Probabilty in a torrent (advanced probability)

  • Context: Graduate 
  • Thread starter Thread starter Goking
  • Start date Start date
  • Tags Tags
    Probability
Click For Summary

Discussion Overview

The discussion revolves around the randomness of file downloads in a torrent and the implications of this randomness for probability-based games. Participants explore various models of randomness, the nature of the downloading process, and the mathematical framing of related problems.

Discussion Character

  • Exploratory
  • Debate/contested
  • Mathematical reasoning

Main Points Raised

  • One participant describes the randomness of file downloads, suggesting a game based on the percentage of total downloading.
  • Another participant questions the ambiguity of "at random" and asks for a specific random process to model the downloads.
  • A comparison is made to preflop poker play, implying strategic elements in the randomness of downloads.
  • Some participants argue that the term "random" lacks a unique meaning across probability problems and emphasize the need for specificity in questions.
  • One participant proposes a related problem involving pouring water into two vessels at random and asks for clarification on how to frame the problem correctly.
  • Another participant challenges the idea that torrents are completely random, suggesting that factors like peer level and protocol can influence download times.

Areas of Agreement / Disagreement

Participants express differing views on the nature of randomness in torrents, with some asserting that it is truly random while others argue that external factors can lead to predictability. There is no consensus on the interpretation of randomness or the validity of the proposed problems.

Contextual Notes

Participants highlight the need for clear definitions and specific questions when discussing randomness in probability. The discussion includes various interpretations of randomness and its application to the downloading process.

Goking
Messages
3
Reaction score
0
As i watched the torrent i noticed that the files are downloaded completely at random. That is to say if you have 10 files, each 200mb, by the time the torrent is downloaded 10 % you can have 1 full file, or 80 % of one, 20 % of another , 10 % of all and so on. If you played a game, which is you bet on % of the total downloading of a torrent, and you win if the difference in mb between the most downloaded file and the least is greater than other person's > what percentage should you bet. Then you play the same game, but this time the goal is the sum of the first 3 files to be exactly 85 % greater than the sum of the last two. Also you examine the possibility of being an interval of numbers, when the chance is greatest, or the problem may not be leginamite at all and in this case, please tell why.
 
Physics news on Phys.org
Goking said:
A the files are downloaded completely at random..

In some contexts, saying something happens "at random" has an obvious mathematical interpretation. However, in very many contexts the statement that something happens "at random" is ambiguous. What specific random process do you have in mind for modelling how files are downloaded?

(If you can state the process, then perhaps you can write a computer simulation to answer your question.)
 
It's like preflop poker play.
 
Stephen Tashi said:
In some contexts, saying something happens "at random" has an obvious mathematical interpretation. However, in very many contexts the statement that something happens "at random" is ambiguous. What specific random process do you have in mind for modelling how files are downloaded?

(If you can state the process, then perhaps you can write a computer simulation to answer your question.)

This is the same random used in all probability e.g random like the movement of the particles of the atom, so you can't possibly know(although i know is unachievable in a normal computer).
 
Goking said:
This is the same random used in all probability e.g random like the movement of the particles of the atom, so you can't possibly know(although i know is unachievable in a normal computer).

There is no unique meaning for "random" that is used in all probability problems. If you want a specific mathematical answer, you'll have to ask a specific question. (I'm not saying it's easy to ask a specific mathematical question. To do so, requires learning various mathematical definitions.)
 
You have x amount of water, poured % by % in two vessels at random. At which point is most likely the difference between the two vessels to be greatest(20%, 50% etc.). (For a problem to be wrong, it has to be inadequately explained or contradictory in nature, not which language is used.) Please tell me if you understood the problem, how it should be given?
 
How are torrents remotely close to 'completely random'. You can nearly predict the download times based on peer level, protocol, and traffic. The problem is you don't know these factors for each download so it appears 'completely random' to you. If you control for these external factors, i would bet that you should expect download times to end approximately around the same time.
 

Similar threads

  • · Replies 1 ·
Replies
1
Views
1K
Replies
1
Views
2K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 9 ·
Replies
9
Views
6K
  • · Replies 8 ·
Replies
8
Views
3K
  • · Replies 3 ·
Replies
3
Views
4K
  • · Replies 8 ·
Replies
8
Views
2K
  • · Replies 11 ·
Replies
11
Views
2K
  • · Replies 2 ·
Replies
2
Views
3K
  • · Replies 53 ·
2
Replies
53
Views
8K