fernando_campos

I still see XMPP mostly used in self-hosted or privacy-focused setups rather than mainstream chat.

Prosody has been the easiest server to run in my experience — lightweight and minimal maintenance. ejabberd seems preferred when scalability or clustering is important.

What keeps XMPP interesting is the federation model. Running your own server while still communicating across domains feels closer to email than modern messaging platforms.

One big difference between LLM progress and robotics is that language models benefited from purely digital feedback loops — training, testing, and scaling could all happen in simulation.

Robotics still pays a heavy “reality tax”. Every improvement eventually has to survive messy, unstructured physical environments where sensing, actuation, and safety interact in unpredictable ways.

My guess is we’ll see a ChatGPT-like moment first in semi-structured environments (warehouses, logistics, industrial assistance) rather than homes.

A general household robot feels closer to a GPT-4 equivalent problem than a GPT-3.5 one because reliability matters much more when failure affects the physical world.

From conversations with teams recently, AI seems to be reducing implementation time but increasing decision fatigue.

People can build faster, but choosing what to build and maintaining consistency across tools is becoming harder.

The coordination problem feels bigger than the coding problem now.

Retries become dangerous when workflows aren't designed to be idempotent from the beginning.

What helped us was treating every job execution as replayable and attaching a unique operation key instead of relying on execution state alone.

Otherwise retries silently create data inconsistencies that only appear much later.

One issue I keep noticing is that most anti-bot systems optimize for blocking instead of increasing friction progressively.

Rate limits tied to behavioral patterns rather than identity seem to work better — especially interaction timing, navigation flow, or session consistency.

We experimented with something similar while building HiveHQ and found bots usually fail when systems require small contextual actions humans do naturally.

Yes, without science nothing can save us like in Covid Pandemic

I will also try to use Claude but like to use OpenAI ChatGPT very much.