jfw: I'm logged up from where I jumped in to present. My pace did pick up after Wednesday, and I found myself engaged. Sometimes I noticed I'd wandered off to composing imaginary responses in my head; I think that's more what happened Wednesday too, as opposed to unrelated daydreaming, and there was also some unscheduled blog reading.
ossabot: Logged on 2019-09-20 04:20:16 diana_coman: iirc you had a question re what does it mean to train here - take the question and read the log with this goal ie as you read try to extract and refine gradually the answer to this question
ossabot: Logged on 2019-09-21 12:52:32 shrysr: diana_coman: re: lonngg - tbh i aborted after.. ~5 min hehe. but yes. hmm the funny thing was that nothing was written into the file after abort. each loop supposed to curl + append to a file .... so i expected a portion to be there. perhaps due to exec from emacs + org mode rather than terminal. will get back to that
jfw: also perhaps worth pointing out that arithmetic done directly in bash uses fixed-width integers and will overflow, though presumably not on the scale involved here.
ossabot: Logged on 2019-09-22 03:34:01 jfw: I'm logged up from where I jumped in to present. My pace did pick up after Wednesday, and I found myself engaged. Sometimes I noticed I'd wandered off to composing imaginary responses in my head; I think that's more what happened Wednesday too, as opposed to unrelated daydreaming, and there was also some unscheduled blog reading.
jfw: Some I did write in notes; it does sound good to capture more of them
ossabot: Logged on 2019-09-22 03:43:02 jfw: also perhaps worth pointing out that arithmetic done directly in bash uses fixed-width integers and will overflow, though presumably not on the scale involved here.
shrysr: From what little I know - an integer or any data type is allocated a defined amount of memory, differing in each language (?) and by overflow you mean that the space is insufficient to hold the integer?
diana_coman: shrysr: specifically: there's a maximum number that you can represent on a given number of bits; one bit can hold only 0 or 1; 2 bits will be enough for 0,1,2,3; n bits are enough for 2^n-1 ; do you know why?
diana_coman: if you try to store a value outside the range that can be represented on that precise number of bits, what happens is overflow/underflow (depending on which side of the available range that value is)
diana_coman: at the very least, on underflow/overflow the given "result" will be rubbish (ie it will be a different value than you expect)
diana_coman: in some cases you can further end up with additional unpredictable effects (eg if the value is written on as many bits as required rather than as many as reserved)
shrysr:
http://logs.nosuchlabs.com/log/ossasepia/2019-09-22#1003297 << haaa! i vaguely remembered this from vocational electronics class though i had to look it up just now. base 2 for eg 2^2 (2 bits) can represent 3 (4 incl 0) base 10 numbers as in 00, 01, 10, 11 and the largest number is 11 which translates to base 10 number 3 (converted via 8421 rule). 2^2 -1 = 3 numbers bc integers start from 0, but we count
snsabot: Logged on 2019-09-22 09:47:12 diana_coman: shrysr: specifically: there's a maximum number that you can represent on a given number of bits; one bit can hold only 0 or 1; 2 bits will be enough for 0,1,2,3; n bits are enough for 2^n-1 ; do you know why?
