jStat: a JavaScript statistical library | Hacker News Reader

jStat: a JavaScript statistical library

jstat.org

79 pointsHirvesh12 comments

Via Functionn - Open Source Resources For Web Developers & Designers: http://functionn.blogspot.com/2012/11/jstat-javascript-stati...

P.S. Functionn contains a whole lot more of awesome resources like jStat. There only a fraction of them I can post here at a time. Take a look if you're interested, and subscribe:

http://functionn.blogspot.com

A solid JS library for statistical computing is definitely long overdue. After skimming through it, JStat seems to only show some trivial examples, and it doesn't seem to be designed for being extended.

What I would really like to see some day is a JS library that extends D3 for graphics and controlling DOM events, that has a good standard library for common statistics (and that is easy to extend), and ideally a possibility to delegate computationally intensive tasks to C (perhaps FFI through Node.js?)

Jason Davies (frequent D3 contributor) started a standalone stats library: https://github.com/jasondavies/science.js/

Originally intended to be part of D3, he's now set it up as its own library.

It's coming soon and much more: http://angel.co/sense, http://www.senseplatform.com. Three of us, full time, for over a year.

It's weird that one can't easily find any links to their github page from the site. Here it is:

https://github.com/jstat/jstat

I almost thought they weren't on GitHub.

Thank you for posting the link to the Github page...additionally for those interested, the library is by John Resig. Put it in the title earlier, got moderated out.

The library is also funded by the European Community's Seventh Framework Programme (http://cordis.europa.eu/fp7/home_en.html).

I didn't edit the title, but I also don't see anywhere that this is by John Resig. See https://raw.github.com/jstat/jstat/master/AUTHORS

The authors merely use some code that he published on his blog a while back at http://ejohn.org/blog/simple-javascript-inheritance/ and that's why they cite his name in the code file.

ahhhh, my bad! thanks for pointing this out too :)

The GitHub version and the version on the jstat website are vastly different. Just thought I'd point that out. I like the way that the GitHub version is laid out in a modular format. Some of the functionality is different so you can't just switch from the old library to the GitHub version and expect all the code to work.

You're absolutely correct. If you look at the commit history, you'll see that it started out as two different projects. We noticed we had the same goals and decided to combine projects. I've been in charge of architecting the API since then.

One of my goals was to make it very modular and easily extendable. In the overview page of the documentation I explain how this was accomplished (http://jstat.github.com/overview.html)

It seems promising, but it also seems like an inactive project, according to the commit dates on the GitHub page linked in the other comment.

Also, there is a sizable amount of dependencies for this library, all due to the use of flot. Since most of the code doesn't use any of the dependencies, I wonder if they would consider releasing any future versions as two parts, jstat.js (containing the number crunching methods) and jstat-flot.js (containing the plotting wrapper methods).

I've been really busy the last several months. The project definitely isn't dead. I check it every day for pull requests and bugs. Just haven't had the time to extend it myself.

The most recent documentation is here: http://jstat.github.com/

Mailing list is here: https://groups.google.com/forum/?fromgroups#!forum/jstat-dis...

Reply on news.ycombinator.com