(2020)

Compare the quality and intentionality of this writing versus 2025 LLM generated/assisted stuff.

Isn't a logical next step to extract the depth field? Possible?

Isn't the depth decoder part of the processing already?

I should have read the abstract.

Right there in the abstract:

"Our approach works by predicting a dense motion flow field and a monocular depth map directly from a single motion-blurred image"

I was hoping the paper was on _learning_ theory.

It is a book on that. Not sure why they left out "machine" from "machine learning".

It is common in Norway to also write 22.7.

They did this in Futurama

https://m.youtube.com/results?sp=mAEA&search_query=futurama+...

I agree. However, I have had some good experience using weakly dotted paper for math. Then it is easier to draw graphs by hand.

Here is an example, 4 different dot sizes:

http://trondal.com/p1.pdf http://trondal.com/p2.pdf http://trondal.com/p3.pdf http://trondal.com/p4.pdf

You need to print them to see which one is suitable.

I remember listening on FM radio to my 100MHz computer running FreeBSD, which sounded like calm rain, and to Windows 95, which sounded like a screaming monster.

Yes, ARexx was big. A lot of popular software had an ARexx API so you could script across programs. Pretty awesome. I haven't seen anything like that since.

From the first example, I don't understand from the syntax why the first formula becomes inlined while the second one is centered on the page.

It's the space after and before the $. A bit weird I guess, but I think it makes some visual sense.

I used a Powershell-script in a recent project and learned some lessons. It gets some data from an API. It is run dozens of times every minute, can run concurrently with itself, has some writing to a json file cache in a race condition free way. Works mostly ok. Development has been difficult because of low volume of forums talking about Powershell. Wouldn't have bothered to make the script without ChatGPT. One time it failed by inexplicably setting system+hidden+read only flags in a project folder. Quite a head scratcher. The worst experience was a bug I still don't understand: The script failed to get an access token when running in "ordinary powershell" but managed to do so when running in Powershell ISE. I checked every possible environment difference and concluded that there was only one: powershell.exe vs powershell_ise.exe. I am not going to be making any more Powershell scripts.