0] seven of them owe their core mechanics to mods / flashes of brilliance from a handful of amateurs.

A more or less correct history of their origins:

- Four of the games on that list are battle royale games, which started life as a somewhat popular Minecraft mod and really took off with "Battle Royale" - a mod of "Dayz" which itself is a mod of "ARMA II". It's mods all the way down.

- Two are tactical FPS games, which owe a huge chunk of their mechanics to "Team Fortress" a Quake mod and "Counter Strike" a Half-Life mod.

- One is a MOBA, which started life as "DotA", a Warcraft III mod.

Of the other three, one is Minecraft. Created by a solo dev, and I expect its moddable nature has helped its multiplayer popularity significantly.

One is Roblox. Created by two devs, and is itself a game creation system.

The final one is Genshin Impact - something of a an outlier in terms of team size and genre origins.

[0] https://twinfinite.net/features/most-played-games


I vaguely remember a mod called tower defense which we played on LAN, I wonder if that was the inspiration.


I've never played the map, but Aeon of Strife from Starcraft is a custom map that has the origin of Dota mechanics... and the Protoss Photon Cannon brought about some earlier versions of Tower Defense custom maps.


Bungie's Myth II supported extensive modding (everything defined in the assets), so there's many total conversions that have been done.


Ace of Spades is one. I usually play with the OpenSpades client[1], but there is also Betterspades[2], and probably many other clients I don't know of yet. There are usually about 10-50 players online on the public servers listed on BuildAndShoot[3], variable depending on the time of day and mostly from Latin America it seems.

One can host the game with piqueserver[4]. I'm not sure if one can still host with the original Ace of Spades server, but the game was 'shut down' in 2019 so maybe not.

It's well worth a go - there is intense satisfaction in digging a tunnel undetected all the way through to the opposing team's base! Playing with friends enhances the enjoyment for me as one can be a little more strategic when in direct communication.

[1]: https://openspades.yvt.jp/

[2]: https://github.com/xtreme8000/BetterSpades

[3]: https://www.buildandshoot.com/servers/

[4]: https://www.piqueserver.org/


There's several for Ultima Online - the original MMO:

- https://crossuo.com

- https://www.classicuo.eu

- https://libregamewiki.org/Iris2_3D_Client_-_for_Ultima_Onlin... - this is older, but really interesting as it's a totally different style to the official one


Second life is tangentially an example. they open sourced (or similar) the official viewer and some enhanced community and custom clients are out there like firestorm

oh, and quake 1's quakeworld client. many excellent quakeworld clients like darkplaces and ezquake. lots of one offs showing off tech, etc


There are several dragonrealms clients, but most use Genie 3, and that's now been open-sourced.


The only UI/UX newsletter I subscribe to[0] also has a a similar problem, but the advice is usually solid. Maybe the first and/or last rule should be ‘seek feedback’.

[0] https://user-interface.io/


I still think of early human progress as this slow march forward rather than what I expect it really was - thousands of years of rediscovery and reinvention by a few million people spread far and wide. Who knows how many groups of people, and their knowledge, were wiped out through bad luck, bad judgement, or worse.


The Tazmanians are a perfect example of this. They were separated from mainland Australia around 10,000 years ago by rising sea level. In that time they regressed technologically to the point where they had only 24 distinct tools compared with the several hundred available to the mainlanders.


Reminds of this awesome talk about technology collapse, has some petty cool historical examples too - https://m.youtube.com/watch?v=ZSRHeXYDLko


Both metaphors seem apt. One is leaning towards how a product makes a person feel when using it, one is about what the product offers.

I hear arguments around what products can offer me all the time, they make sense, but I generally go with how a product makes me feel. Probably more than I even realise.

I don’t use Emacs, I do whisk eggs and fish doughnuts out of oil with chopsticks, I couldn’t fully say why for either.


oh i get it and i will contradict my own saying "you dont need GPU for handling source code" when i talk about my personal projects, but there is a time and a place for all the tools there are: some have janky handles, some have single-purpose nibs, others are as customizable as your regular homeowners house.

i am still a bit iffy about emacslisp, but less so than before i grokked CL (thanks for that goes to "practical common lisp" by peter seibel)

but: whatever floats your boat will keep your boat afloat :D


If we’re at a place a LLM can make and use tools I’m excited about the limits of its ability to create tools... specifically to create new tools.

For me that kind of entailment is one of the more exciting pillars of general intelligence and unique among humans (I think?).


The next step could be pairing multiple LLMs with code synthesisers. The LLM could describe the target API, a second LLM would build said API, and resolve the query.

I think at that point we will be very close to an AGI.


I'm never going to stare at TV static and see deeper meaning in random noise, and while this is almost the opposite of that, I'm surprised by how I felt watching this.

Awe and mild discomfort aren't the usual feelings I get watching information being processed.


SEEKING CO-FOUNDER | Metaverse/Collaboration | Technical Founder | UK | Pre-Seed

A couple of years ago I started to build https://www.temin.net/ for my own personal use. Describing its current feature set as "Miro/Mural in 3D" wouldn't be too far off. Really interested in helping teams with externalising thought/visibility/community/serendipitous collisions.

Give me a shout if you're interested in turning it into something - email is in my profile.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: