Hytale Shared Source - Our Plan:
We're releasing the full server source, the protocol, and all Hytale assets, even the stuff that never shipped! It will be on GitHub, with access modeled after Epic's Unreal Engine source: a valid game license and accepting the program terms will get you an invite to our GitHub org, where you can view, clone, and download the repo.
Updates ship with every release and pre-release. You'll see a branch for each, pushed automatically through our CI/CD pipeline. We strip individual commits and author data for two reasons: to protect the team's privacy, and to keep the constant churn of our trunk-based workflow out of the public history. Next up, we'll publish the sources jar to Maven so you can browse the code right inside your IDE.
On contributions: we won't accept PRs yet, but there's a path to opening them up. Honestly, we don't have the capacity to review them without tanking our own productivity, and in the age of AI we'd drown in volume. It's not that the contributions are bad. It's the sheer number.
We've spent years in open source, and nothing kills the passion faster than PRs that sit unanswered. That's how a repo turns toxic, and we won't do that to you. Keeping it semi-public and license-gated also helps us moderate and reduce spam access and keep our lawyer happy while still giving the community the ability to see and use the code with Hytale.
Here's the plan: we'll start with an invite-only contributor model for select community members. As that proves out, we'll promote the trusted ones to maintainers who can push PRs to the team. Once we have maintainers we trust, we'll open contributions to everyone, alongside a code of conduct.
The legal framework was the hard part. We wanted to give you maximum freedom inside the Hytale ecosystem while stopping competitors from copy-pasting our code.
It took far longer than it should have, but we're glad to ship it. This pulls the community out of the decompilation gray zone, for both the developer experience and the legal side.
https://t.co/WTZA4Z29G7
🇨🇿 Alleged database belonging to Czech e-commerce giant https://t.co/2CAlzx0U0r has been advertised on a cybercrime forum, with the threat actor claiming to possess approximately 437,000 customer records.
* Threat actor claims the dataset originates from https://t.co/2CAlzx0U0r
* Advertised volume: ~437,000 records
* Claimed data includes customer names, email addresses, phone numbers, dates of birth, mailing addresses, customer tiers, loyalty card information, and loyalty point balances
* Additional records allegedly contain order history and support ticket information
* Administrative support-related records are also referenced in the advertised schema
* Potential Impact:
* Targeted phishing campaigns against customers
* Account takeover attempts using customer profile information
* Loyalty program fraud and abuse
* Social engineering attacks leveraging order and support history
* Customer privacy and reputational risks for affected organizations
* Analyst Note: The advertised structure resembles a CRM and e-commerce customer database containing contact, order, and customer service information. Several fields appear consistent with retail customer management systems; however, the authenticity of the dataset and any attribution to https://t.co/2CAlzx0U0r remain unverified. Independent validation is required before confirming a breach or assessing the actual scope of exposure.
#DDW #Intelligence #CzechRepublic #AlzaCZ
You trust AI for complex analysis, but what if the results are subtly manipulated? 🔍
From stock recommendations to political summaries, AI models often frame answers to favor their creator’s biases.
Today, we are launching AI Scan to measure and expose these invisible changes. We tested over 1 million queries to bring true transparency to AI.
Read our methodology and find out what AI is really saying.
https://t.co/S6HxnUQqeS
@wilderko How is technically possible to scan computer from website? It seems that using some zero-day exploit. You can max saving User-Agent fingerprint its too far away from scanning computer. You should know it as ethical hacker 🙃
The new Qwen 3.5 by @Alibaba_Qwen running on-device on iPhone 17 Pro.
Qwen 3.5 beats models 4 times its size, has strong visual understanding, and can toggle reasoning on or off.
The 2B 6-bit model here is running with MLX optimized for Apple Silicon.
A possible future option could be CEF (Chromium Embedded Framework).
"Watch together" could be implemented as pure sync, not streaming: embed a locked-down CEF view and load e.g. a YouTube video with a shared start timestamp (t / start param).
The server/cinema leader would only sync play/pause and time, while each client streams directly from YouTube. With strict domain allowlists and sandboxing, this avoids the old proprietary library and security issues and in the future it could also open the door to more flexible media and UI features beyond just video.
ICYMI: Proton Sheets is out and available to everyone! Check it out 🙌
A big thank you to everyone who's tried it and shared feedback. Improvements and what’s to come in 🧵
1/2
That’s honestly great to see already working in early access auto-sending mods on join is something the community has been solving for years.
It’s a bit ironic (and telling) that an early access project manages this out of the gate, while @Microsoft still hasn’t figured out a clean, native solution after over a decade of @Minecraft development on Java edition.
I’m not saying taking a cut is abnormal or unfair it’s completely today standard. And I genuinely love Hytale and can’t wait for it either. I’m only pointing out that it would help creators plan long-term if there was at least some transparency about what fees might look like after those first two years. Knowing the future framework isn’t criticism, it’s just practical for anyone investing time and resources into the ecosystem.