Mathematical Atlas Online

vitaly.friedman writes "Whether you are looking for Complex analysis, Differential Equatiions or Probability and Statistics references, the Mathematical Atlas covers the whole area of mathematics and related subjects. The so-called 'Gateway to Mathematics' is a collection of articles about aspects of mathematics at and above the university level, but (usually) not at the level of current research. "The goal of this collection is to introduce the subject areas of modern mathematics, to describe a few of the milestone results and topics, and to give pointers to some of the key resources where further information is to be found. Like any good atlas, we try to present several ways to look at each area and to show its relationship with neighboring areas and sub-areas."

ANd what's wrong with Wolfram site? ;-)

[PaulBu]

http://mathworld.wolfram.com/ [wolfram.com] -- great resource, not necessarily only about Mathematica (TM), but math in general.

Paul B.

F.P.???

Re:ANd what's wrong with Wolfram site? ;-)

[dickeya]

Navigating the Wolfram site is a walk in the park when compared to the Math-atlas. I tried to find information on both sites regarding least-squares solutions of linear systems. Wolfram took me directly to what I wanted simply by using the search box, whereas the Math-atlas directed me to listing on the search page from which I had to choose from several options. After thinking about this for several moments, I clicked one that directed me to a bibliographical reference with no immediate gratification.

Re:ANd what's wrong with Wolfram site? ;-)

[VStrider]

There is a search box [niu.edu], it's just not so easy to find. Overall, a very good site, though it badly needs some web design improvements.
Wolfram is quite good as well. I've used it several times and found exactly what I wanted, fast. I'll be using both from now on.

And for your particular problem...

[PaulBu]

(least-squares solution, that is), probably the best resource is "Numerical Recipes", they have been changing their location for as long as I remember, but Goodle says it is currently at http://www.nr.com/ [nr.com]

I remember dealing with that particualr issue myself... ;-)

Paul B.

Thanks for the link....

[dickeya]

I'm using the info to solve some image registration issues I'm having.... primarily perspective transformations.

Ugh.

Re:And for your particular problem...

[ChadN]

Well, since you mention Numerical Recipes, it is obligatory to post one of the many rebuttals [colorado.edu]. Basically, the books have some okay discussions (and cover a very WIDE range of subjects) but their code is crap. I say that boldly, since I must maintain code that was originally developed using their C libraries. There have always been better alternatives, and especially these days when so much is available on the web.

http://en.wikipedia.org/wiki/Numerical_Recipes [wikipedia.org]
http://en.wikipedia.org/wiki/GNU_Scientific_Librar y [wikipedia.org]
http://en.wikipedia.org/wiki/Linear_least_squares [wikipedia.org]
http://en.wikipedia.org/wiki/Singular_value_decomp osition [wikipedia.org]
http://www.netlib.org/ [netlib.org]

Re:And for your particular problem...

[radtea]

Basically, the books have some okay discussions (and cover a very WIDE range of subjects) but their code is crap

This assessment is insufficiently extreme on all counts. I have used Numerical Recipes for over a decade, including using it as a course textbook, where it was particularly good because students could buy it in the language they knew best and focus on the the algorithmic issues rather than learning a new language. But I can't get over how good and bad the book is.

They have extremely good disc

Re:ANd what's wrong with Wolfram site? ;-)

[WaterBreath]

Well, there's nothing wrong with it. The Atlas doesn't appear to be intended to provide the level of detail that MathWorld does.

But it does give a good overview of what a given broad topic is all about, in relatively layman terms. (Something MathWorld is not particularly concerned with.) And one thing it does offer that MathWorld doesn't is lists of resources for learning about the topics. Mathworld just has citations.

ANd what's wrong with M$? ;-)

[hadaso]

Same kind of comparison...

Frighteningly bad design

[Anonymous Coward]

It's often said that programmers make terrible web designers.

I think we can safely add math geeks to that list too.

Oh yeah

[poopdeville]

That's Davin Rusin's website. He just moved it to a new domain. It was on his edu website before. He's a nice guy. He's a regular poster on sci.math and (through cross-posting) sci.logic.

Re:Oh yeah

[poopdeville]

Errm, Dave Rusin.

How recent is this site?

[musselm]

From the site: "This site last updated (and all links checked) 2002/04/1"

It's a cool project, but man, it's ugly.

Specific design suggestions

[HerbieTMac]

It has been noted that the design of this website needs work. Let me reiterate that and add some specifics.

• Color. Forget the dithering. It renders horribly on LCD monitors and low-refresh rate CRTs.
• Search. Wow does that search need work. A google "site:math-atlas.org" thingy will work better.
• Navigation. Some may not agree with me but tables would be a definite plus on this site. Keep a sidebar visible so that I can see where I am and easily click around.

Overall, kudos for the content and generally improving the web.

PlanetMath and Wikipedia

[pfafrich]

Two other good maths encyclopedias are PlanetMath [planetmath.org] and Wikipedia [wikipedia.org] both are open content, open source etc. PlanetMath is pear reviewed and at a high level.

good tourist site

[opencity]

This is a good introductory site for dilettantes like myself. The Wolfram site goes over my head pretty quickly.

Wiki beats both, although the pages aren't neon.