ATC people are surprisingly resistant to change. They eat up whatever bullshit of a UI they are fed by the few companies producing that specialized software and when the time comes for a change they are too used to the old systems they want an exact replica. My current project is literally a huge ATC system and the more we try to bring in actual controllers for feedback the more we realize they just constantly yap about how they are doing it CURRENTLY in their ancient systems and ask of us that we make it the same. THEY ARE USING MICE WITH 3 buttons!
I am in the sector, I develop ATC software. It is not rocket science (which was also solved with money actually) you can actually solve ATC with more money.
I work on ATC software in another country. In my experience the biggest hurdle is the way the software is being developed and sold. ATC authorities and service providers buy these systems as a product but they don't have the code. Developers of the products (people like me) constantly have to maintain different versions deployed all around the world for different areas or countries or service providers. And there are hundreds of different systems working in parallel for no reason. For example there is an airport with 5 runways here and they installed a specific software just to monitor the speed and altitude of the planes taking off from this runway. They already have 5 different survelliance monitors feeding this info as well as direct view from the tower. Every new software now also has to consider and integrate with this specific system and many like it.
This is wonderful. Anything built that long ago that's still around is a testament to the amazing engineers who made it, and also it probably should be replaced or severely upgraded!
it's probably because one bot is the limit for a conversation to be intelligent. DeepSeek realized that he was more sentient and intelligent than another murican who keeps talking about bullshit and shut down the conversation before it got even dumber.
it's 2025 and this is what we are still reading on HN forums lmao... if you are not a historian trying to get this model to write a propaganda paper that will earn you a spot in an establishment backed university then I see no reason why would this be a problem for anyone. Imagine that OpenAI finally reach AGI with o-99 and when you ask chatgpt-1200 about deepseek it spits out garbage about some social credit bullshit because that's what supposedly intelligent creatures lurking HN forums do!
tech people are notorious for being ignorant assholes about anything outside of their field of expertise. There are multiple very reputable research showing smart people to be more susceptible to propaganda and brainwashing.
depends on the content length tho, I can download 1 GB of data in 100 seconds but if I am watching it for 20 minutes (which is not unreasonable at all given that torrents for 20 minute TV shows are generally around 1gigs at 1080p) I have 1200 seconds to upload that 1000 MB of data back. That's about 7Mbit upload which is not crazy at all
