8,360 karma · joined March 21, 2014
goisser94@gmail.com
https://hobofan.com
Mastodon: @hobofan@toot.berlin (Twitter: @hobofan)
---
Currently building https://github.com/EratoLab/erato
Most match-based FPS don't do that anymore, as it's susceptible to DDOS. Whether a LAN-like mode is otherwise still feasible/acceptable by todays game quality standards is debatable.
> Provide server binaries for self-hosted servers
This is only feasible if the multiplayer backend is a simple server binary, which in many cases it's not anymore, but a full cloud architecture you would find for any SaaS app. There additionally is the issue of licensed libraries, which may prohibit redistribution of the server binaries (and may e.g. be bound to a per-host pricing).
> Now, please give me an example of a game that doesn't fit within that framework if you want to continue this argument.
Take your pick from[0] or a competitors website: PEAK, Content Warning, Gorilla Tag; All games from indie developers that heavily rely on good networking that wouldn't be feasible to be replaced.
> you can just make your game into a service requiring a monthly payment
Ah, yes the simple option of "completely tank your playerbase".
_some_ multiplayer games can, many can't, as they are using a cloud-based multiplayer backend that isn't easily replaceable (see other discussions in this thread). SKG makes no effort to address those differences.
They may have never made any money to begin with, as they ran out of money during the development phase of the game because they were trying to comply with the regulation, and never got to release the game. Regulation almost always places a higher proportional burden on the smaller players, while larger players can afford it, which is why sensible regulation has carveouts for smaller players.
> maybe there will a lot of lag, maybe something will break. But it is still better than not being able to play the game at all
How is that better? A multiplayer game with awful lag isn't enjoyable anymore, and a game without joy is just a chore.
So that excludes ~95% of addressable playerbase.
There is no guarantee that they did!!! Yes, the examples we are pointing out (typical "friendslop" games from the last years) made bank, and should be able to afford to afford and EOL path.
However for every successful game that uses those technologies there are ~100 that "didn't make it", or barely broke even that are now also forced to do additional work on something they either post-hoc now was financially unfeasible, or have to do up-front work on something where it's a gamble whether it will be financially feasible.
In my personal opinion, completely downplaying the effort and financial reality that comes with making games compliant, and based on that creating carveouts for e.g. sub-$100k-revenue games was the downfall of SKG. If they would have made an effort to recognize that, they would be able to mobilize a large base of the indie developer community as well.
If the law were to be passed, Photon would at the maximum be incentivized to produce a self-hostable API-compatible alternative that would be neutered to such a degree that Games still qualify as "playable" on paper but would be unenjoyable to actually play. More likely, they won't do anything, as they are not the game developer and not responsible for compliance.
On top of that, there are also instruments that help the voters track whether politicians are engaging in corrupt lobbyism like voting records + donation / campaign contribution records, though few countries do that to a degree that it forms a cohesive anti-corruption framework. None of those measures exist for individual voters.
Yes, they do. Small developers disproportionally have to rely on online services to make their multiplayer games work to a playable standard acceptable to the users, as they can't afford to write them from scratch (and couldn't even afford to do the devops work that comes with a self-hosted alternative).
Example: PEAK, on of _the_ multiplayer hits of last year from a small studio is built on top of Photon[0] for their multiplayer. If you were to remove that component you might as well completely rewrite the game.
The few examples you point out as "open source released with stubs" are also usually games that are decades old and cultural landmarks, where there was economic incentive from the right holders (good PR) to release them (e.g. Quake). This isn't tennable for your typical game that has to shut down online services because it's financially unsustainable.
Presentation-first judging has been a thing for a long time, and unless there is a organizing party that explicitly makes code reviews a part of the scoring, and the organizers ensure attendance quotas for different personas (engineer vs. product vs. designer) it will always drift that way.
The only ones I've seen switch to 3.1 Flash Lite were from 2.5 Flash Lite, and all for the most simple use cases, e.g. small UX enhancements.
Or Google. I'm working with multiple customers right now that are very pissed at Google for deprecating Gemini 2.5 Flash, canning the GA release of 3.0 Flash and now have to decide whether to bite the bullet of the 5x price increase for 3.5 Flash or switching providers. Quite a few of them will likely fully pivot to open models.
- I think the comparison to TCP/DNS/BGP is the more apt one compared to MCP/A2A
- Those protocols negotiate capabilities and exchange information about themselves, but not in a self-serving manner of just talking about themselves, but with the goal of ultimately transporting data for a higher layer. Ask Protocol lacks that.
- Objects don't exist in a vacuum, but in a context. As the objects will only know about themselves they will always be limited in how to describe themselves best. An LLM that lives on the outside and just gets a static description of an object will be in a much better description to answer an "ask" query.
- Given that the existing agent protocols you are putting it in a context in already come with "description" fields and the like, the protocol seems too little of a value add to actually target. e.g. there is no benefit for a MCP server to conform to the prescribed manifest rather than implementing a freeform "ask" tool.
- If you want to actually bring the point across that it "occupies a different position" than transport/agent protocols, don't put it into a comparison matrix where you force it into the same schema
- ("Open Source" doesn't count as governance)
MCP servers on the side of the consuming organizations fit into the existing IT landscape, with centralized safeguard on who can access what a lot better and are easier to administer than letting their employees run arbitrary agents against arbitrary sources and destinations and cause chaos.
Last year the government launched an goverment owned alternative "myGov" and now has to claw back market share, which I don't see working out.
With things happening in general, and with Bun's LLM-aided move away from Zig in particular, there is bound to be some interest in talking about LLMs and how that impacts Zig's future.
I think this was a well measured "hey, let's focus on thing we are coming together to celebrate and advance: Zig".
It doesn't have to be sophisticated and they don't need to implement it themselves. They piggy-back on OpenFeature where the client libraries have a simple targeting rule evaluation engine integrated.
With both of those combined they are currently just redistributing wealth to the elderly that have created this mess.
I'm working on an self-hostable LLM (web) UI[0] that aims to provide a comparable good UX to e.g. ChatGPT, and you are right that there is a decent amount of fragility involved, and more management overhead than most people would expect.
However, we usually find that those details happen a lot more in e.g. the harness (= out application), or some prompt tuning that's required for each of the models, rather than model quality itself. We have seen customers using self-hosted LLMs with similar user satisfaction across their organization to other customers that heavily lean on latest GPT-5 models on Azure. Especially given that you have to do some level of tuning and setup anyways, you might as well invest it in "local"/self-hosted AI (if you can make the financials of the inference cost work out for you).
I think it should also be noted that the inference providers on hyperscalers also tend to be quite fragile, each in their own way (e.g. Google with a horrible rate limit system or Azure with almost weekly intermittent 500-error incidents).