[proFit-list] alpha testers wanted

Mon Apr 27 14:19:35 CDT 2009

Dear pro Fit users

We did expect Python to be a controversial issue, but it was not a  
decision that we took lightly. The language that we were to adopt had  
to fulfill the following criteria:

1) The language should be available on a default MacOS without the  
need to install any special software.

2) The language should provide a convenient mechanism to pass named  
arguments to functions. Think of a function such as PlotData, which  
has nearly 100 possible arguments, such as the color of the plot, the  
line style, the point style, etc. etc. In a language that does not  
support named arguments, such as C, you would have to pass all 100 of  
them each time you call PlotData. A call to PlotData would e.g. look  
as follows:

  PlotData(scatterPlot, 1, 2, 0, FrontWindow, true, true, false, 0, 1,  
0, 1, 0, 1, linScaling, linScaling, linScaling, ...etc)

With named arguments, you only pass the arguments whose default values  
you want to change, such as
  PlotData(xColumn 1, yColumn 2)

3) It should be possible to integrate the language into our built-in  
debugger. therefore, the language should provide a reasonably easy to  
use mechanism to obtain the names and values of variables, to set  
breakpoints, etc.

4) The language should be such that it is at least difficult to make  
pro Fit crash or hang.

5) The language should be suited for scientific use, should have a  
large base of users in scitech and there should be a large pool of  
readily available scitech software for it.

When evaluating the languages, we did not have any bias ourselves. (In  
fact, personally, I do not really adore Python's indentation-based  
syntax, but after having worked with it for some time I see that it  
does have its advantages.)

Basically, the languages fulfilling criteria 1 through 4 are Python,  
Ruby, JavaScript, Apple Script and PHP. In addition, Python excels in  
criterion 5 (it supports a rich math library, complex numbers, and  
there are very powerful packages for numerical analysis, such as numpy  
(which is installed by default in MacOS 10.5) and scipy). Ruby was  
close, but definitely second, and Java Script also had its appeal  
because it is bound to see some major speed optimizations in the near  
future. (PHP is too web centric in our view, and Apple Script provides  
poor numerical capabilities and is slowish.)

When integrating Python into pro Fit, we redesigned the pro Fit  
scripting architecture in such a way that most of the application is  
agnostic of the actual scripting language in use, with a defined  
interface between pro Fit and the scripting language. Hence, the  
integration of a third language in the future should be much easier.  
Therefore, if you provide us with good reasons to integrate a further  
scripting language, we may consider this. However, first we will first  
work on optimizing the Python and Pascal scripting.

And please, as mentioned by Chris, do not start a language war here.  
There are enough of these out there.

Regards

Kurt Sutter
QuantumSoft

On 27.04.2009, at 08:17, Chris Lee wrote:

> I am not going to start a language war, but I think there are some  
> very good reasons for choosing python. It gives you a lot of  
> language integration for free. You can call C and fortran code with  
> ease and receive the results as a python type variable. So,  
> depending on what the final implementation looks like, this will  
> actually do exactly what you want with a lot less work for the  
> quantumSoft dudes.
>
> Another reason for choosing python might be numpy, scipy and  
> scientific. These three libraries basically give you an open source  
> matlab (sans visualization... you need other packages for that). It  
> handles arrays and complicated operations much more cleanly than  
> current pro fit coding operations. By integrating python pro fit can  
> take advantage of all of this without having to reinvent the wheel.
>
> That said, there are good reasons for including other languages as  
> well.
>
> Cheers
> Chris
>
> On Apr 26, 2009, at 10:54 PM, s.g. prussin wrote:
>
>> Thank you for this. Once again, I'd like to ask that you consider  
>> the direct importing of FORTRAN code. For many in the scientific  
>> and engineering community, we still use this language and it must  
>> be so because of the very many legacy codes that are the basis of  
>> many calculations. I know it's a big request but please consider it  
>> - sgp
>>
>>
>> On Apr 26, 2009, at 3:32 AM, pro Fit Support wrote:
>>
>>> Dear pro Fit users
>>>
>>> We at QuantumSoft are presently working hard on the next version  
>>> of pro Fit, which will be version 6.2. The release date of that  
>>> version is still uncertain, but my guess is that it will be  
>>> sometime in late 2009 or in  2010, so don't hold your breath. It  
>>> will provide a number of interesting new features.
>>>
>>> One of these new features will be integrated support for the  
>>> Python programming language for defining Programs (Macros) and pro  
>>> Fit functions, as an alternative to pro Fit's own Pascal-like  
>>> compiler.
>>>
>>> We feel that some user feedback would be valuable at this time. We  
>>> therefore have decided to run an "alpha" testing program with  
>>> version 6.2, with emphasis on its Python programming capabilities.  
>>> So, if you have some Python knowledge, or if you are an  
>>> experienced pro Fit user willing to spend some time digging into  
>>> Python and testing pro Fit with it, please send us a note to
>>>
>>> profit at quansoft.com,
>>>
>>> with a brief description of who you are and why you feel you could  
>>> contribute to the alpha testing. If you are accepted to the alpha  
>>> testing program, we will then send you a pro Fit build to test.
>>>
>>> Please do not expect the build that we will send you to be fit for  
>>> normal use. It will be an internal build that has not undergone  
>>> much testing and that will have various parts unfinished or  
>>> disabled. It is for testing only. It will require MacOS 10.5.6 or  
>>> better.
>>>
>>> Best regards
>>>
>>> Kurt Sutter
>>> QuantumSoft
>>>
>>>
>>> _______________________________________________
>>> proFit-list mailing list
>>> proFit-list at quantum-soft.com
>>> http://quantum-soft.com/mailman/listinfo/profit-list_quantum- 
>>> soft.com
>>
>> S.G Prussin
>> Department of Nuclear Engineering
>> University of California
>> Berkeley, California 94720
>>
>> prussin at berkeley.edu
>>
>>
>>
>> _______________________________________________
>> proFit-list mailing list
>> proFit-list at quantum-soft.com
>> http://quantum-soft.com/mailman/listinfo/profit-list_quantum-soft.com
>
> _______________________________________________
> proFit-list mailing list
> proFit-list at quantum-soft.com
> http://quantum-soft.com/mailman/listinfo/profit-list_quantum-soft.com

Best regards

Kurt Sutter
QuantumSoft

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://quantum-soft.com/pipermail/profit-list_quantum-soft.com/attachments/20090427/3c7b2f22/attachment.html>