[kde-edu]: Dylon: New developer in training
aleixpol at kde.org
Thu Dec 25 12:14:45 CET 2008
I am the KAlgebra maintainer for now and I'm glad to see anyone interested
on such a tool. :P
I am not planning to change any underlying technologies such as the parser
or anything similar by now (it is not being revamped), but it would be great
if you wanted to take a deep look at the project and help us take it
forward, providing new interfaces and ideas to make it more useful to users.
If what you plan is to make a frontend for maxima, well, this is a whole
different project and you might want to talk to koffice guys, because there
the interesting thing would be to have a KOffice maxima flake, which would
be great too.
You don't want to write yet another CAS, probably.
2008/12/24 Dylon Edwards <integr8e at gmail.com>
> I'm working on writing a fontend for computer algebra systems -- like
> Maxima -- to learn KDE+Qt, however, I'm very new to the programming world.
> It seems there are plenty of excellent free software CAS's, but nobody's
> developed an intuitive interface to any of them yet. Maxima does an awesome
> job of formatting its solutions at the command line, but many people are
> weary of using consoles; I would like to develop an application that would
> make using them as easy as (or easier than) their commercial counterparts,
> such as Maple and Mathematica. Such an application will take some time to
> develop, especially by a new programmer, but I'm prepared to spend a while
> on it, and have actually been working on it for about a year now (but only
> an upcoming 4th semester sophomore, working on my 1st Bachelor's at a local
> university, I haven't generated as much code as I would have liked). My GUI
> would be much further along than it is, but until last week, I was writing
> my own CAS because I was unaware of Maxima.
> Some of my goals for my CAS were:
> - It needed arbitrary precision
> - It should be able to break the solution down into easy-to-understand
> - It needed a clear, natural syntax that would be easy for anyone to pick
> For the 1st, I noticed floating point calculations were not very accurate,
> and had begun working on my own version of what I've now come to know as
> Java's LargeDecimal class, which broke the operations down into steps and
> stored the result of each sub-calculation in Strings. I thought I'd
> discovered something revolutionary until I learned about LargeDecimals a
> month ago (which made such operations much easier) :P
> I developed a good algorithm for the 2nd; basically, it solved an infix
> expression as we do, parsing through it using the order of operations to
> determine whether to perform each calculation. The parser would pass
> through the expression multiple times until it solved the whole thing,
> printing the current state of the expression upon the completion of each
> pass. Pretty simple stuff, if you think about it, but for somebody who had
> just completed his 2nd semester, the algorithm took some work. I'm still
> not aware of anything like it, and being so simple, I will probably try to
> get Maxima's devs (or those of whichever CAS I decide to use) to implement
> The 3rd is self-explanatory.
> In the last week, I've taught myself C++ and have begun studying, more in
> depth, Qt and XML. Originally, I had planned to write the application in
> QtJambi, but have become aggravated with its current state and have decided
> C++ would be better.
> My basic plan is to obtain a user's expression from a TextEdit widget, pass
> it to Maxima, which will be embedded and running in the background, have it
> solve and export the solution in MathML (which it already has a class for),
> and send the solution to KFormula's FormulaElement class which will draw the
> nicely formatted solution beneath the user's expression on the canvas. Does
> that sound like a plan?
> Also, if any of you would be interested in helping me get something off the
> ground, that would be cool too; I would be more than willing to help revamp
> KAlgebra, if its devs like the direction I'm heading, or I could create a
> separate project, it doesn't matter to me; I'm of the opinion that working
> with people on already existing projects is good, and that I should only
> branch out and start my own thing if absolutely necessary (thus the reason
> I've decided to go the Maxima route). I have a MUCH larger vision for my
> application than I've included with this email, but I figured this is
> probably a good starting point.
> Hello World!
> Dylon Edwards
> SSWJ - Stay(ing) Strong With Jesus, Always!
> kde-edu mailing list
> kde-edu at mail.kde.org
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the kde-edu