Hi Philipp, thank you for your reply. Now I have much clearer view on the project you would like to do.
> Ok the question I did not answer why I do not work on sage? > > I like sage and I think it is really great software. I want to create > an alternative to the sage webbased interface because I think GTK is a > superior technology for creating an interactive interface. I think in > the long run a gtk interface will be easier to maintain and extend > (things like autocompletion, syntaxcheck, etc can be easily > implemented with a custom gtk.TextBuffer and gtk.TextView ). I think > an easy to use and powerful GUI is the key feature to bring open > source mathematical technology to the mass market. This is a concrete motivation, which the previous application was missing. Having light weight version (but full of features and user friendly) of SAGE notebook is an interesting idea. Note however that Basti have done a lot work in this field, so I think cooperation with him would be very fruitful. Then you would have a basis and could implement more advanced features and, what is more important, it would be more likely to finish your project in GSoC time frame. Basti, could you tell us about symbide's status (last commit was on 05 January). > are you determined to use GTK ? symbide is written using GTK, so if you would like to extended symbide's functionality, then GTK is must. I won't judge which is better, GTK or WX, but if you have good understanding of GTK and know how to use it efficiently, then it would be better to use it. Of course there is a point in using WX, as Sebastian suggested. So, Philip, please write new application, change the title to something meaningful eg. "GTK based GUI for SymPy" (or something similar), add motivation why not SAGE notebook, in which details projects will differ. Also don't forget to say a few words about SymPy, ie. why for SymPy, what improvements it will gain eg. better MathML generation code etc. Mateusz 2008/3/31, PhilippStrack <[EMAIL PROTECTED]>: > > Hi Sebastian, > > at the moment I am really favouring GTK, but I am not determined. I > will use what ever technology makes it possible to achieve good > results. The Pro's for gtk are that the combination of gtk.TextView > and gtk.textbuffer is really well suited to do what I want to do. Even > better there is already some code ( reinteract ) that can be used as a > starting point wich will enable me to get started really fast. What > are the drawbacks from GTK apart from the Platform thing? > > Thank You for your feedback, best wishes > > Philipp Strack > > P.S.: > I think Mac Applications need not to depend on X > http://developer.imendio.com/projects/gtk-macosx/ > > > > > On 31 Mrz., 15:25, "Sebastian Haase" <[EMAIL PROTECTED]> wrote: > > Hi Philipp, > > > > I did not read all of your post. > > But, are you determined to use GTK ? > > I would be much happier if you could go with wx -- wxPython. > > It is also a quite powerful GUI gramework, and it runs much better on > > non-Linux platforms: > > I think there is a GTK version for Windows, but it surely is non-native. > > Also for Mac -OSX, as far as I know, GTK apps would always depend on > > X11 and not be native. > > > > Matplotlib supports wxPython. > > > > I understand that GTK vs. wx is almost a religious question for some > > people. But I hope I'm not alone in seeing the advantages of wxPython. > > It certainly is a very well developped platform and comes with a super > > helpful community. Many SciPy tools (e.g. the new MayaVi2) are build > > around wxPython. > > > > Thanks for writing free code, > > Sebastian Haase > > > > On Mon, Mar 31, 2008 at 12:58 PM, PhilippStrack > > > > > <[EMAIL PROTECTED]> wrote: > > > > > Hello, > > > > > I want to create a GTK based GUI for sympy in the GSOC Program 2008. > > > it would be really nice to get some feedback on the application: > > > > > Gnome Summer of Code Application > > > > > Target of this application is to combine available open source > > > technologies to build a modern computer algebra system with a focus on > > > easy usability and easy expandability. I will take python as the base > > > language and use simpy to give it scientific capabilities. To build a > > > Mathematica notebook like interface i will modify reinteract to render > > > MathML formulas and Plots inside the script code. For the rendering of > > > formulas I intend to use gtkmathview. As a plotting technology I will > > > use matplotlib, because it is widely used and provides wide range of > > > plot styles. > > > > > technologies: > > > > > python http://www.python.org/ > > > sympy http://code.google.com/p/sympy/ > > > scipy http://www.scipy.org/ > > > reinteract http://www.reinteract.org/ > > > gtkmathview http://helm.cs.unibo.it/mml-widget/ > > > matplotlib http://matplotlib.sourceforge.net/ > > > > > I the first phase of the GSOC i will work on integrationg all the > > > technologies. The following things need to be done: > > > > > Work on reinteract to better integrate with Sympy > > > > > 1) make sympy formula objects plottable > > > Write a function that plots sympy formula objects using matplotlib and > > > give the gui the ability to render the plots inside the document. > > > > > 2) make reinteract render sympy formulas as MathML > > > sympy objects can already be converted into MathML by the function > > > print_mathml(). I will rewrite this part of sympy to get high quality > > > mathml. The next step is to get the MathML rendered into reinteract, I > > > will use gtkmathview. There are at the moment no python bindings for > > > gtkmathview I will either write python bindings or render the mathml > > > into an svg which can easily be embedded into a GtkTextView. > > > > > 3) provide a command for reading and writing CSV (comma seperated > > > files) > > > most scientific data is exchanged as a CSV, providing a command for > > > handling csv files is therfore essential. There are multible python > > > classes for reading and writing CSV. I will use one of them and make > > > sure that the resulting functions are easy to use. I want them to > > > automatically recognize the seperator and the structure of the csv > > > (the csv's generated by octave, R, exel, matlab,.. are very > > > different ). It is essential that the data from the csv file can be > > > easily accessed from scipy and sympy > > > > > In the second phase of the GSOC i will work on the GUI > > > > > 4) Provide a Gui that makes it easy to access the CAS commands > > > (Plotting, differentiation, Loading Files, etc..) without knowing the > > > exact commands > > > > > 5) Modify reinteract such that it allows non executable code inside a > > > document > > > > > 6) Define a file format for the worksheets > > > > > In the third phase i will make sure it is easy to get started using > > > the newly created program > > > > > 7) Write documentation > > > > > 8) Set up a website . Create example worksheets and present them on > > > > the website to encourage use (a bit > > > likehttp://demonstrations.wolfram.com/) > > > > > > 9) Create a Forum to collect possible use cases in order to direct > > > future development > > > > > Mateusz Paprocki > > > A few comments about goals of this application: > > > > > 0) Integrate sympy, scipy and matplotlib > > > > > How will you achieve this? Will you start a new project > > > or just add support for scipy and matplotlib to sympy? > > > > > 1) make sympy formula objects plottable > > > > > This feature was implemented long ago > > > (see sympy.utilities.lambdify). > > > > > 2) make reinteract render sympy formulas as MathML > > > > > MathML generator wasn't updated for ages and needs > > > rewriting almost from scratch. Will you do this? > > > > > 3) provide a command for reading and writing CSV > > > > > SciPy and Matplotlib have this feature. Why do we > > > need another approach to CSV. How it will differ > > > from existing implementations? > > > > > So could you tell us what is the main goal of this project? > > > > > It seems that you would like to: > > > > > [1] create new CAS, or glue together existing products > > > > > But there is SAGE which has bindings to most (if not > > > all) CAS available in the market (also it supports > > > SymPy). How we will benefit from your system? > > > > > If you want to improve SymPy's numerical evaluation abilities, > > > using SciPy for this purpose, then you should have discussed > > > this with SymPy's developers. > > > > > [2] write GUI for [1] (or maybe for any CAS?) > > > > > There is beautiful SAGE notebook (www.sagenb.org), > > > which you should have mentioned in first place in > > > your application. > > > > > Rather than doing things from scratch, you could > > > help SAGE reaching its goals. Quite a few GSOC > > > ideas can be found here: > > > > > http://wiki.sagemath.org/gsoc08notebook > > > > > btw. "Abstract" and "Detailed description" should differ. > > > 03/31/08 08:03 > > > Philipp Strack > > > Hi, > > > > > thank you very much for your feedback on my GSOC proposal. I will try > > > to answer your questions and rewrite my application to be more precise > > > > > > Mateusz Paprocki > > > >A few comments about goals of this application: > > > > > >0) Integrate sympy, scipy and matplotlib > > > > > >How will you achieve this? Will you start a new project > > > >or just add support for scipy and matplotlib to sympy? > > > > > I dont want to start a new Project! Sympy is the mathematical base > > > technology behind this proposal > > > > > >1) make sympy formula objects plottable > > > > > >This feature was implemented long ago > > > >(see sympy.utilities.lambdify). > > > > > oh i did not find it, sorry > > > > > >2) make reinteract render sympy formulas as MathML > > > > > >MathML generator wasn't updated for ages and needs > > > >rewriting almost from scratch. Will you do this? > > > > > Yes > > > > > >3) provide a command for reading and writing CSV > > > > > >SciPy and Matplotlib have this feature. Why do we > > > >need another approach to CSV. How it will differ > > > >from existing implementations? > > > > > I have no interest in reinventing the Wheel. I just had some problems > > > using the different python classes for reading csv files as they do > > > not recognize delimiters automatically i just want to add this feature > > > because I think it will make handling csv files much easier for not so > > > tech savy users > > > > > >So could you tell us what is the main goal of this project? > > > > > >It seems that you would like to: > > > > > I want to bring existing open source technologies together to create a > > > software that makes it easy for users that are not tech savvy to use > > > sympy. To be more concrete I want to create a gtk interface for sympy > > > (based upon reinteract). This Interface will be able to do indocument > > > rendering of formulas and graphs. To get the ability of formula > > > rendering I need to work on sympy's MathML generator as well as on > > > gtkmathview. If it is needed I will rewrite the MathML generator from > > > scratch. > > > 03/31/08 09:03 > > > Philipp Strack > > > Ok the question I did not answer why I do not work on sage? > > > > > I like sage and I think it is really great software. I want to create > > > an alternative to the sage webbased interface because I think GTK is a > > > superior technology for creating an interactive interface. I think in > > > the long run a gtk interface will be easier to maintain and extend > > > (things like autocompletion, syntaxcheck, etc can be easily > > > implemented with a custom gtk.TextBuffer and gtk.TextView ). I think > > > an easy to use and powerful GUI is the key feature to bring open > > > source mathematical technology to the mass market. > > > 03/31/08 10:03 > > > Philipp Strack > > > some information about myself: > > > I am a 23 year old student from Bonn, Germany. I have a Diploma/master > > > degree in economics and I am currentely in the last year of my diploma > > > in Mathematics. I am specialiced on numerics and I am writing my > > > diploma thesis in Mathematics about the efficient solution of Optimal > > > Control Problems with restrictions over PDE's using wavelets. Durin my > > > studies I have been teaching Mathematics at the Economic Department > > > for 2 Years. During this courses I have used a variety of different > > > mathemtical Programms (R,stata,matlab,maple,Matematica). > > > > > --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "sympy" group. To post to this group, send email to sympy@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sympy?hl=en -~----------~----~----~----~------~----~------~--~---
