Important Research Project

dave_w dave.willmer at gmail.com
Thu Aug 30 20:55:39 CEST 2007


On Aug 30, 2:54 pm, Paul McGuire <pt... at austin.rr.com> wrote:
> On Aug 30, 8:12 am, "E.D.G." <edgrs... at ix.netcom.com> wrote:
>
> > Important Research Project  (Related to computer programming)
>
> > Posted by E.D.G. on August 30, 2007 edgrs... at ix.netcom.com
>
> >     This report is being posted to a number of Internet Newsgroups
>
> Always the hallmark of a considerate poster.
>
> > to see if
> > there are any experienced computer programmers who would like to provide
> > some assistance with an effort to develop a Perl language computer program.
>
> >     Interested parties can try contacting me by e-mail or by posting a
> > response note to the comp.lang.perl.misc newsgroup.  They would need to
> > download a recent (free) MSI copy of Perl from the ActiveState Web site and
> > get it running on a Windows XP or Vista system.
>
> >http://www.activestate.com
>
> This is a little backwards, one usually presents their research topic
> *first* and their contact info *last*.  The reason?  SO PEOPLE KNOW
> WHAT THE @#$(&#!@ YOU ARE WORKING ON!  Ok, I'll bite, keep reading...
>
> >     I am presently using Perl 5.8.8
>
> Whoop-de-doo for you.
>
> > but plan to upgrade to the latest
> > version as soon as possible.  People can use Windows 98 if that is the only
> > operating system available.  Perl also runs on other operating systems.  But
> > at this time I specifically need help with the Windows version.
>
> I suspect Perl is largely the same Perl on all those platforms.  Win
> 98?  Is this a zombie spam from the turn of the century?
>
> >     The goal is to have a single Perl program (or modules) perform functions
> > that have been done by a sizeable collection of other language programs in
> > the past.
>
> Doing what!?  Grrr..., keep reading, there's gotta be a punch line...
>
> >     Help is presently needed with learning how to get Perl to generate
> > charts and also produce standalone .exe copies of itself.  The plan is to
> > then make those .exe copies available to other scientific researchers around
> > the world for free use along with free use updates when they become
> > available.  If other researchers wish to get Perl running on their own
> > computers then they will probably also be given the source code for the
> > original program for free use so that they can do their own development
> > work.
>
> Ohmigod, is Google broken?  Or has Perl gone this long and this far
> without support for creating charts and graphs?  Sounds like about 10
> minutes of research.
>
> >     Perl was originally chosen because it is quite versatile, is a free
> > download, and is supported both by ActiveState and quite a few independent
> > programmers.  So other researchers could get their own versions running
> > without having to worry about viruses or cost.
>
> (Why is this posted on all these non-Perl newsgroups, then?  I've
> *seen* Perl already, and never want to again!)
>
> >     So far the work is fairly advanced.  The effort has been underway for at
> > least a decade.  
>
> ... and we are just getting around to plotting some data.
>
> > The core data generation program was formally copyrighted
> > several years ago.  
>
> Red flag #37 - "formally copyrighted", wooo-ooooh
>
> > My present version of Perl will send data to Windows as
> > if it were being manually typed into the keyboard (important for controlling
> > other programs).  And it can directed to respond to most keystrokes even
> > when another program is the active one.  Unfortunately, Windows also
> > presently responds to those keystrokes.  And that complicates things a bit.
>
> ... and has for the past decade, and I still haven't figured it out.
>
> >     Not being a professional computer programmer I have been finding it
> > difficult to get new features such as a chart generating ability merged with
> > and running with Perl.  And the entire research project is now being slowed
> > as a result.  One of my colleagues has done an extensive amount of work with
> > Basic.  And I even offered to pay him to help with the Perl development
> > effort.  But after he downloaded a copy of Perl and examined it he decided
> > that this would involve too much effort.  I have to agree with him.
>
> Maybe that up-front language choice could stand a review...
>
> >     Once it is possible to create charts and .exe versions the plan is for
> > researchers around the world to develop Perl modules for generating a
> > variety of data related to sun, moon, planet, ocean tide crest and trough,
> > and Solid Earth Tide locations.  Most of those data can already be generated
> > with other programs.  Some of the data are not yet available anywhere as far
> > as I am aware.  If the effort is unusually successful the Perl program (or
> > modules) might eventually be converted to CGI programs that will run at one
> > or more Internet Web sites.
>
> AHHH!  The "research", I almost forgot!  Why is it left to the end of
> the message?  And do we still know what the @#(*&$ this "variety of
> data" is for?  I'm going to take a wild guess here - earthquake
> prediction?  perpetual motion?  picking guaranteed-to-win lottery
> numbers?
>
> After a pitch like that, who could not be tempted at the prospect of
> "generating a variety of data related to sun, moon, etc. etc."?  Sign
> me up - NOT!
>
> GTFA,
> -- Paul

Thanks for the amusing reply Paul, i think you summed up a few
people's feelings quite fairly, but maybe a little more tact next
time? :-)

I'm going to begin my answer with the age old taboo of 'use a
different language', but since you cross-posted to comp.lang.python, i
feel justified.

so... use python.

you didnt exactly state what you wanted out of the project, so i'm
going to take a few guesses.

if you want astrophysical data, there is a python module which is
based on NASA JPL's ephemerides system, you can find it here:

http://www.cv.nrao.edu/~rfisher/Python/py_solar_system.html

and yes, i know there is Astro::SLA on CPAN.

if you want (basic) ocean/tidal modelling, you can use this (again...
python)

http://geosci.uchicago.edu/~cdieterich/swom/

or take a look at this PDF entitled 'python for assembling climate
modelling toolkits':

www.esmf.ucar.edu/presentations/pres_0305_jacobs.pdf

if you want to connect to online databases etc., using urllib2 (a
standard python module), you can fetch the data you need in only a few
lines of code (and i mean 2 or 3 lines for basic requests)

for plotting... use matplotlib, in most python programmers minds it
has reached legendary status.

if you're still thinking about plotting in Perl, look here

http://biptest.weizmann.ac.il/course/prog/graphics/success.html

it is the graphing page from the Weizmann institute in Israel which
does a course entitled 'Perl programming course for bioinformatics and
internet'; perhaps you could use some of your research money to send
someone on the course?

and if you want to put graphs online, you can use matplotlib server
side to generate a jpg for example, and then show that using html
(again, not very many lines of code).

without knowing exactly what needs accomplished, i ts difficult to
help, but it seems like most of your programming has already been
done, and just needs some code to glue different modules together
(trust me on this one - use python).

with python, matplotlib etc, everything is cross platform, if it is
web-based access you're after, it doesnt matter what platform you
develop on
you also talk about generating standalone .exe files - use py2exe w/
NSIS, or pyInstaller, both of which have tutorials online.

most importantly of all... re-read Paul's post above.
if you don't find it as amusing as the rest of us, here's the gist...

"Ohmigod, is Google broken?"

cheers
Dave




More information about the Python-list mailing list