There was a while in college I was like that. It's like you confuse your circadian rhythm enough that it stops bothering you. It was super convenient, but once you sleep the several days in the row it starts breaking down. You just need to constantly be changing it up a bit.
My experience is the same. If I constantly change it up, I have no problems.
On weekends I try to hold to a more timezone specific schedule, but include odd naps at odd times to sort of keep the schedule of broken sleep.
One other aspect I've noticed is I used to feel more tired and people would always comment on how tired I looked. Now I feel much more awake and in the moment, although honestly I have no idea why this would matter.
I think I'm always kept at a constant state of having slept not all that long ago, even if it was a short period of time, and that helps? This is guessing.
The other aspect is, I try to keep my sleep intervals at an hour 30. so, 1:30, 3, 4:30 hours. Not sure if that has anything to do with it, but easier for me to wake up when I do that.
"The UT 3000 is the very first underwater communication system combining analogue and digital communication in one unit. In addition to the telephony and telegraphy mode, the UT 3000 offers unique features such as own noise measurement, horizontal distance measurement and transmission of SOS signals.
"The new digital mode for the first time allows fast and reliable transmission of digital data in water. The possibilities for digital data transfer are nearly unlimited. In the future, own position data, SMS, maneuver reports or target data can be exchanged between submarines in a more secure way.
Also: according to the UT 3000 MASQ brochure, it can transmit at up to 1000 bits/sec from 12km to 17km
Edit more: the brochure says the communication protocol overhead is typically 20%. Error correction is included - and features "excellent burst noise rejection". I would think underwater burst noise could be quite long. Would this require an interleaved code? Could someone comment on this?
Strang's hand-wavy teaching style in my opinion kinda falls apart towards the later part of the course. I feel like the whole second half is kinda half baked. I didn't come away really having a good intuitive understanding of what the SVD was or why I should care about eigenvalues. I definitely think it's a good place to start though, but if you want something a little more organized I'd really recommend.
It's very clearly written and all the proofs are not too long nor too short. It pretty quickly goes through all the stuff Strang covers and moves on to more difficult things.
Thank you for that recommendation; I had not heard about the book before.
An extra nice thing: the book's chapters are available for free --- for downloading and viewing, and not for other uses --- from the book's website: http://matrixanalysis.com
But you should read it cover to cover. You should buy a notebook; and some basic electronics equipment (and there's probably free simulation that's good enough online now) and work through it.
Draw the diagrams (freehand!), do the math (on paper, not with a calculator!), do the examples.
It's dense because it's complete (for the time, there's a lot of stuff that's missing for modern world) and it's terse because you're supposed to be doing the work at the same time.
And, if you really want to get in depth understanding you buy the AoE Lab-book and build stuff while you're reading the main text.
Yes, that's sort of the point. ;) Most textbooks that contain a lot of math and science-y stuff are that way. Also, you don't have to read a textbook cover to cover - I know I didn't in college.
Constructing the paradox is just the means of proving the theorem. The paradox itself is not particularly important. Also, the paradox in Godel's theorem is "This statement is unprovable" not "This statement is false".
