mycall

8 ms·

61.
by mycall 4mo ago
h4ckernews also accessed an Azure Function App that provided direct download URLs for internal FIFA files, including transfer reports and board level data, due to a lack of RBAC access checks.
62.
by mycall 4mo ago
For one thing, the rich were as rich in proportion as today. That alone would need to be reversed.
63.
by mycall 4mo ago
Does taking a skin cell, turning it into a stem cell, dividing it into specialized sperm and egg cells then making a living being provide some proof to Animism?
64.
by mycall 4mo ago
The biggest problem I see is that web browsers don't natively support .ans inside <canvas> without Ansilove.js or Xterm.js hacks. Having a native AV format that comes from ANSI, pre-rendered via FFmpeg, is the missing link for &l
65.
by mycall 4mo ago
Maybe the training approaches taken to date are wrong for decentralized systems. Setup a virtual subnet you can trust and do training on that. Create a AI model island in a trusted/federated model system -- definitely slower than the t
66.
by mycall 4mo ago
Nature desalinates seawater all the time in the form of water vapor without using electricity. There must be another way for humans to command and control the release of fresh water out of the ocean for our own needs.
67.
by mycall 4mo ago
Would you say it is homoiconic, similar to LISP where the syntax of the language is the AST; so, data can become code (Macros) and code can be data (the S-Expression)?
68.
by mycall 4mo ago
Yet they still love ordering from Amazon.
69.
by mycall 4mo ago
Do the same principles apply to diffusion-based modeling?
70.
by mycall 4mo ago
While it seems like a good idea, resending a growing context window is very inefficient and costly. Instance pinning would make a huge efficiency gains but also collapse LLM provider revenue. This is something open models could better solv
71.
by mycall 4mo ago
Wait until the agents start assigning you real world tasks to earn a paycheck. They can easily become the boss when they hold the strings. I feel this side of the discussion isn't happening.
72.
by mycall 4mo ago
Wouldn't it be easier if MCP supported vector embeddings as input/output? It would shift its integration from a Broker Agent to a Deterministic Semantic Router. That by itself would prevent context bloat.
73.
by mycall 5mo ago
In-app notifications settings should do this if they are trustworthy.
74.
by mycall 5mo ago
> orchestra and symphony pop You can find a ton of this on archive.org [0]. One of my favorites is from K-Mart background vinyl. AI can't quite do this yet. [0] https://archive.org/search?query=subject%3A%22Kmart%22
75.
by mycall 5mo ago
Does 802.11p work in any of these mesh networks? It could amplify their usefulness.
76.
by mycall 5mo ago
Let's do it today
77.
by mycall 5mo ago
The church was a great archive of knowledge for the longest time. They were the powerful few too.
78.
by mycall 5mo ago
Don't all nonprofits need more money to improve their sustainment?
79.
by mycall 5mo ago
The "Fiction" part should be obvious to the AI, what's wrong?
80.
by mycall 5mo ago
Could False = FileNotFound and this is a case of AST symbol constraints?
81.
by mycall 5mo ago
Do a deep dive into how OS/360 formalized to having DOS.
82.
by mycall 5mo ago
Google's AI search uses your previous inquiries and discussions to guide its response. It should be more precise than it is.
83.
by mycall 5mo ago
That's interesting as I was considering GGUF --> ONNX conversions (via Olive), but if this creates unknown distortions in the effectiveness and stability, it might be a dead-end idea.
84.
by mycall 5mo ago
You should double check Project 2025 before solely blaming Miller.
85.
by mycall 5mo ago
How does it compare to something like VexRisc [0] with respect to its capabilities? I know having x86 support is a metric all by itself. [0] https://github.com/spinalhdl/vexriscv
86.
by mycall 5mo ago
Not exactly DOS but give Balls of Steel a try [0] https://steamcommunity.com/app/358430
87.
by mycall 5mo ago
When you have AI models finding vulnerabilities by themselves in a for-loop, it isn't long before there is no human to go kinetic on.
88.
by mycall 5mo ago
I do hope that JEPA can help resolve the nonsense from AI models.
89.
by mycall 5mo ago
#2 is impossible now that oss models are readily available and nobody would know you are using them.
90.
by mycall 5mo ago
Google Maps is my favorite human harness.

More ›