Comments for A blog on math

Comment on Youtube Video Ratings by Mariano M. Chouza

Mariano M. Chouza — Mon, 09 Nov 2009 16:58:37 +0000

“The median can be computed in linear time but the computation is significant,…”

As the ratings of individual people fall in a small number of values (0, 1, …, 5), the median and the average can be calculated in almost constant time (technically logarithmic) from the aggregate numbers

Comment on Mono Asp.Net MVC and Nginx by Russell Kay

Russell Kay — Sun, 08 Nov 2009 09:28:36 +0000

I was having similar problems using fastcgi and Debian (the hosting setup on dreamhost), and I found the answer to the problem was not to comment the code out, but to replace the code with

if (pathInfo != String.Empty) {
	path = path + pathInfo;
}

This allows pages to be routed correctly (looks like fastcgi directs the HttpRequest differently from mod_mono, so the check is incorrect).

I hope this helps, took me a day or so of debugging to work this one out.

Russell

Comment on Youtube Video Ratings by Statistically Accurate Ratings « A blog on math

Statistically Accurate Ratings « A blog on math — Wed, 28 Oct 2009 20:54:51 +0000

[...] probability, rating, ratings, stackoverflow, statistics, wilson confidence interval In my previous post on ratings I noted some issues with using the mean vs the media for the rating. A few days ago Jeff [...]

Comment on Mono Asp.Net MVC and Nginx by ASP.NET MVC Archived Blog Posts, Page 1

ASP.NET MVC Archived Blog Posts, Page 1 — Wed, 21 Oct 2009 04:14:51 +0000

[...] to VoteMono Asp.Net MVC and Nginx (10/16/2009)Friday, October 16, 2009 from Bryan BellI’ve been using Mono ASP.Net for a while with Nginx as the [...]

Comment on Protected: Solving tangram puzzles by A Heuristic Solution to the Tangram Puzzle « A blog on math

A Heuristic Solution to the Tangram Puzzle « A blog on math — Thu, 20 Aug 2009 22:02:51 +0000

[...] me I’ll remove the document. ironically it’s taken me over a year since my post here to finally read the [...]

Comment on Line Segment Intersection by Bryan Bell

Bryan Bell — Wed, 05 Aug 2009 17:06:15 +0000

No you’re correct it’s a typo. I’ve updated the post.

Comment on Line Segment Intersection by DaveS

DaveS — Tue, 04 Aug 2009 23:41:46 +0000

Specifically…
(P + t . R) x S = Q x S
PxS + t(RxS) = QxS
t(RxS) = QxS – PxS
t = ((Q-P)xS) / (RxS)

Am I doing something wrong there? (I’m rusty, hence my visit to this site. )

Comment on Line Segment Intersection by DaveS

DaveS — Tue, 04 Aug 2009 23:19:19 +0000

You have

(P + t . R) x S = Q x S

… solve for t ….

t = ((P – Q) x S) / (R x S)

But wouldn’t it be…

t = ((Q – P) x S) / (R x S)

Comment on Solving Tangrams Using the Java Topology Suite (jts) by Antonio

Antonio — Tue, 14 Jul 2009 15:01:01 +0000

I have implemented a practical solution for Tangram puzzle with an OpenGL interface that is really fast, in pratice less than a minute for all silhuets i have tried (more than 10). My code uses “Positon” and “Clipping” method that is fast in a backtraking implementation. I have some simple files with input silhuet that I can share with you and I’d like to know if you have input silhuet for Tangram that you could share with me.

Antonio.

Comment on Protected: A Modified Algorithm for Solving Tangram Puzzles by Antonio

Antonio — Tue, 14 Jul 2009 14:48:54 +0000

Have you implemented a full algorithm for solving tangram puzzles ? did it work ?

I have been working with this subjetc for a month and got with an intersting backtraking solution using a polygon clipping algorithm (maybe the same way you do vert_solve) and a position method. I built an animated interface using OpenGL and would like to use compile it in Linux. Dow you know how to compile OpenGl in linux ? Do you have more deails about your implementation of this puzzle?

Antonio.