When it comes to scientific software for Linux, everyone would shout: yes, we have it, we have it all, and we have alternatives!
But today I realized: well, we have several, but the alternatives are not as it would look like in the first moment.
The closed side
The closed side, maybe in your eyes the dark side, is well known by the most people who deal with such kind of software. And the most thing you will find there will have versions for Linux: Matlab, Maple, Mathematica. And for statistics we have SPSS
These programs are good, reliable and have a well working company behind them which continues developing the software. Whoever dealt with such a piece of software nows what I mean.
The “open” side
And the other site? Normally you will now hear some famous names like scilab or Mupad. But other’s, more informed will point out that these are not really free as in freedom of speech.
Even if some people now about programs like Magma, there must be said that these are all (!) not free!
The real open side
Some people might refer to this side as the light side – but in the area of scientific computations it is not so light as it should be. There are several programs around, which are really free, and they can do a good job – but most of them had seen the last news, infos or development months and years ago. And even the ones which has a current development has a poor interface.
Sure, several of these programs do a really good job, if you now how to use them. R is even able to run S code and is a competitor to SPSS – as long as you study how to use it and as long as you can afford it to get used to a computer and the command line.
The same is for gnuplot, maxima and gap – they all do a nice job in their field of science, but the interafces are poorly designed.
Others, like euler, had a nice interface ages ago. The actual versions only support windows, the latest linux version with an old gtk version is ages old.
It’s funny that on the one hand, on the hand of the desktops, we are diving into the world of usability tests and Human Interface Guidelines while on the other hand some specific, very powerful programs do not even have a interface.
I do not say that every programm should have one – but these programs shouldn’t think they can competete against the “dark side” as long as they do not have a nice interface.
I would like to get a linux distribution with a pack of powerful science programs which have nice guis – it does not matter if it is gtk or qt/kde based, it should just fit. This would be a nice way to get back into the universities, back to the science students, to show them what free software is, and why it is important in a science world to have free software.
But at the moment it does not look like that…
But their is still hope, as everytime.
All the projects are asked more and more, and some will realize that there is a need of building such interfaces. Projects like axiom still need a lot to go, but are looking a little promising because they lern how to present themselfes.
So what I would advise them to do: start polishing up your webpage. Build a new one up which is in some nice, easy colours and which is not to messy. If you want to take the easy and fast way: take a wiki. It is easy to use, can be used and changed by almost everyone, can be improved easily and could improve the way you work if you use it in the right way.
After that, try to keep the “news” section up2date. It’s always important to show that you are still alive. Some projects might do a lot with hardly anyone recognizing that they are even still alive.
After that, start doing some press relations – there is not much to do but sending out a new-release-x.y mail to every press linux/open soure press site you know (slashdot, pro-linux.de, newsforge.net, whatever), when you do the next release. Add some additional information how active the project is, what have changed, and what the future steps will be.
And if you have done this, then try to start a simple gui – it would be nice to have a gui where you can choose the background color, the font, and store the last session to have a look back. Yes, I know, normal gui-terminals can do all these – but you have to start somewhere, and this would be a starting point.
I would love to see the free science programs comming back 🙂