Currently there are few applications in KDE explicitly designed for use in science, mathematics, and engineering above the high school level. Most existing applications are part of the edutainment category and tend to be skewed towards less advanced audiences. So please post ideas about applications and tools you think would be of use to more advanced levels. Feel free to comment on the ideas presented by others as well.

Here are some of the ideas brought up, in no particular order (see here):

• An IDE for script (interpeted) programming languages, particularly python
• A matlab, maple, and/or mathematica clone
• A good reference and paper manager
• A 2D molecular model drawer
• A mapping/GIS tool
• A 2D/3D design/CAD tool
• A lexical analysis program
• A circuit simulator
• An economics simulation program
• A generic simulation program
• A sequence alignment program
• A cladistics program
• A generic object categorization and display program

So does anyone have an additions, or any comments on what they would like to see in any of the programs above?

Python-IDE: Well, somebody would have to work on the KDevelop-Plugin…
There is already a Matlab clone called Octave und Sage is really good. And there is Cantor. I do not think there is need for new KDE-math-software.

Paper manager: On top of Kile+Nepomuk? Kile definitely needs nice reference-management.

GIS: Marble?

Lexical analyzer: I am working on it, I can already create NFAs, convert them to DFAs and minimize them. It is pretty fast and it is generic enough to allow Unicode-support. ( see http://gitorious.org/~the-user/kdevelop ... mits/lexer ) It is also quite fast, few hundred states are not a problem.

Circuit-simulator: Wasn’t there KTechlab or something like that? • Oh, I am really surprised, it is still under active development.

Economics simulation: For example what should be simulated? Pricing for certificates?

What is generic simulation?

Generic categorization: Maybe on top of Nepomuk?

In some fields there exists very elaborated free software from academia, e.g. I know about projects bio-informatics and theorem-provers with several 100k lines of code (ISABELLE, BALL, and there are others…), I do not think such stuff should be reinvented, but some of them could need a nice frontend.

I am passing along the ideas people had in the comments I linked to, there may or may not be ways to integrate these into other programs, and where there seemed to be I tried to put those ideas in the other thread.

Cantor is currently missing a lot of critical features so it isn't a replacement for existing mathematics software yet (see the other thread for my take on this).

For GIS/marble and python/kdevelop, see the other thread, I already mentioned those there.

I don't know what the economics simulator is, the person who mentioned it didn't really go into any detail that I could understand.

The "generic simulation" program would be a generalization of things like Step and a circuit simulator, allowing someone to define their own objects and their own interactions using their own equations.

The generic categorization isn't for categorizing files, it is for categorizing objects and displaying those categories. So, based on certain properties it can organize and display data in various ways. On one hand it could act as a generalization of programs like kalzium and a similar subatomic particle display program, as well as a cladistics programs. These all have in common the need to break sets of objects into groups based on different properties, display those groups in various ways, and show relationships between the groups. On the other side programs like cladistics programs and sequence alignment programs need to determine how similar objects are based on a large number of variables, but this can also be used, for instance, for categorizing spike rasters from neurons. I see being able to categorize arbitrary objects and being able to group those categories to be of general usefulness to a wide range of fields.

Both the simulator and categorizing program were attempts by me to look at some of the programs people were suggesting or that already exist and try to find underlying tasks that multiple programs had in common in order to come up with a small number of general programs that could be useful to a wide variety of fields rather than a lot of highly specialized but very similar programs.

For a paper manager, this is how I envision it:

• Strigi is used for searching for papers already on the hard drive, extracting citations from papers based on layour and/or DOI, and for full-text indexing the document's content.
• Akonadi is used to store and retrieve papers. Authors and journals are similar to contacts and papers are similar to emails, so the change necessary to implement this should be minimal. This also allows easy development of alternative front-ends and sharing papers over a network or storing them on a remote computer.
• Akanadi is also used for retrieving citations from online indexes
• Okular kpart is used for displaying papers
• Konqueror/rekonq has integrated system similar to Zotero for retrieving papers and citations from web sites and storing them on the local drive
• A kio slave is available to find and work with documents
• A program similar to kmail is used to find, organize, and display papers as well as export bibtex files
• A plugin in koffice is used to format and integrate papers into office documents (a similar plugin could be used for openoffice or even MS word).

TheBlackCat wrote:Here are some of the ideas brought up, in no particular order (see here):

• A good reference and paper manager
• A 2D molecular model drawer
• A sequence alignment program
• A cladistics program

So does anyone have an additions, or any comments on what they would like to see in any of the programs above?

For the points above, I would suggest

• A Bibus or Kbib integration with Kwrite would be GREAT (the only thing making me stick to openoffice is Bibus).
• UGene (http://ugene.unipro.ru/) can do lots of stuff related to bioinformatics (including cladistics, sequence alignment and protein structure view, see http://www.youtube.com/user/UniproUGENE for examples).

Some added functionality that I would love to see, either to UGene or as a stand-alone program (KDE port of GENtle?) is a graphical plasmid editor with functionality similar to Vector NTI (Invitrogen).

I hve noticed that one:
http://github.com/scummos/kdevelop-python/

Bioinformatics, chemistry etc.: http://www.netsci.org/Resources/Software/index.html
Bioinformatics: http://www.bioinformatics.org/softwaremap/?form_cat=2
Theorem provers: http://en.wikipedia.org/wiki/Automated_ ... g_software
Some interesting other fields from computer science came into my mind, tey may be used in other KDE-applications and it may be feasible to implement a new application for this problem:
• A tool for evolutionary/genetic programming
• General gaming (that can be definitely fun for many differen kinds of people)
• CSP-Solving
• Programming language+compiler
• Texture synthesis

I would love to see OSIRIX , a DICOM viewer. Currently its made only for MAC OS X, however there is an opensource version , I dont know if anybody has ported it to Linux
windows systems are buggy ,they are slow. It does a wonderful job on mac os. There is no other worthwhile dicom viewer for Linux. Though there are many programs which can open dicom files ,but none are useful in clinical scenarios , reading CT, MRI.

Medicine: http://www.idoimaging.com/index.shtml
All of them are free software, e.g. some DICOM viewers AMIDE (Gtk+), KRadView (KDE 3)…

A Bibliography manager. JabRef is just about the only non-KDE app that I use.

An ideal bibmanager has basically all the functions that TheBlackCat mentions, but it should also keep in synch with a BibTeX file.

Personally, I would also like improved support in KDevelop for Fortran 90/2003, but I can work my way around that with Kate.

Cantor with octave background is something that I would start to use as soon as it is stably released

Well, I think there are not that many people interested in Fortran…
That does not mean there is no chance (there is an Erlang-plugin, too), but somebody has to do it.

The User wrote:Well, I think there are not that many people interested in Fortran…
That does not mean there is no chance (there is an Erlang-plugin, too), but somebody has to do it.

Fortran use in scientific fields is huge. It is not used in commercial software development much anymore, but it scientists still use it a lot for internal software and there is a massive amount of existing software written in it.

Maybe, but I have never noticed anybody in the KDE-community using it and for KDevelop-support this is somehow important.
Most newer research-projects I have heard about use C++, Java or Python…

OK, let me add my "two cents"...

First let me introduce myself: I'm a physicist working on plasmas and head of a research group. On our group we make a lot of modeling as well as experiments.
And in each of these two areas the Kcommunity can help us a bit with:

a) Fortran support on Kdevelop! WE DO USE Fortran 95/2003! We don't care about fancy-new-languages as the scientific community has a huge legacy of good and well tested (Fortran) codes. We don't have time to learn "new languages"... Besides modern Fortran 95/2003 does a very nice job (although sometimes is slower than good old FORTRAN 77).
I've actually used Kdevelop 3 to write and compile Fortran and was displeased when Kdevelop 4 drop Fortran support. This has made me move to Eclipse...
To tell you the full story we also use Python - a lot - for pre- and post- processing of modeling data. Mostly using numpy / scipy for number crunching and matplotlib for plotting. Finally we also find very convenient to developed libraries in Fortran, glue then to Python with ctypes or f2py and use Python as a front-end.

b) A very usefull tool (but also a BIG project) would be an ELN (Electronic Laboratory Notebook http://en.wikipedia.org/wiki/Electronic_lab_notebook)!
This is a horizontal application for all fields of science and one area where is very difficult to find Open Source applications. If there was something like that in KDE this would make KDE THE desktop in a lot of labs.

Just one more comment on my "request" for a ELN:
I have used "Basket Note Pads" (http://basket.kde.org/) as a personal ELN and it has some interesting characteristics:
- take notes;
- Integration of text, images, worksheets and links;
- Password protection;
- Tasks.

However it lacks a lot of things (network awareness; group activities; templates; database; data treatment tools) to be a good ELN and it is too flexible for this purpose...

If someone is interested in picking this as a project I can gladly contribute to the definition of requirements and maybe some code. Unfortunately I'm not familiar with C++ and the KDE internals.

I think it would be much easier to just add the missing feature to an existing project like BasKet rather than starting a completely new one. I'm not familir with ELN program...do you think there is no way one can improve BasKet to work as ELN?