[help-3dldf] contributing to 3DLDF

[arnuld]

> C++ is probably the most important programming language in a commercial
> sense at present.  However, it seems that there are more and job postings
> requiring languages such as Perl, PHP, Visual Basic, Visual C++, and C##.

Well, from that experience I have also observed a pattern, a future-pattern
that shows that  an indistrial shift will happen after next 5-6 years from
"My Favorite Programming Language" or "This is Standard/Best Programming
Language for everything" to "Better tool for the problems at hand" and
"Problem Space Abstraction" in relation to lower level details of a
language" . e.g Java vs Haskell and C++ vs Common Lisp. not to mention the
Mercury and OCaml.


> I believe I'm a good programmer and I know that I use most of the features
> of C++ and many of the data types and functions in the standard library.
> At present, I'm the sole author and all it's gotten me so far is the
> occasional job interview.

:(

that scares me here in India


> I now have a job, except that I can't start because of a problem with my
> residency permit.  The job involves porting old Fortran programs to a new
> system.  If I'm allowed to start, this will make it easier for me to work
> on 3DLDF.  I plan to stop adding features and start preparing a new
> release.  I've been working on getting the source files fixed up so that
> CWEAVE can process them.  CWEAVE is part of CWEB, which you will also have
> to learn.  Normally, I just run CTANGLE to generate plain C++ code and
> don't bother running CWEAVE to generate the documentation.  Now I have to
> go through each file and correct any errors.

congratulations for your new Job and I pray for you get residency permit.
I will better learn CWEAVE too.


> What I'd like to do next is try to get access to a system with the most
> recent versions of GCC, Bison, Flex, the standard library and especially
> the GNU Scientific Library, and try building the package on that system.

hmm... well, I can build them from source and try that on my system Arch
x86_64. May be I will switch to Debian after 10 or 15 days.



> You could skip reading _The TeXbook_, at least for the present, but
> everything else makes sense.

I don't have the TeXbook because that is not available in India. All I
have is the manual available with the TeX source.


> You may need to read _The METAFONTBook_, but
> you should at least try to get yourself a copy.  If you have trouble
> building 3DLDF on your system, please let me know.  I'd appreciate it if
> you'd subscribe to the mailing list and send any questions there.  It
> would be nice if there was a little traffic on it.

I have subscribed both to help-bison and help-3dldf :)



> No, it all makes sense so far.  However, please be aware that as far as I
> know, 3DLDF has _no users at all_ except for myself.

I know that because latest message on the mailing-list is of April 2007.


> I know that a couple
> of people have used it, and not only people who've contacted me.  I want
> to continue developing and using it for my own purposes, but if it wasn't
> for that, it would be a dead project.

One thing that always confused me is that Why did you start 3DLDF ?

    I mean we all know why Ekiga exists or why gcc exists. Just don't have
any idea about history, goals. mission and future of 3DLDF.


> It's also possible that a potential
> employer wouldn't allow me to continue working on it.  If I had a choice,
> I wouldn't agree to this, but I don't have a choice.  This current job is
> only for a year and there is almost no chance of an extension.  I'm only
> mentioning this as a possibility;  it isn't very likely that an employer
> would force me to stop working on it.  If it did happen, I'd ask the Free
> Software Foundation to find a new maintainer, but I doubt very much that
> anyone's ever read even a fraction of the source code.

If it is possible I will keep on working on it :)


> You should also be aware that there is a very large amount of source code
> and the package is rather complex.  You won't have to read the whole
> thing, but you will have to learn your way around it.  Be warned that it
> will be very boring to read.  You don't have to read all the log entries.
> Computer graphics is interesting but involves a great deal of tedious
> drudgery.  A working package like 3DLDF has to do a lot of boring things
> that one can skip over in a book about the mathematics and algorithms used
> in computer graphics, such as keeping track of dynamically allocated
> memory and accounting for error conditions.

I am not interested in computer graphics or 3D drawing. All I work for is
because 3DLDF is GNU project and sooner or later GPLv3 will apply on it
and 2nd, it is written in C++ and that is only language i know and
understand better than an average C++ programmer.

The negative side is that 3DLDF code is huge and complex. I have already
spent 2 years in understanding and sharpening my craft for C++ and I m
jobless. I do not want to carry 1 more year just to get familiar with some
piece of software. I am theoretical man right now because I have no
experience in writing software. After 1 year I want to stand as a better
technical C++ programmer who understands the practical aspects of both C++
and Software Development. I want to stand as the one who can go in front
of an interviewer and tell him that "I am an experienced developer and I
can
write C++ code better than an average Post-Graduate out from some #1 or #3
university/college".

Will 3DLDF help in achieving that aim ?


> If you do work on the package, it's more likely to be working on new
> features or making changes that I've planned rather than bug fixing.  For
> example, these are a couple of things things I'd like to do but have never
> gotten around to:
>
> 1.  Write a parser for a machine-like language that reads binary files and
> write a `dump' function for outputting the state of a run in the
> corresponding binary format.
>
> 2.  Write an `output' function to write output in OpenGL and/or PNG
> format.
>
> 3.  Find out how to turn 3DLDF into a plug-in for GIMP.
>
> There's also some housekeeping I'd like to do.  For example, there are
> some parser rules that I'd like to get rid of.

 well, I can do it if you mentor me.


> Thanks again for your interest and if you have any questions, please feel
> free to write to me anytime, preferably on the mailing list unless it's of
> a private nature.  If you have any questions about Bison, it would be a
> good idea to subscribe to the help-bison mailing list, too.  I subscribe
> to it and contribute frequently.  There's a mailing list for MetaPost, but
> I no longer subscribe to it.


thanks for the information.




-- arnuld
http://lispmachine.wordpress.com


-- arnuld
http://lispmachine.wordpress.com


--
    
Email and shopping with the feelgood factor!
55% of income to good causes. http://www.ippimail.com