Resilience Theatre

Never backup. Only bootstrap. Fast. #opsec

VisualPTT - generated code to test high latency link delivery. Meaningful application where user can record message by keeping AltGr pressed down and on release, message is released as normal file for transport. When received, another end plays it back automatically. Operates fully offline, includes speech-to-text with whisper.cpp and annotates received video message. All this on Debian 13 laptop, without internet connectivity. Check source here: https://github.com/resiliencetheatre/visualptt #offline #preparedness #resilience #space #deepspacenetwork #mesh #manet #tak #atak

@0x00string@infosec.exchange Thanks for comment! This is awesome. LLM usage is new thing for me and I've done fully manual coding for ages, LLM assisted ones lately and one project without even seeing the code. This gives good perspective where we are going with Coding. Somehow I have gained a feeling that "humans should not program computers". I am mostly wondering why Codeberg won't host repositories with all of these approaches. Would be nice to setup public forgejo for non human coders and say that it only hosts LLM created code :)

@0x00string@infosec.exchange Like I said, I believe future is that we don't read or write code. But we're there not yet. I just wonder, why generated code is not hosted by Codeberg. There are even build systems which do generate code on build. What becomes lying, I do that constantly to myself. It's only way to innovate and keep going, because 99% of things are useless what I do with computers :) It requires a lot of lying that things are worth of doing.

Updated project git [1] with system diagram. If you like to have full Offline capable situation map for your projects, feel free to check this project out. It allows you to have full world map and easy API to provide target data to map. It runs from single Python file. [1] https://codeberg.org/resiliencetheatre/map #OfflineMap #TAK #ATAK #SituationScope #ControlRoom #Resilience #CriticalCommunications #MESH #MANET

"Protecting our FLOSS commons from LLMs" is very interesting approach. It seems that if I create PoC tooling with LLM, Codeberg is not for me: https://blog.codeberg.org/protecting-our-floss-commons-from-llms.html #codeberg #llm #opensource

Single python file experiment for Situational awareness. Test server available at: https://vps.resilience-theatre.com:7443/ and source is here: https://codeberg.org/resiliencetheatre/map #TAK #ATAK #Meshtastic #offthegrid #preparedness

While you're selecting a product, I am advising strategy. https://resilience-theatre.com/permission/ #preparedness #strategy #dfir #resilience #redteam #opensource

@GrapheneOS@grapheneos.social You're just doing it wrong. Using current mobile hardware and android as base is just wrong approach. Happy to talk about this f2f.

@0x00string@infosec.exchange Yeah, luckily I have my own instance as well. But having a feeling that AI in coding is not going away.

@0x00string@infosec.exchange What was the question?

@0x00string@infosec.exchange This Codeberg's terms of use is actually very well formulated: "7. You must not share projects that mostly consist of code written by "generative AI"-tools (including services such as *Claude*, *OpenAI Codex*). Such projects having an unclear copyright status (see requirements § 2 (1) 1 and § 2 (1) 3) and furthermore have little safeguards to ensure that they do not include harmful code (c.f. § 2 (1) 5)." And I fully agree on this. I would not host publicly any AI generated code after seeing this. But equally, what prevents humans to act similarly towards code they publish? This is new era of liability for service providers!