111

and then we have a shitty app 😂

Comments
  • 29
    The issue with academic papers is that their math fulfils every definition of shitty coding practices, especially all these non-speaking variable names that totally look alike. There is a reason why Fortran code was always a total mess.
  • 10
    I spent a few hours on multiple days trying to decipher what Wikipedia was telling me about the algorithm I was supposed to implement and present for school, then tried multiple other sites and finally found a YouTube video using a totally different, way easier formula that apparently does the same.
    But maybe it doesn't work for imaginary numbers in five dimensions or whatever Wikipedia was trying to generalise.
  • 4
    Math as a language was invented for humans, not machines.
    The variables follow a strict set of definitions, therefore you do not need long names.
    Translating this into readable source code can be quite challenging...
  • 4
    @Fabian Wikipedia is by design NOT a learning book. Sorry to say that, but you are to stupid to learn that way. You can read algorithm explanation, for example of JPEG, then read on cosine transform, and quantisation, but will have totally no idea on how to implement it. Not to mention edge cases of any sort. This is the reason they teach math at CompSci. You'd better learn about concrete applications, like tensorflow, and then, having experience, learn about underlying mechanisms.
    You use this scheme everywhere else, even without realizing. Look at databases, how do they build indexes? Store their data? How does the memory paging work? :) you don't have to know for 99% cases :)
  • 3
    "Which yields the deglusion curve of: BULLSHIT."

    That point killed me 😂
  • 0
    @mt3o If it wasn't for school, I would have taken a totally different approach anyway. The algorithm I'm supposed to implement is a matrix solving algorithm, but it requires matrix inversion, which requires matrix solving.
  • 1
    @Fabian there are math books explaining this. They start with simple examples, then approach more difficult examples, edge cases, quirks and gimmicks and relations with other principles. Wikipedia is not a math book to teach those.
    Go to the library and check with physical encyclopedia then compare it with wiki. It's even harder with paper encyclopedia.
  • 0
    @mt3o The task was a project for about a week, among other homework. I'll not go to the library for that. :D
    No worries, I found a proper formula in the end and can implement that.
  • 0
    I had to design a centrifugal water pump. I didn't find a paper with th design rules, measures, or some other useful info about it. Only a lot of formulas to calculate flow rate, velocity, pressure and other shits. Went to youtube and found a video explaining exactly what I need with the maths simplified. I designed it, printed and works perfect.
Add Comment