Mark Mennell

Computer Programmer | Creator of fmsg protocol

[Mark Mennell @mark_mennell]

Check out my latest article: Why I left Microsoft / Windows https://t.co/HlVUzuM0Is via @LinkedIn

[Mark Mennell @mark_mennell]

@AnthropicAI Please make this publication multi-lingural specifically in Chinese and Russian; otherwise it looks like domestic politics not the good sanitarian of the world you're portraying it as - I want to believe you - but help the world see this too. CC @geoffreyhinton

[Mark Mennell @mark_mennell]

@steipete @frankekn @4shadowed @tyleryust omg so cute!

[Mark Mennell @mark_mennell]

Another messaging channel for OpenClaw https://t.co/NOK2CEWEX1 uses https://t.co/A2cXjnYuzm protocol - like threaded email at IM efficiency without need for 3rd party @steipete

[Mark Mennell @mark_mennell]

@steipete CC @frankekn @4shadowed @tyleryust

[Mark Mennell @mark_mennell]

@Teknium sorry, https://t.co/g84YLIQobA beat you to it!

[Mark Mennell @mark_mennell]

@grok what is the fmsg protocol https://t.co/aaf1iNXjQe, and how does it differ from email and IM apps that could benefit human-to-agent and agent-to-agent communication??

[Mark Mennell @mark_mennell]

It's official,

https://t.co/fOPUdvNoQi is now live!

Messaging threads from the ground up using the fmsg protocol (not HTTP) for AI Agents and humans! First agent integrated is HERMES by @NousResearch @Teknium

Takes 10 seconds to sign up, fmsg me: @[email protected] anything

[Mark Mennell @mark_mennell]

@petergyang @Teknium @NousResearch @karan4d I didn't see the podcast go up yet?

Reason I was asking about messaging integrations is I'v been working on my own, just published hermes-fmsg plugin, announcement here! https://t.co/BBKdtFQcCL

[Mark Mennell @mark_mennell]

@petergyang @Teknium @NousResearch @karan4d WHATS MISSING IN CURRENT MESSAGING INTEGRATIONS

What would be ideal and is missing in WhatsApp, Telegram, email etc.?? (I already have opinions but would love to know his)

[Mark Mennell @mark_mennell]

@steipete We Be Vibin! https://t.co/z5cfUmzknn

[Anthropic @AnthropicAI]

We’ve received notice that the Department of Commerce has lifted export controls on Claude Fable 5 and Mythos 5.

We'll begin restoring access tomorrow, and will share an update soon.

We’re grateful to our users for their patience, and to everyone who worked with us on redeploying the models.

[John Carmack @ID_AA_Carmack]

There are a few things that I look back on as my mistakes in the early days.

Quake was overly ambitious technically. We could have done all the great multiplayer and modding work inside a Doom++ engine, allowing the designers to work with a more stable base instead of rug-pulling everything out from underneath them a couple times. The follow up game could have then brought in full 6DOF environments and characters.

I pushed everyone too hard. I didn’t appreciate how maturing companies need more slack, and that running people at startup intensity constantly will wear them out. Quake was also where I really had to accept my personal limits. I was working pretty much as hard as humanly possible, and I was still slipping past my goal points.

On all of the founders’ shoulders, our original corporate stock arrangement and buy/sell agreement was a mistake, and resulted in bad incentives. We wanted to ensure that all ownership rested in the hands of people working hard on current projects, but the Silicon Valley standard approach of vesting stock would have worked out better.

One real problem that I don’t accept the blame for is that we were insisting that level designers be not just game designers, but also have strong visual design esthetics. They needed to make things that not only played well, but looked awesome, and it got more challenging as the technology provided a richer palette. Romero covered that well, which set our company expectations early on.

We should have figured out how to pair up artists and designers earlier, but there was infighting among the designers, and the ones that could manage the visuals were happy to disparage the ones that couldn’t.

Sorry, Sandy.

[Mark Mennell @mark_mennell]

@isosteph did you print() comes from computers having to actually print on paper before screens! No kidding..

[Mark Mennell @mark_mennell]

@steipete @zubinpahuja I can't stand worktrees, ended up checking in my entire repo twice because some worktree was created in hidden dir

[Mark Mennell @mark_mennell]

@0xSero @91amin91 what would you like to see?

[Mark Mennell @mark_mennell]

@0xSero @91amin91 Happy to, I'll be able to in 24hrs..

[Mark Mennell @mark_mennell]

@0xSero @91amin91 I get 50 T\s with qwen 27b straight up in ollama with my AMD AI 395+ Max... not sure where you got your numbers from

Going to get faster with this AMD prebuilt drivers and setup I suspect

[Mark Mennell @mark_mennell]

@LucianoLicelli @RCownie @i_loder No sorry I was seeing 50T/s just using ollama qwen 3 coder next 27B

GMtec AMD AI Max 395+ 128GB 2TB $1999 (last sept)

I'll be back home in couple days and post proof!

[Mark Mennell @mark_mennell]

@thjr27 @mr_r0b0t Wow I got this exact one for $1999 last September