> I would rather tell that you think biology is molecular biology
> and molecular biology is the study of sequences and atomic structure,
> which is wrong. Thank you for the population biologists,
> the physiologists etc., all using a lot of computing.
> However, I am currently working to port a stochastic simulator all
> written in C++, with a nice interface (written with visual C++).
> And (1)- beyond the nice interface, it is the hell (no dynamic
> allocations, just macro def, hundredth of strcat and strcpy etc...),
> which means a good interface does not mean a good program,
> (2)- I am forced to rewrite the interface, because it is absolutely
> not portable.
I never told that you need to use visual C++. My choice would be wxWindows
(if using C++) or Java for any program that will require portability as well
as GUI. My original post was for that- I was not being able to choose
between these two!
> How can you say that! Gimp was developped a lot faster than
> photoshop! But nevertheless to say you can write it in C++ in
> less than 2 years is not very humble. Hey genious, you have to
> contact Adobe to write Photoshop 6, they miss you over there.
Not exactly! I am not very sure how long it took to write Photoshop but GIMP
has a long history. GTK anf GNOME were part of it! Consider that too.
> Are you prophet as well?
No! Just a realistic and pragmatic individual
> No. They are bioinformatics programs written for bioinformaticians.
Who are these people??? Bioinformaticians????
> In my PhD lab, they did their sequence alignments with MS Word!
You support my idea! Gurus failed to deliver it to masses!
> You have to possess computer skills to make bioinformatics.
Sure to make bioinformatics but use the facility?? I am not very sure!
> why? Because the programs are continuously changing, and to use the
> best one, you need to understand the algorithmes below. All the
> crystallographer I know are using academic softwares and prefer
> the one with scripting languages in order to write extensions.
> And on reverse, you need biological background to write good
> bioinformatics programs, otherwise you get a nice interface
> providing bat results. This is why the best programs, whatever
> is the biological field, are always academic (or assimilated)
I am talking about millions of people using softwares in general not
customising for special purpose.
>>> At home I am working with a AMD K5 and 24 MB under GNU/Debian slink
> (OK, at work I have an AMD Athlon and 128 MB, still not a SGI)
> And I AM NOT A COMPUTING GURU. I am a former experimental biologist
> who progressively migrated to bioinformatics throught image analysis
> and sequence and structure treatment. Then I know how work the
> experimental biologist without computing background, I know how
> work the biologist who wants to use bioinformatics tools, and I
> know the problems occuring to the programmer of such tools.
>And still you disagree for GUI and portability???? Strange!
>> Do you want to produce useful programs or not, and for who? Mac users
> will not use your programs. They will use remote servers (running
> under unix), or commercial programs!
That's pessimism! :-(
>> Yes, it is, and Linux (or other unices for PC) is becoming the
> OS of choice for bioinformatics, replacing the SGI and Sun you
> talked about ...
>I agree! But I want to include all!
>> Because diversity probably means Windows with visual C++?
> OK, no hope. As I tell to my daughter "you'll understand later".
> Write your first program and we'll talk later.
Not at all ! Diversity means all of them, all languages and all OSs. I have
already written my first project and you'll be surprised to know it is
written in none of the languages that we talked about. It is written in
about that- it runs in all systems under all platforms! I'll let you know
when I release it(first week of next month). I'll wait and see!
Thanks for reacting to my post!
Malay Kumar Basu
Centre for Cellular and Molecular Biology
I N D I A
Cataholic: Can't stop bringing cats home.
curiouser at ccmb.ap.nic.in