A/V sync and latency issues are as usual driving me insane. There's an opening for an amazing mobile app to be rewritten for dealing with this which apparently doesn't exist yet, here's how it would work: (thread)
The app has three modes: tick, tock, and calibrate. When it's in tick mode once a second it flashes a specific color light on its screen and makes a ticking sound. The tick sound should be high pitched and should be specific and the color should \
be a specific one. The details of what's best for those will be up to you, the engineer writing this thing. In the tock mode it has the camera on and waits for ticks to happen. When it hears and possibly sees a tick it calculates the offset between the light and sound for \
display and makes a tock response sound and color flash. That brings us back to the other thing the tick mode does, which is it calculates the round trip time for the tock sound to come in and displays that, along with the A/V sync offset of the tock.
For calibration you hold it up to a mirror and it runs both the tick and tock modes itself to calculate its own internal latencies so it knows to subtract those out.
The mere existence of this app would probably guilt trip the A/V world into getting their act together. People have no idea what the round trip time and A/V sync issues no their video conferencing software are. Once anyone can simply \
hold up phones on either end and be told how awful it is things might start to improve. It would also be an amazing since experiment. Sound travels at about a foot per millisecond, so it could reasonably accurately measure the distance of two phones in the same room.
Someone please write this. I ask for no cut or even credit, I just want it to exist so I can use it and video conferencing and playback starts sucking less.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
An engineering question I like to ask people is: What are the ultimate sources of the power we the human race use? (thread)
The challenge is both to understand the second law of thermodynamics ('gravity' is not a source of energy) and to get all of them. The last one is a doozy. See if you can name them all before reading on.
Chemical stores of energy are relatively small and need to be replenished or they get used up quickly, which is in the process of happening. They mostly occur in the form of Carbon which isn't fully oxidized.
Now a few highly speculative thoughts on better ways of getting things into space (thread)
The most important thing to understand about getting things into space is the role of the atmosphere. It both slows things down with air resistance and provides propellant which can be conveniently grabbed onto and flung backwards
As it happens, Earth's atmosphere is designed exactly wrong for space launches. It's so thick at the surface that hitting escape velocity will melt anything before it gets out, and becomes thin so fast that you can't use it to get any significant height
Why is the unit of specific impulse in seconds, and how does it wind up being proportional to exhaust velocity, which seems like it should be kg*m/s ?
Summarizing the helpful links people have posted: The starting point is impulse, which is how much 'push' you can get out of your fuel, which is force*time. For the same fuel you can do less force for longer, or more force for shorter, but the product is the same \
I for one have a tendency to call this 'power' which is incorrect, it's 'impulse'. Since force = mass*acceleration and impulse = force*time, impulse = kg*(m/s*s)*s = kg*(m/s) \
Trying to figure out what happened in the Avenatti case it's completely bananas and I have questions (thread)
The story is that Avenatti approached Nike threatening a lawsuit over them having violated NCAA rules that college athletes must be treated like slaves, and offered a settlement including him personally getting paid $20 million (or so) to \
run an internal investigation at Nike making sure that they continued to treat college athletes like slaves moving forwards. Clearly he personally really, really cares strongly that college athletes continue to be treated like slaves.
Writing computer programs to play snake is very interesting! Here's an overview, which I have many thoughts on including a straightforwardly implementable clear improvement (thread)
A much algorithm thing which works by dividing the board into 2x2 cells which makes calculation easier for reasons is here github.com/twanvl/snake/
The inefficiencies added by the limitation to cells are extremely small and not really worth discussing, there are vastly larger optimizations to be had for much less effort and risk.
And now for an explanation of how some classification algorithms work and an honest question (thread)
For classifying data like was used in the Netflix prize you have a big problem: There are lots of people and lots of movies, and the number of overlapping movie ratings different people gave is very small
In data science parlance the data is sparse and high dimensional. This makes it not very useful for guessing if a particular person would like a particular movie.